68% of Google searches end without a click.

I know, you’re an overachiever and you know all about this. But still, it’s not common knowledge. When there are this many zero-click searches, it’s generally because the user got what they wanted right there in Google. Which means they didn’t go past Google’s search results page. Which means a website didn’t get a click.

AI Overviews might be turning search results into a place for deeper consideration.

When an AI Overview appeared, people who scrolled in both directions spent 47.5% of their scrolling going back up the page, compared with 27% without an AI Overview. They also paused their mouse cursors more often.

These behaviors suggest that people may be spending more time reading, comparing, and thinking about the information they’re seeing. And they’re doing all this right there on Google, before they ever even visit your website. Again, clickless intent.

1 in 5 searches happen somewhere other than traditional search.

About 80% of search still happens on Google or Bing (but let’s face it, mostly Google). So it’s not as though search is dead. Far from it. But search has evolved. It’s a behavior now, not just a channel. So where else do people search? On ecommerce sites (like Amazon, Etsy, etc), social media, AI tools, and other websites like Wikipedia and iMDb. In that order. Which means that one in five searches happens somewhere on the internet.

If you’ve been here for a while, you know what I’m getting at. We all need to get used to optimizing for views and engagement in the places where people are already hanging out. And we have to get used to doing this in ways that we can’t perfectly measure.

Not that we shouldn’t try.

Just that we have to get comfortable with the imperfection, and be accountable to our decisions. That means developing good judgment through experience, and learning to have conviction in our ideas.

That’s a big part of what Zero Click Marketing is about.

Now if you’ve been following really closely, you might notice that my friend Kevin Indig and I have each started to solve this measurement problem pretty similarly. Our ideas converge in this measurement framework:

The idea is to measure three things:

1. Exposure: Are people encountering your brand in the places where they’re already spending time?

2. Behavior: Are they engaging, talking about you, or seeking you out?

3. Outcomes: Are you seeing more traffic, customers, and revenue?

You won’t be able to draw a perfect line connecting every impression to every sale. But when you look at these signals together, you can make a much more informed assessment of what’s working. And by the way, that chart comes courtesy of Kevin Indig’s and Amanda Johnson’s article, which you should read next:

Growth Memo

How to plan 2027 when traffic is down and AI won't attribute

Read more

8 days ago · 40 likes · 10 comments · Kevin Indig and Amanda Johnson

I mean, let’s be honest. Attribution has always been deeply flawed at best. The problem is only getting worse now.

A case study in measurement: Growclass’s word-of-mouth success

I talked about measurement and attribution with my friend Sarah Stockdale, founder and CEO of Growclass, on this week’s episode of the Zero Click Marketing podcast.

Sarah shared that much of Growclass’s business comes from referrals and word of mouth. Someone takes a course, loves it, and recommends it to a friend over text or Slack.

Guess where that traffic shows up in Google Analytics? Direct.

So how does she figure out what’s actually working?

She asks customers how they found Growclass at checkout. She uses social monitoring tools to find conversations about her brand. And because Growclass is a community business, she and her team actually talk to their customers.

Even with all that, she acknowledges that attribution is imperfect.

As Sarah put it: “We won’t always have perfect data as to whether or not it worked, but we will have general signal.”

And I think that’s exactly the mindset marketers need to adopt.

Measure what you can. Investigate what you can’t. And make informed decisions even when you don’t have all the answers.

We talked about so much more in this episode, too: the importance of customer empathy, the marketing fundamentals AI can’t replace, and why experienced marketers know when the data they’re looking at is probably wrong.

🎧 Listen to The Marketing Skills AI Can’t Give You, with Sarah Stockdale.

And if you’re in LA next week...

Come hang out with me!

I’m joining Wistia for LA Tech Week on Thursday, October 15, for a conversation about AI and creativity.

Wistia is also buying 100 copies of Zero Click Marketing to give away at the event! I’ll be signing books and hanging out with attendees.

These might be the first physical copies of our book that anyone gets their hands on. Possibly even before I get mine. 😂

The event is bigger than just my panel, so you’ll also get to meet other folks in the LA tech and creative communities.

Thursday, October 15, 1-6pm (my panel is at 3:30pm) at the Rose Room in Venice Beach.

» RSVP for free!

Hope to see some of you there! ❤️

The zero-click trend is accelerating fast. But it’s not really that things are changing every day. It’s more like the change that has already been happening — platforms hoarding users/attention, fewer and fewer clicks being available, referral traffic being misreported — is still going on.

Just a lot faster.

And lately, I feel like a tide has really shifted. Marketers and business leaders are painfully, truly aware of this now. And I think AI is both to blame and to credit.

The availability of AI answers — and I’m saying “AI answers” broadly here, not just Google, but all the LLM tools we use — is suppressing links even more.

We have pretty stark evidence of this. Pew Research Center found that when a Google search produced an AI summary, users clicked a traditional search result just 8% of the time, compared with 15% when there was no AI summary. And the links inside the AI summary? They got clicked just 1% of the time. Not to mention that Google users are more likely to end their browsing session entirely after visiting a search page with an AI summary than on pages without one.

Everybody knows it now. So I think we’re all fully realizing that marketing is going back to first principles.

We’re seeing that PR could be the future of marketing. The rise of brand marketing. Content marketing is even having some kind of resurgence. Because for content marketers, the job isn’t just to rank for keywords and increase organic traffic anymore. It’s to be findable, talked about, and to show up in the places where people — and increasingly, machines — are looking for information.

Anyway. Here are a few threads I’ve been pulling on lately. I’ve been struggling to bring them all together, so this is my best attempt (in a voice memo with minimal typing edits. My wrists are achey and I want to blame my computer behavior, not my Pokemon Pokopia addiction).

1. Backlinks are really just proxies for human behavior.

uSERP just published its latest State of Backlinks report, surveying 116 SEO professionals. 82% said they believe backlinks influence whether AI engines cite a brand. Interestingly, 87% said they would keep building links even if Google removed them as a ranking signal. And adding to my point about PR earlier, 53.4% of respondents said that digital PR delivered their strongest results in the past 12 months.

The irony is not lost on me. In this increasingly clickless world, we still desperately want those links.

But it makes sense when you think about something Wil Reynolds said when he joined me on the Zero Click Marketing podcast recently: these things are really just proxies for human behavior.

A backlink to your site, your research, your content, whatever it is, means somebody thought it was worth linking to. It’s relevant in some way to the source that linked to you. That’s a proxy for influence, quality, relevance —all the things that might help a search or AI tool understand whether something is useful to a person.

2. We still want humans in our recommendations.

Kaleigh Moore’s newsletter pointed me toward a new IAB report, “The Creator Signal: How AI Systems Interpret Influence in Commerce.”

IAB surveyed 2,200 people across five countries who had recently used AI for shopping research.

And 56% of people who shop with AI say they prefer recommendations that pull from creator or influencer reviews. Even more interesting: 65% said they’re more confident in AI recommendations when they’re informed by credible creator reviews.

That’s another proxy for human behavior.

More than half of people shopping with AI still want some kind of third-party human validation of the thing. It makes sense because we already know people like word of mouth. We like recommendations from friends and family. And, absent a specific family/friend referral for a product you’re considering, what could be a better proxy than a creator you trust?

3. And then there are Blue Spoons.

That’s in this week’s podcast episode is with Josh Ho of Referral Rock. We end up talking about Unreasonable Hospitality, Will Guidara’s book about creating experiences so memorable that people can’t help but talk about them.

Early in Guidara’s career, he was running a café at the Museum of Modern Art, and he became obsessed with finding these distinctive blue spoons for the ice cream cart. They weren’t cheap. But he thought they were worth spending money on because they were different enough to get noticed.

People would see someone walking around with one and ask: What’s that? Where did you get that?

Josh called these kinds of things “spikeable, shareable experiences.”

I love that phrasing. Because the Blue Spoon does a bunch of marketing jobs at once. It gives people something to talk about. It surprises and delights them. It becomes a distinctive branding element. Someone walking around with one is basically billboarding the experience for you.

And you can’t really measure all of that. You don’t know how many people saw the spoon. You don’t know how many asked about it. You don’t know how many eventually bought ice cream because somebody else was walking around holding this unique little blue spoon.

But you know it did something. Here’s where I think these three threads come together:

We’re spending an awful lot of time figuring out how to market to machines. How do we get cited by AI? How do we rank? How do we show up in an answer? How do we measure visibility when nobody clicks?

Those are reasonable questions. But the signals we’re chasing keep leading us back to people. A backlink exists because somebody thought something was worth linking to. (Ok fine, sometimes you can buy them, but Google really hates that and you’re not supposed to do that, and if you get caught you can get in big trouble, mister.) A creator recommendation matters because somebody trusts someone else’s experience. The Blue Spoon works because somebody notices and remembers it, and maybe tells somebody else about it.

The machines can crawl all of that. They can summarize it. They can cite it. They can try to make sense of it.

But first, somebody has to care.

So maybe that’s the first-principles marketing question worth asking right now:

What are you doing that’s worth linking to? What are you doing that’s worth talking about? What are you doing that’s worth remembering?

Or, put another way:

What’s your Blue Spoon?

📚 LA friends: Come see me at LA Tech Week! (And get my book!)

This one is kind of a big deal for me!

On October 15, I’ll be joining Wistia for The New Rules of Video Marketing, an afternoon of conversations about video, marketing, and where AI is actually useful right now.

I’m speaking on a panel about using AI in your marketing and creative efforts. But perhaps more importantly...

Wistia ordered 100 copies of Zero Click Marketing to gift to attendees!

So if you’re in LA, come hang out with us, grab a copy of the book, and I’ll sign it for you! (Er… if you’ll let me?)

It’s free, but you’ll need to get on the guest list.

» RSVP for The New Rules of Video Marketing

Chris Savage, CEO of Wistia, thinks that’s backwards. On a recent episode of Zero Click Marketing, he said:

His logic is simple. Everyone wants numbers. So everyone takes the budget and puts it into the things most likely to show a return which are, of course, the same things everyone else is funding. And now you’ve signed up for the most expensive, most saturated marketing efforts. Meanwhile, the stuff that no one can track is probably where most of the opportunity is, because this is the heavily under-leveraged channel.

Savage has the receipts. Ramp, a B2B finance company, put Brian from The Office in a cubicle in Times Square and filmed ads around him. No dashboard told them to do that. But it was funny, it tugged a very specific nostalgic heartstring for the millennials who are buying expense software, and it got people talking in a way no LinkedIn carousel ever will.

Wistia’s version of this: a full-company goodbye song for the Canon 5D camera — thirty people singing a corporate, “Because You Loved Me.” Savage’s review: “the most ridiculous video. Makes no sense, and it just did so well.” And here’s the part that never makes the case study: the team loved making it. “When you give yourself permission for this type of work, it really does lift everybody.”

That’s the tell, by the way. The ideas that scare your CFO are usually the ones your team is itching to make.

So what’s the actual playbook? Savage’s is refreshingly concrete: if you have a budget, carve off 10% and “treat this stuff completely differently. We’re gonna hold it to a completely different bar. It’s gonna be like a taste bar. It’s gonna be a creativity bar. It’s gonna be, like, what literally makes us laugh.” The other 90% of the budget still answers to the numbers. The 10% answers to taste.

I asked him how someone should pitch one of these ideas to a skeptical leader, and his answer was pure operator: don’t lead with the idea, lead with the competition. “Ramp is eating our lunch.” Frame it as FOMO — competitors are already doing this, the talent that knows how is scarce, the investment is small next to the saturated channels that aren’t working anyway.

And as the marketing practitioner, the one who’s usually in the weeds with the rest of the time, I’d add one more angle: morale. If the team’s energy is flat, say so — “the team would be really excited about this.” Chris confirmed it from experience. The Canon 5D video wasn’t just a hit with the audience; it was a hit in the office. Permission to make weird stuff is a company culture boost. It’s just… fun.

None of this is new, exactly. It’s the budget-line version of what Jay Acunzo told me on this show: don’t market more, matter more. It’s unsolicited response rate (URR) with a spreadsheet column. The ideas that make people seek you out were never going to fit neatly in an attribution model. You can’t fake the thing, and you can’t fully track the thing… and those are the same thing.

The pipes are saturated. The dashboards all measure the same five channels. The opportunity is in the stuff that makes your CFO nervous and your team light up. Fund the untrackable. There’s a reason the untrackable stuff wins, and Wil Reynolds — on today’s episode — named it: “Google and AI are just proxies for human trust.” Savage’s 10% is the human-trust budget. The 90% optimizes for the proxies. The 10% earns the thing the proxies are trying to measure.

And now, two calls to action because, I am, of course, a marketer. A chaotic one as of late. (IYKYK, more than one CTA is bad, I know, I am not perfect, and do as I say, not as I do.)

1. Listen to the full conversation with Chris — we get into why he launched Talking Too Loud as audio-only at a video company, his “I just watched you on a boat” theory of trust, and what he’d do starting from scratch tomorrow (spoiler: hand-to-hand combat, duh).

2. And if your CFO needs something to look at while the 10% does its thing, steal Wil Reynolds’s favorite metric: for every piece of content, what share of its traffic comes from social and direct — the WhatsApp forward, the Slack drop, the link someone pasted into a Google Doc — versus the algorithmic proxies for trust? Somewhere, a human decided your thing was worth remembering.

🍬 Petits Fours

Four delicious considerations because this newsletter is, after all, called the “Menu.” (I know; I have too many names for things. The cobbler’s children have no shoes. I, on the other hand, have too many shoes, too many purses, and too many Pilates grippy socks.)

📖 The Zero Click Marketing book is almost here. My and Rand Fishkin’s book comes out next month — last call to preorder through our publisher for the bundle deals and bonuses (including a $50 SparkToro credit). But also… we’re officially live on Amazon for preorders!!! So if the shipping cost is a no for you dawg, you can order through Amazon. (Again, too many CTAs, I am chaotic evil right now.)

🎥 Hang out with me next Tuesday, and let’s make better buyer personas: Office Hours with Rand. I’m co-hosting SparkToro Office Hours on making better personas (and doing better marketing) — Tuesday 9/29 at 11am PT. Register free on Goldcast, or sign up to get the recording.

🎧 Listen to or watch the backlog of episodes on the Zero Click Marketing podcast: Life has been crazy lately, and I haven’t sent this newsletter in a while. Mea culpa. I think you’ve still been keeping up, but if not, be sure to listen! We’ve had some great guests on the past few weeks, and still more to come, including Sarah Stockdale, Ben Putano, and Rand Fishkin (again! He’s sick of me and my bs, I know it.)

🎮 Are you hooked on Pokemon Pokopia the way I am? I suspect most of you will not care, but I’m really into this game, and I’ve been building little tributes to the other environments within Palette Town. Here’s my beach build!

If you liked today’s newsletter, forward it to a friend who’s tired of defending every dollar with a dashboard. If you hated it, reply and ruin my day.

Thanks for reading The Menu by Amanda Natividad! This post is public so feel free to share it. Every time a post gets shared, a marketer gets a click. (Just kidding, they don’t.)

When I saw the chart, my reaction was: I don’t care.

I don’t care because the week-by-week volatility of an LLM’s retrieval system doesn’t influence my marketing strategy — and it shouldn’t influence yours either.

By the way, if you want the audio version of this (and no, this isn’t a transcript), you can listen to that on the recent new episode of the Zero Click Marketing podcast. Oh! And if you’re new here, hello and welcome! Yes, I have a podcast. It’s often short-form monologues, so easy to binge or easy to check out one before deciding whether ya wanna commit.

If you’ve been investing in Reddit because your audience genuinely spends time there, then that means you’re reading the conversations in your category. You’re learning how people describe their problems. You’re answering questions when you have something useful to contribute. Maybe you’re watching people compare products, recommend vendors, complain about the incumbent solutions, or tell each other what they wish existed. That’s valuable marketing. It’s audience and customer research, rolled into one. You found value there, and increased AI visibility was the happy byproduct of your efforts.

Let’s talk a little bit more about how these underlying citation systems are incredibly unstable. Semrush recently ran the same 100 prompts through two different ChatGPT reasoning modes. Only 25.6% of the cited domains overlapped between the two. When ChatGPT used higher reasoning, it ran 4.6 times as many internal searches and cited nearly twice as many sources per answer.

So… we don’t even have one consistent thing called “ChatGPT citations.” Two modes of the same product can take the same basic question and go searching through very different parts of the web. Now add Gemini. Perplexity. Google AI Overviews. Claude. Copilot. Future models that haven’t been launched yet. Add product updates, ranking tweaks, retrieval changes, licensing agreements, and whatever happens next Tuesday.

Trying to organize your marketing strategy around whichever source is currently over-indexing in one of those systems is exhausting and a waste of time.

The wonderfully boring news

There was another piece of research I found much more interesting than the Reddit drop.

Profound recently analyzed 11.84 billion citations across eight AI systems. Across all models, roughly 57% of citations came from brand websites. Even ChatGPT, which relies on those brand sites the least in its answers, still accounted for 47% of all citations. 69% of all citations on Gemini are from brand sites.

Promptwatch found a similar pattern from a different angle. In its analysis of millions of classified ChatGPT citations during July, product pages were the biggest content type at 32.8%. Back in March, product pages had accounted for roughly 18%. So citations for product pages are up.

While marketers (or… our bosses?) have been furiously asking which external platform the AI systems love today, an enormous amount of the information these tools retrieve continues to come from… websites. And it’s awesomely boring. Make your website gooder. Explain your products well. Publish things worth knowing.

If you have original research, make it available. If you have expertise that nobody else in your industry has, put it somewhere people can find it. Make your documentation clear. Give journalists and creators reasons to reference you. Participate in communities where your customers already spend time.

Over time, you start building your various homes and apartments across the internet. Your website is the property you own. And then you rent your audience on LinkedIn, YouTube, or on a niche Subreddit. You don’t control those rented properties, and you don’t need to because together, they create a visible digital footprint. They give people multiple opportunities to discover you, understand what you know, and decide whether they trust you. And of course, they also give AI systems evidence to read and use.

Increased AI visibility is a benefit of good marketing

Today’s citation winner can become tomorrow’s 80% decline without your company changing a single thing. So if you like, sure, keep up with that news, but only as long as you’re focusing on the most durable, lasting question of all:

Where does my audience spend time, and am I doing something valuable enough there that they know who I am and trust me?

If the answer includes Reddit, keep doing Reddit.

If it includes your website, LinkedIn, YouTube, industry publications, newsletters, podcasts, communities, events, or some combination of all of them, build those things too.

Create enough useful, credible presence across the internet that when people go looking for information about your category, they — and the AI systems — can find you.

Then if ChatGPT changes its mind about Reddit again next quarter, you can look at the chart, learn something from it, and carry on because you’ll still have a marketing strategy.

🥮 Petits Fours

4 random/fun things that I think are worth your time because this newsletter is, after all, called, “the Menu.”

💙 One of my love languages is a solid, healthy, loving roasting by a dear friend. The friend in question is Jay Acunzo. I recently joined his show, “Why They Resonate,” where he gave me feedback on my MicroConf keynote talk. He helped me think through my context-setting (did we need THAT much of it? Maybe?), and I called myself a f-in idiot. That’s always fun.

🗣️ That MicroConf talk, by the way, now has over 60,000 views which is pretty crazy in my world! If you are one of the lovely people who watched that talk and became a subscriber, thank you very much for your support. I try to make this newsletter worth your while, and I don’t force it… meaning if I’m really strapped for time, this falls to the wayside because I don’t want to send you garbage.

😰 Why Is Everyone In Tech So Sad? My friend Sarah Stockdale sent me that essay and it’s quite good. (Yes, my little analog hobby is making sourdough bread.) Sarah, by the way, is hosting a personal brand summit with her company Growclass. Reply to/ping me if you need a discount. I may be able to get one.

👾 I really want to show off my Pokopia build: This is the cozy interior of my house in Withering Wasteland. How cute is this?! Ok, I’m sorry, this really wasn’t worth your time if you’re not playing Pokopia, but at least I didn’t make you click elsewhere to see it.

But the episode opened with an example I’d given them: how old is Paul Rudd? You Google it, Google tells you, done. You don’t need to visit Wikipedia, or IMDb, or a People magazine article — unless you’re ready to pore over beauty secrets to Paul Rudd’s apparent immunity to aging. That little example ended up framing the whole episode. They used the stat that 68% of U.S. Google searches now end without a click, and the rest of the story chewed on the same question we’d been kicking around: what happens when the platform just hands you the answer, and the website that actually did the work never gets the visit?

So, yeah. My literal quotes were brief. My ideas went further. And that’s what influence usually looks like — a lot smaller on the tape than it was in the room.

We love to measure the visible thing. The click, the quote, the referral, the nine minutes of airtime, the tracked conversion. It’s the part we can point at. But the part we can point at is almost always smaller than what actually happened. Someone reads your LinkedIn post and repeats your argument in a meeting you’ll never sit in. A reporter builds a story around your example and runs one sentence of the interview. Somebody hears your company on a podcast, files the name away, and searches for it three days later — showing up, helpfully, as “direct” traffic from nowhere. A buyer reads a genuinely good idea of yours, never clicks a single link, and still walks away understanding you better than they did that morning. The work did something. Your dashboard just can’t retrace its steps.

Which is a big reason I’ve gotten skeptical of treating attribution like a record of what worked. Attribution can tell you what got the credit. It can’t really tell you what caused the result — and sometimes it can’t even tell you where the thing came from. A couple of years ago at SparkToro we ran an experiment: we sent more than 1,100 visits across 11 social networks, then checked what Google Analytics had to say about it. A big chunk showed up as Direct, with no clue which platform actually sent it. It’s not that the traffic was untrackable. The platforms just weren’t telling us.

And even when you can see the activity perfectly well, the attributed number will happily walk you off a cliff. Dropbox’s data science team ran month-long blackout experiments on mobile advertising and search engine marketing. On paper the attributed return on ad spend looked great — 1.53 for mobile, around 2.0 for SEM. But the causal ROAS, the return that actually depended on the ads running at all, came back at 0.70 for mobile and 0.92 for SEM. Both underwater. As a result, Dropbox moved $25 million away from that low-incrementality spend, and its portfolio LTV:CAC improved by 53%. The attributed story and the causal story weren’t just a little different. They flat-out disagreed about whether the money was working.

Note that I’m railing against attribution specifically. I’m not mad at measurement. I love receipts. Given the choice I’d take an itemized one, ideally with pictures... but that’s not a thing. But be honest about what a receipt can actually prove. Sometimes it undersells the work — the way nine minutes of NPR undersold a 40-minute conversation. Sometimes it gives first prize to whichever activity happened to be easiest to see.

So look at your numbers with your eyes wide open. What’s missing from the picture? Who owns the data you’re leaning on? What would you genuinely do differently if the number moved? And when being wrong gets expensive — millions of dollars, a whole channel — can you actually test whether the activity caused the result, instead of just taking its word for it?

Petits Fours

Four small things worth your time this week.

1. I went deep on attribution on Humans of Martech.

I joined Phil Gamache and Darrell Alfonso to talk about why attribution keeps quietly lying to us: dark social that shows up as “Direct,” HubSpot’s alligator graph, and the Dropbox study where a perfectly healthy-looking ROAS was actually losing money. If you saw my post this week on attributed vs. causal ROAS, this is the hour-long version with all the tangents left in.

» Listen to the latest episode of Humans of Martech

2. Inc. asked me how to actually pick an email platform.

I got to use my favorite framing for this: “A CRM is the customer ledger. An email marketing platform is the communication engine.” The piece also has my actual advice, which is to prioritize integrations first, segmentation second, automation third, and to stop paying for a complex marketing operations system when what you needed was a few reliable automations.

» Inc’s article on choosing an email platform (great for non-marketers!)

3. Zero Click Marketing, explained for my mom (or your mom… or for all the normies).

This week’s episode of my own podcast is the one I made after going on NPR’s The Indicator and realizing I’ve never explained what I do in plain English. No jargon, no mention of LTV:CAC. Just Paul Rudd’s age and the 68% of US Google searches that now end without a single click. If someone in your life doesn’t know what zero click whatever means, this is nine minutes that might help.

» Listen to the 9-minute episode of Zero Click Marketing for YOUR MOM (lol)

4. Zero Click Summer School starts next week.

…which feels like the right moment to mention the book is up for pre-order. Zero Click Marketing, co-written with Rand Fishkin, out this October. If any of the above resonated, the book is where the whole argument lives.

» Last chance to preorder the Zero Click Marketing book and join us for our exclusive webinar series!

I’m often more interested in the philosophies behind marketing — how we choose our campaigns, which ideas are worth exploring, what deserves more of our attention — than in the campaigns themselves.

So when Tim Soulo asked me a question on the Ahrefs podcast that I couldn’t immediately answer, I did what any thinkboi would do: I’ve been turning it over ever since.

The question was:

If you can’t justify spending money to promote your content, how did you justify spending the time and effort to create it in the first place?

I mean, if an idea is valuable enough to research, write, edit, publish, and distribute organically, shouldn’t it also be valuable enough to advertise?

Maybe. But I don’t think creating and promoting are the same decision.

You create and build upon an idea because it deserves to exist. You pay to promote it because you decide a specific audience, channel, and campaign are the best use of a finite budget right now.

An idea can be worth making without necessarily being an efficient advertisement. Some ideas hit immediately. Others travel slowly through blog posts, podcasts, webinars, presentations, conversations, and social posts before becoming part of how people understand a subject.

Still, “the idea matters” can become a convenient excuse for publishing and hoping. So the better question may be:

What kind of support does this idea deserve?

Sometimes the answer is paid media. Sometimes it’s a webinar, podcast tour, keynote, a book (cough cough), or a permanent home on your website.

The obligation isn’t to advertise every idea you believe in. It’s to give the best ones more than one chance to matter—and to be honest about whether you’re actually doing that, or just hoping.

I wrote more about this on the SparkToro blog…

🧁 Petits Fours

1. The debate that started it

Tim Soulo invited me onto the Ahrefs podcast to debate Zero Click Marketing, content distribution, attribution, personal reputation, and whether three great posts really beat five mediocre ones.

Despite the confrontational episode title, this was a generous, thoughtful conversation — and Tim asked questions that forced me to examine some of my own assumptions.

» Listen to/watch my conversation with Tim Soulo on the Ahrefs podcast

2. Every marketing yes contains a no

On my own podcast, I explored one of the defining skills of modern marketing: deciding what not to do.

Every time you choose a campaign, channel, optimization project, or new initiative, you are also choosing what will receive less time, money, and attention. The episode is only 8 minutes, because I respect both opportunity cost and your podcast queue.

» Listen: Marketing Is Choosing What Not to Do

3. Make what people want. Let your values lead.

I also joined the Do Less Marketing podcast to talk about building an audience, making work people actually want, and letting your values shape how you grow.

We discussed audience research, useful content, reputation, and the tension between doing what performs and doing the kind of work you believe should exist.

» Listen to my appearance on Do Less Marketing

4. I can’t stop thinking about Obsession

I recently watched Obsession, and I can’t stop talking about it with my husband. We’ve been happily gobbling up every interview with the lead actors and director, and we’ve become huge fans of Inde Navarrette. (I can’t remember the last time I was this captivated by an actor!) Every choice she makes, the control over her voice, the control over facial expressions, is incredible.

I won’t spoil the movie. But should you watch it, please come back and discuss it with me because I need more people to discuss it with.

» Streaming on Peacock (subscription required)

What decision will this measurement help you make?

In my recent conversation with Rachel Elsts Downey, founder of podcast agency Share Your Genius, she explained that her company doesn’t obsess over every newsletter open, blog visit, or individual content conversion.

She looks at the bigger picture: whether the company’s audience is growing over time. More importantly, she looks at how many leads the business is generating, whether those leads are qualified, and whether they’re turning into the number of new customers the company needs to hit its revenue goals.

That’s it.

I’m not arguing against looking at newsletter performance. Open rates can reveal deliverability problems. Clicks can show which subjects interest readers. Replies obviously tell you a lot.

But before you spend hours inside a dashboard, ask what you’ll do differently because of what you find.

Would one weak newsletter convince you to abandon email? Probably not.

Would one low-performing LinkedIn post change your broader content strategy? I hope not.

Would learning that your team spends 20 hours a week producing a newsletter while qualified leads are falling short cause you to reconsider the investment? Yes. That information should change a real business decision.

The closer a metric gets to a meaningful decision, the more attention it deserves.

Rachel’s approach also creates room for experimentation. Her team knows the number of leads and customers the business needs. As long as marketing is helping produce them, the person responsible for marketing gets room to test ideas rather than endlessly optimize small things.

That freedom matters because each marketing channel requires more work than it appears to from the outside. A Pinterest account isn’t just a Pinterest account. It needs a strategy, expertise, production, distribution, and enough time to learn what works.

Trying to appear everywhere usually means doing mediocre work everywhere.

A better approach is to find the intersection between two things:

1. Where your audience already pays attention

2. Where you’re willing and able to create consistently

Then choose a channel. Build a repeatable system. Give it enough time and volume to teach you something. Only then consider adding another.

You don’t have to pursue every plausible marketing opportunity and monitor every available metric.

You need enough information to make good decisions about what to keep, what to improve, and what to stop doing.

That’s what measurement is for. We covered lots more in this week’s episode of the Zero Click Marketing podcast.

📖 Preorder Zero Click Marketing and Join Us for Summer School

Google answers questions directly. Social platforms would rather keep audiences inside their feeds. AI tools summarize the useful parts of your work without necessarily sending anyone to your website.

The clicks are disappearing. The influence isn’t.

That’s why Rand Fishkin and I wrote Zero Click Marketing: a practical guide to earning awareness, trust, and demand in the places your audience already pays attention — even when there’s no neat trail of clicks leading back to you.

And for preorder customers, the book comes with several extras.

Zero Click Summer School begins in just a few weeks.

We’re hosting two live, one-hour sessions on August 4 and August 11 to help readers put the book’s ideas into practice. We’ll go deeper into the frameworks, work through real marketing questions, and give you a chance to learn directly with us.

Both webinars will be recorded, but the recordings will only be distributed to people who purchased the book.

So yes, come October, you’ll get the book. You’ll also get two hours of practical instruction designed to help you start applying it before the hardcover even reaches your shelf.

Preorder Zero Click Marketing today to reserve your place in Summer School.

And should you happen to know a marketer who keeps asking where all their traffic went, consider forwarding this to them. We may not be able to restore the old internet, but we can help them stop marketing as though it still exists.

🥮 Petits Fours

4 random/fun things that I think are worth your time and if you disagree at least I didn’t take up too much of your time.

This is what I mean when I say good PR is your new AI visibility strategy: My friend Clare Sayas interviewed me for her Substack, the Boilerplate! It was fun digging into zero click marketing or rather, zero click PR with a seasoned PR pro. A great read especially if you’re new to my work.

Join this Content Marketing Institute webinar with me: I’m speaking at Content Marketing World this fall (swoon!) and in advance of that, I’m presenting how to track success of zero click marketing/content in next week’s webinar. CMI is legit so you know this will be the perfect (free) way to learn from me!

An at-home cocktail tip: I love making simple but mostly-proper cocktails at home. And I’m an even bigger fan of making these things easy and accessible. So one of my go-to “recipes” is this: a 1:2:4 ratio of sweetener, citrus juice, and spirit. I rarely even make simple syrup anymore; I usually use orgeat (you may be able to find this at any big box grocery store now!) or maple syrup. In my suggested ratio, a cocktail might be: 1/2 oz maple syrup, 1 oz lemon juice, and 2 oz bourbon. Or you can sub bourbon for gin, vodka, tequila, or brandy. Yum.

Oh also, did you know Hibiki Harmony is not super expensive anymore? You probably did; I’ve just been out of the loop. This is the last bottle I bought — it was February 2020, my last grocery run before shelter-in-place orders were enforced. And I’ve rationed out the whisky since. I bought it for $75 and back then, that was a steal! There was undersupply of Japanese whisky and for a time (2018-2020, maybe?), not only was Hibiki Harmony nearly impossible to find, when you did, it would cost $125+ (at least in the Bay Area). I came across a new bottle at Whole Foods just the other day for $75! And then I learned that Japan had build more distilleries to meet demand. Hibiki shortage no more.

His summary of the space, verbatim: it’s “a 10,000-piece puzzle that has no blueprint and that changes parts every week.” Comforting.

Here’s what stuck with me most…

Oh, and of course, you can listen to our full conversation on today’s Zero Click Marketing podcast. (YouTube link if you wanna see our cute faces.)

People really don’t click anymore, ugh

From Kevin’s own studies and from outside work (he pointed to Pew): people are not clicking citations or sources inside AI answers. They’ll click a buy or purchase button. They mostly won’t click through to go read the underlying page.

The exception is what Kevin calls “high risk” prompts — medical, insurance, finance — where people do more fact-checking before they trust the AI’s answer at face value.

Ghost citations: the AI version of getting used but not credited

This was my favorite part of the conversation. A ghost citation is when your brand is cited as a source behind an AI answer but never actually named in the answer itself.

Kevin walked through the mechanics: an LLM retrieves content, decides what to cite, and separately decides what to mention by name — and those are three different filters, not one. In his research, only about 13% of brands were both cited and mentioned in the same answer.

The clearest example he gave was G2. Ask an AI “what’s the best CRM?” and it’ll pull heavily from G2 to form the answer — G2 gets ghost-cited constantly. But the answer itself will say “the best CRMs are HubSpot, Salesforce, etc.” — not “according to G2.” G2 gets used, not named. Reddit and LinkedIn work similarly: retrieved constantly, credited rarely.

If you’re a review site or an aggregator, that tradeoff is probably fine — it’s the business model. But if you’re a SaaS company or a DTC brand, Kevin’s point is blunt: citations barely matter because almost nobody clicks them. What matters is whether you get mentioned by name in the answer itself.

One more wrinkle: prompt length changes the behavior. Short prompts (”best CRM”) get more brand names thrown around because, absent greater, more specific context, the AI is trying to give you as many options as it thinks you need. Longer, more specific prompts narrow things down to fewer, more targeted mentions.

AI reads like a rushed editor, not a patient reader

Kevin’s data (drawn from analyzing over a million search results) points to AI systems reading content top-heavy — a lot of weight on the intro, a smaller bump of attention at the conclusion, and not much patience for anything else. His advice: write like classic journalism. Lead with the punchline. Structure it like a table of facts, not a story.

Here’s a great quote from him:

“If you wanna write for AI, it needs to be Wikipedia. Facts clearly structured, no storytelling, just all the most important information in the beginning.”

He used Malcolm Gladwell as the counterexample — a writer who deliberately withholds information to build toward a reveal. Great for humans. Terrible for AI, which wants the nugget up front, not a story arc. That zero-click tension — write for the machine that decides your visibility, or write for the human so they can enjoy reading it — is one neither of us has a clean answer for.

Trust is the currency now

This is the thesis Kevin kept coming back to, and the stat that backs it up is the one I keep repeating to people: when Google shows an AI Overview at the top of results, clicks to Reddit and YouTube go up by roughly two-thirds. Not because those sites are cited more (though YouTube increasingly is) — it’s because people read the AI’s answer, and then go looking for a human to confirm it actually sounds right.

The second data point: when people use AI mode to shortlist options (best laptop, best insurance plan), they pick the first item on the list about 75% of the time — rank still matters. Except when they recognize and trust a brand on that list. Trust beats rank order every time it shows up. As Kevin put it:

“Trust is not just like a conversion booster. I think it is the most important good in this AI world, since content is easier and faster to create than ever before, and we’re competing with more synthetic content.”

Worth sitting with if you’ve noticed the flood of obviously-AI-generated “storytime” threads doing numbers on Threads and LinkedIn lately — the ones that read a little too smooth, land a little too neatly. We both agreed: it’s working right now, and it’s also eroding trust that’s expensive to rebuild once people clock it.

How to actually measure this stuff

If you’re just starting to track AI visibility, Kevin’s cut against a lot of the current tooling default (which is mostly “run a prompt once, see what comes back”). His recommendation, in order:

1. Share of voice / mention rate — how often you’re named in answers, against the total field of brands being named. Citations matter far less than mentions unless you’re an aggregator or review site.

2. Shortlist rank — are you showing up when AI narrows a category down to a handful of options? That’s the closest thing to bottom-of-funnel intent this channel has.

3. Sentiment and attributes — not just “are we mentioned” but how — are you being described as trustworthy, cheap, reliable, effective? AI mirrors the consensus about your brand back to users, for better or worse.

He also flagged that single-run prompt tracking is close to useless — model outputs are noisy enough that you need to run the same prompt multiple times (and ideally across different personas / jobs-to-be-done) to see what’s actually stable versus random variance.

His closing argument, and mine too: performance marketing spent two decades training marketers to stare at dashboards instead of talking to people. AI search is forcing that muscle to come back. If you don’t actually know what your audience believes about you — or whether they’ve even heard of you — the AI answering their questions doesn’t know either.

More from Kevin: he writes Growth Memo, currently going out to just over 20,000 subscribers. If you found this lil ol’ newsletter edition useful, that’s the best next read.

Listen to the full conversation wherever you get your podcasts.

🧁 Petits Fours

Four random things that I hope you’ll enjoy. If not, maybe dip these in ranch.

There’s still time to preorder my book and join us for Zero Click Summer School: Rand and I are hard at work on our presentations for our upcoming exclusive webinars. They’ll be in early August (and yes, you’ll get the recording!). If you want an invite to this sort of zero-click boot camp, then preorder our Zero Click Marketing book directly from our publisher. If the shipping fee is too hefty for you (lookin at you, my lovely community outside the US), you can opt for the digital bundle of audiobook + ebook. Grab the book, join us for the webinars, and support indie publishers. Yay!

An ode to ranch dressing in WaPo (free link, I think): Coming back from America’s 250th birthday weekend, I’m proud of us for producing ranch dressing. I have loved it with french fries, sandwiches, and occasionally sprinkled over chicken then roasted, for years. It tickles me to know tourists from around the world have been given friendly TSA warnings to make sure they pack those ranch bottles in their checked luggages.

A lesson in girl math: I never shopped at Lululemon. And then one day my husband told me we’d get a $75 American Express credit if we bought something there. I begrudgingly looked around and got a pair of their Daydrift wide leg trousers. They’re comfy as hell (which annoyed me because I think I’m supposed to be forever-angry at them for their CEO being a jerk… but upon googling I have just learned that he’s no longer CEO so yay). And then months later, a different color went on sale for $79. This means the pair that just arrived at my doorstep the other day only cost me $4! I will spend the rest of my life chasing this high.

And finally, the smartest thing I’ve ever seen in my life:

If you liked today’s newsletter, the next best thing you can do is forward it to a friend who’s looking for AEO and marketing advice. If you hated it, then feel free to reply and ruin my day.

Boring intros aside, today I have a Field Notes episode of the Zero Click Marketing podcast for you. I talk about what Similarweb’s Downstream Impact of AI Visibility research means for marketers. My co-author Rand Fishkin has the solid breakdown for ya.

On my podcast, I talk about the implications for AI on the buyer journey, which includes how AI creates, captures, nurtures, and protects demand.

Although I suppose if you don’t have time for any of this, then these are the important parts:

Similarweb was trying to answer the question: when AI introduces or recommends a brand, does that recommendation appear to change what people do afterward?

Yes.

Users who received an AI recommendation were 2 and a half times more likely to visit that brand’s website within the next 7 days.

And when they did visit that website, how did they show up? Through branded search.

Somehow, I promise there is more on the 13-minute episode of the podcast. I talk too much, probably.

Anyway, that’s kind of all I have for you. My wrist is really acting up, I think from just typing and mousing too much. Yes, I have the Logitec MX vertical mouse. Yes, I love it and it helps, but sometimes enough is enough for ya body. And we are in copy edits for the zero-click marketing book, so I need to save my wrist strength for that. Exciting!

🧁 Petits Fours

Four things I’m doing/consuming/enjoying in some way…

🐟 What I’m eating: salmon, the Creole-inspired sheet pan edition. Via Epicurious. I decided to just buy an Epicurious sub for the year and rely on that for my weeknight dinners. I’m too drained and overstimulated most days to think of a new recipe, but I also want to change it up more often. For some strange but wonderful reason, my kids really love fish. So I try to cook it often.

👯‍♀️ What I’m listening to: “Deja Vu” by Olivia Rodrigo. I know, not new. I still hate running, but the storytelling in this song is excellent and it’s fun to run to. You don’t have to be an Olivia Rodrigo fan to appreciate her lyricism.

😂 What I’m laughing at: This little girl doing the “Gigachad” face is my new hero.

🎮 What I’m playing: Pokemon Pokopia. I love cozy games and this is my favorite one right now. Are you playing? If so, my lil pro tip is to go to a dream island, fully break down the house that’s there (walls, windows, roof and all), and then rebuild it in one of the story areas.

I think a lot more CEOs, CMOs, and founders are about to start asking this. And I get it. For 20 years, “how visible are we?” had an answer. Maybe not a great one, but an answer: rankings, organic traffic, branded search, share of voice, reviews, the analyst report someone paid too much for.

But buyers aren’t only searching anymore. They’re asking. They’re asking ChatGPT, Gemini, and Perplexity for the shortlist, the comparison, the tutorial. So the executive question becomes: are we showing up there?

I think the better question is: are we being described correctly?

That’s what we dive into on today’s episode of the Zero Click Marketing podcast.

This piece is sponsored by HubSpot, and I took them up on it because their AEO Grader gave me something concrete to poke at: my own company. Quick caveat before you screenshot me — I don’t think any AI visibility tool is gospel, and I’d gently suggest you not let a chatbot reorg your positioning. But marketers need a starting point, and a score is a fine place to start. It’s just not the whole story.

Here’s what I mean by mirror…

When I ran SparkToro through the grader, some of it was affirming. It pegged us as a niche audience research tool, picked up on our credibility with marketing practitioners, and understood that we help marketers figure out where their audience spends time — what they read, watch, listen to, and follow. Useful. Not because I needed the ego boost (though I’m human, compliments are nice), but because it confirmed the broad idea is landing. The public signals around us are strong enough that AI systems mostly get what we do.

Other parts needed a human read. We got lumped in with Semrush, Ahrefs, Similarweb, Brandwatch, and a few others. Some of those are fair-ish neighbors — if you’re researching a market or sizing up competitors, I can understand the overlap. But we’re not trying to be a broad SEO suite or an enterprise social listening platform. So the reflex is to say “the AI got it wrong.”

And maybe it did. But it’s more useful to ask why. Where did that comparison come from? Is the category itself confusing? Are buyers genuinely using those tools as mental substitutes for us? Is our public record (the third-party articles, review sites, comparison posts, podcast transcripts) sending mixed signals about where we belong?

This is where positioning becomes really important. April Dunford (you can’t talk about positioning and not mention April Dunford, this is marketing law) describes positioning as context-setting: it tells people what market you’re in, what you replace, who you’re for, and what they should compare you to. So if an AI tool drops you in the wrong competitive set, that might be an AI problem. But it might also be a *you* problem — the public internet may not have enough clear, consistent evidence about where you actually belong.

Which means you have to resist two instincts, both wrong.

The first is taking the output as truth. “Gemini says we’re in this category, so I guess that’s the category.” No.

The second is waving it off. “That’s dumb, the AI is stupid, moving on.” Also no.

Because even when the output is wrong, the wrongness is information. If AI systems believe something off about you, there’s a decent chance some slice of your market believes it too.

So when I read one of these AI visibility reports, I’m not just asking whether we’re visible; I’m asking whether we’re understood.

Here’s the practical version. If your CEO asks about your AI visibility, don’t hand them a number and call it done. Run the grader, then sit with three questions:

1. What’s accurate? Those are the signals worth reinforcing.

2. What’s wrong or incomplete? That’s where you’re getting miscategorized, or where the nuance is falling out.

3. And what does all of it say about your public record? Maybe you need clearer website copy, a better comparison page, sharper case studies, more third-party proof.

Then pick a short list. Nobody will actually work through your 47-item AEO roadmap, and they’ll resent you for building it. Pick the handful that matter: one product page, one comparison page, one customer story, one partner description. Fix it. Then get the idea off your own site. Talk about it on a podcast, put it in a newsletter, post it on LinkedIn, work it into a webinar.

This is the part I care about most, because it’s what I keep saying about the public record: if your best proof is trapped in a sales deck or your founder’s head, it’s not helping you yet. (Wil Reynolds ran a great experiment on this — an old “high turnover” complaint kept surfacing in branded AI answers until Seer published their actual retention data, and the citations flipped.) Give the web — and the systems that learn from it — more accurate evidence to work with. Then run the analysis again in a month or two and see what moved.

You’re not optimizing for robots. You’re clarifying what your market, and the machines that read your market, believe about you. And if they’ve got it wrong, well.. now you know what to fix. That visibility report is a reflection of what the internet thinks about you. And if you don’t shape the story the web tells about you, something — or someone — else will.

🧁 Petits Fours

Four things I’m doing/consuming/enjoying in some way…

🍛 What I’m eating: Current obsession is basil chicken stir-fry by Shilpa Uskokovic (recipe on Epicurious and Bon Appetit; pick your freemium poison). It’s cheap, easy, and super fast to make. I didn’t have sambal oelek so I used the chili garlic oil in my pantry. And I couldn’t find holy basil at the grocery, so I substituted regular Italian basil and mint, and a heavy pinch of black pepper. It doesn’t yield the exact same flavor, but it hits the various sweet, spicy, bright, and peppery notes you get in holy basil. Oh also, I threw in a handful of frozen spinach as well to sneak more veggies into the meal. My toddler, little kid, husband, and I loved it. The only other universal winner in my household is my basic spaghetti with meat sauce.

🏃‍♀️ What I’m listening to: I don’t identify as a Swiftie, but I’m discovering a love for a select few Taylor Swift songs. In particular, it’s “Cruel Summer” (I know, not new), and boy, is this song fun as hell to run to. I hate running, I hate cardio, I try to avoid it. But I also want to be able to move more without wanting to die of fatigue. Now I’m doing regular run-walks to sad-ish music that makes me feel like the main character. Sprinting to: And I snuck in through the garden gate / Every night that summer, just to seal my fate / And I screamed, “For whatever it’s worth / I love you, ain’t that the worst thing you ever heard?” / He looks up, grinnin’ like a devil is as cathartically good as letting out a good cry. You’re welcome.

😂 What I’m laughing at: That feel when you accidentally use the wrong personality at work…

⚽️ What made me smile: “Somebody said that World Cup is just the US having a giant sleepover with the cousins we never see because our parents hate each other.” [Twitter/X]

Most likely, you look them up online to learn a little more about them. It’s why my friend Jay Acunzo said to me in this week’s episode: public speaking is the ultimate form of zero click marketing.

And the more I sat with it, the more I realized it’s a stress test for everything we make.

Jay has been giving keynotes professionally since 2016. And in a talk, a “click” is quite literally physically impossible. A speech forces you to do the one thing most marketing tries to avoid: deliver discrete, undeniable value right now, in the room. You can include a QR code in your deck, and yes, attendees might use it in real time, but that isn’t really the goal for anybody. It’s not any attendees’ goal to transact in that moment, and when people speak publicly, they don’t always have something to sell. (Er… except for people like me who are trying to sell their book. But that’s sort of an exception.)

“Your ideas — and therefore, the value and originality of your thinking — should compel people to take an action. You shouldn’t need an overt call to action.” —Jay Acunzo

That’s zero click marketing in its purest form. Not “never have a CTA,” but: be so good that the seeking-out happens on its own.

He even has a (made-up, and frankly perfect) metric for it: unsolicited response rate, or URR. The rate at which people come to you without being asked. Because when the response is unsolicited, it’s more qualified. The catch, of course, is that you can’t fake URR. You actually have to be good.

The bar for “good” isn’t reach. It’s resonance.

This is where Jay said: “Don’t market more. Matter more.”

And related to that:

“Reach is how many see it. Resonance is how much they care.”

“Don’t be the best — be their favorite.”

I added one amendment, which I told him on the show: the more you obsess over resonance, the further your reach eventually goes anyway. Resonance isn’t the opposite of reach. It’s the on-ramp to it.

But mattering is harder than marketing. It takes creativity, risk-taking, and often, just sitting with the problem a lot longer than your competitors are willing to. Jay’s diagnosis for why so much content underperforms is brutal and correct: you’re at letter Z. You’ve lived with your idea so long that you’re standing at the very end of understanding it. Your audience is back at letter A. And most of us, because we’re so close to it, skip straight to “here’s how to do it” — the solution — over and over again.

“You have not gotten them ready to care or to act. Maybe that’s why it’s hard.”

Good communicators do the slow part. They meet people at A. They name what the audience wants, agitate the real problem, surface the deeper problem underneath it, and only then offer a premise. By the time they say “here’s how,” the audience is already leaning in, asking for it.

(This is, not coincidentally, the exact conversation that nudged me to stop treating “clicks are disappearing” as obvious and actually build the argument from the ground up. It’s a big reason the book Rand and I wrote exists.)

If you take one thing from this episode, make it this. Jay asks:

For a lot of marketing — especially the churn-and-burn, brute-force, dig-the-hole-before-it-caves-in kind — the honest answer is no. Which is the actual reason a traffic decline feels existential. If you’re not saying anything that compels people to seek you out, you were always going to be exposed the moment the pipes changed.

And the pipes are always changing. Jay’s reminder that distribution channels aren’t neutral systems of pipes — they’re ad networks with opinions about what should spread — is something I think about weekly. Instagram rewards what makes it more ad inventory. LinkedIn wants you teaching 101-level stuff to the widest possible room. None of that necessarily maps to the audience you actually want, or to the work that’s actually best for them.

A keynote, though? Forty-five minutes, no algorithm in between. As Jay put it: “I’d much rather say one thing one time that my audience thinks about 10 times a day than get in front of the audience 10 times a week.”

Influence and impact. Not followers and traffic.

If you’ve ever put someone you admire on a pedestal for how effortless they make all this look, Jay would like you to know:

“The pedestal is made of crap.”

It’s built from bad drafts, dead-end soundbites, and rants that went nowhere. Creativity is trial and error — try the idea on like a shirt off the rack, decide it doesn’t fit, add a little seasoning, try another one. The people who look like they have a superpower mostly just refused to skip that part.

Listen to the Zero Click Marketing podcast wherever you get your pods, or watch it on YouTube.

As for Jay, everything he teaches lives at jayacunzo.com — his essays, his newsletter, and his relaunched podcast, Why They Resonate, where he breaks down what makes great speakers actually land. If you want to get better as a speaker, thought leader, or storyteller, start there.

Watch me get gently roasted. In a couple of weeks I’m going on Why They Resonate so Jay can pick apart a talk I’ve been giving this past year. I get to learn; he gets to roast.

🍪 Petits Fours

Four delicious considerations because this newsletter is, after all, called the “Menu.”

Another reason to stop treating “more traffic” as the goal: The people who still click are often the ones with harder, more specific questions that AI couldn’t fully answer. That means generic “what is X” content is less useful, while original research, real experience, proprietary data, and sharper analysis become more valuable. More by Jon Kelly in Hacker Noon.

Voice of reason Anil Dash on the “inevitability” of AI: “…it’s possible for creators working in their own communities to now make AI tools that serve their specific needs, without causing all the harms that make people object to Big AI. This feels like the true alternative to the narrative of “inevitability” that so much of the hyper-funded AI industry is trying to push, while also not forcing people into a quiet life of AI guilt if they still find some utility in some aspects of these tools.”

Nintendo Switch 2 price increasing on September 1: I like cozy games but I don’t play often, so I was unsure about getting the Switch 2. Buuuut… since the price is going up in a few months, my husband surprised me with the console plus Pokemon Pokopia. The game is absolutely delightful. Highly recommend.

Is it weird to listen to your friends’ podcasts to keep in touch in a lazy way? Yes? Yeah, I thought so. Anyway, my latest favorite listen is my friend Asia Orangio’s In Demand podcast. Asia’s one of the most brilliant people I know, and an original thinker. Check it out if you wanna get smarter about B2B growth.

I’ve made the case before that AI optimization is mostly just good marketing. Today I want to zoom in on the part that I think matters most — the part AI flat-out cannot do for you, no matter how big or advanced the model gets.

It’s lived experience.

This is something my friend Kaleigh Moore and I got into on today’s episode of Zero Click Marketing. Kaleigh’s a writer and content strategist who spent seven years on a Forbes retail column before going deep on AI search, and she thinks about this stuff more rigorously than almost anyone I know.

A large language model is a phenomenal line cook. Hand it a recipe that already exists somewhere on the internet and it’ll plate the dish faster and cleaner than you can. What it cannot do is taste. It wasn’t on the customer call this morning. It didn’t watch the feature flop and ruminate on why. It has no idea what it actually feels like to do the job — because it has never once done the job.

Kaleigh put it about as plainly as it gets: when you find the person who’s hands-on with the product, talking to customers, solving the weird problems, running the experiments — that’s a goldmine, because it’s a point of view that isn’t published anywhere else. It’s all firsthand. That’s the whole thing. AI can remix what’s already out there. It can tell you the most statistically-likely word that should follow the previous one. It cannot generate what only exists in someone’s head.

Which means the scarce, defensible, un-copy-pasteable input in your entire content operation might be you. And the people already on your payroll.

To be clear, this more than a “let’s protect humans” argument. The models themselves seem to reward it. The material that gets pulled into AI answers, cited, surfaced, skews toward the authentic and the experience-based — precisely because that’s the stuff a model can’t fake its way into. When everyone is generating, the un-generatable becomes the premium ingredient. Scarcity goes up, not down.

So the question isn’t about how we can 10x our output. It should really be: what do we know that no one else could possibly know?

A few ways to actually act on this:

Stop using your experts as editors. Start using them as sources. Most companies save their sharpest operator for the very end of the process to “review for accuracy.” That’s backwards. The person closest to the product and the customer isn’t a fact-checker — they’re the richest raw material you have. Put them at the start.

Capture; don’t manufacture. The experience already exists. It’s in the company lunch-and-learn, the answer someone gives in a meeting, the Slack thread where your engineer explains why the obvious solution is actually wrong. Kaleigh’s line that stuck with me: film one hour-long internal talk and you’ve got weeks of content. You’re not creating from scratch; you’re writing down what’s already being said.

Make it sound like a person, not a press release. We talked about a company whose employee posts work precisely because they read like coworkers who happen to like their jobs — not a commercial in a trench coat. The lived-experience advantage evaporates the second you sand it into corporate beige. Keep the specifics. Keep the opinions. Keep the slightly-oversharing part.

(Getting a reluctant expert to actually open up is its own challenge — a whole separate issue I’ll get to. For now, assume the gold is in there. It almost always is.)

The freeing truth is that the content arms race isn’t won on volume. It’s won on the stuff only you could say. As the cost of cranking out generic content drops to roughly zero, the value of the genuinely firsthand goes up. Your competitive edge isn’t a tool you need to buy. It’s already there — in your brain, in your emails, in your support tickets, debating you in the weekly standup.

Go write them down.

…I mean, after you listen to or watch the podcast. 🙃

🥖 Petits Fours

Four delicious considerations because this newsletter is, after all, called the “Menu.”

🚨 My co-author Rand Fishkin’s latest research is out, and oof: in the first four months of 2026, 68.01% of Google searches ended without a click. That’s up from 60.45% in 2024, and it’s the fastest acceleration we’ve seen in the last decade. Traffic really is no longer the clean proxy for marketing value it used to be. Read the full breakdown on the SparkToro Blog.

🌶️ I’m speaking on Zero-Click Marketing: The Secrets to AEO: Sadly, I had to miss Chili Piper’s in-person Chilipalooza (a sneak-attack norovirus will do that to ya). But on the bright side, this means you’ll be able to join my virtual make-up session! We’ll talk about what actually happens to marketing strategy when AI answers your buyer’s question before they ever click your site — and what marketers should optimize for instead. See ya there!

🎟️ For marketers trying to keep up without losing their minds: I’m speaking at Proofed, Growclass’s one-day virtual marketing summit for mid-career marketers who feel like everything is changing way too fast. We’ll be talking AI, search, content, social, and what actually matters now. My session is The Zero-Click Playbook: How to Build Demand Before the Website Visit, where I’ll share a practical framework for creating value where your audience already is, building preference off-site, and measuring beyond traffic. Tickets are $75, and it’s happening on Zoom in a couple days!

🥚 I’m making Mayak Eggs, or Korean marinated eggs, this week. I’ve never made them before but I love rice and eggs so I have no doubt they’ll become my new obsession. Have you made them before? I’m using Cookerru’s recipe. Send me your fave recipe!

See ya next week, cuties.

But my good friend Talia Wolf says it’s the other way around. When she said that to me in today’s Zero Click Marketing episode, my eyes lit up as I eagerly waited her rationale.

Talia is a conversion optimization expert, the founder of GetUplift, and the author of Emotional Targeting. She’s spent years digging into why people actually buy — not the reason they give, the real one. Her claim is blunt: a B2B purchase carries more emotional weight than almost anything you’ll buy for yourself, because the stakes are personal in a way a new cardigan never is.

Think about what’s actually riding on it. Buy something dumb for yourself and you’re out a little money and some pride. Pick the wrong software for your company and the fallout has your name on it — the budget for the year, and your standing with everyone who signed off. Talia’s example is the one that stuck with me. You’ve been told to switch email platforms. Now you’re up at 2am eyes wide, catastrophizing the what-ifs: What if nobody can use the new one? What if everybody hates it? What if you migrate and the whole list vanishes in some freak accident, and everyone knows it was your fault?

That’s not a rational buyer.

Talia splits it into two ideas:

• Self-image: how I want to feel about myself.

• Social image: how I want everyone else to see me.

The person choosing enterprise software doesn’t just want the product to work. They want to feel like they made the right call — and they want everyone else to see that too. None of that lives in a feature list. Brands lead with “we have AI” and “all-in-one platform,” then wonder why their genuinely good product isn’t clicking with anyone.

If you’re competing on features and pricing, you’re probably already in a knife fight with five companies that look a lot like you. Once a buyer confirms you fit the budget and check the boxes, the thing that tips them toward you over the other two finalists isn’t another spec. It’s whether you make them feel understood: whether you clearly know what they’ve tried before, why it failed them, and what’s keeping them up at night.

In our interview, Talia gets specific about how you actually pull that off — including a comparison page she rebuilt around what an agency owner is really afraid of (not “here’s our profitability report,” but “here’s how you’ll finally know whether you can take on the next client”) that drove 54% more free trials from organic. We get into why “save time” is an empty promise until you know whose time, and what they’d do with it once they got it back. And why hiding your pricing in 2026 just hands your story to ChatGPT, which will cheerfully get it wrong on your behalf.

It’s one of my favorite conversations we’ve had about the part of marketing that lives underneath the funnel — the actual human on the other end of the purchase, doing math about their own job.

🎧 Listen to the full 30-minute episode wherever you get your podcasts, or watch it on YouTube.

And if you want to go deeper, Talia’s book Emotional Targeting is worth the short flight it takes to read.

🥖 Petits Fours

Four delicious considerations because this newsletter is, after all, called the “Menu.”

A dip for your next picnic, gathering, or pity-party of one: I’m obsessed with Trader Joe’s Parmesan tapenade. I’ve already gone through two lil tubs of it, smeared lovingly over my homemade sourdough bread. It’s cheesy, it’s herby, it’s salty, garlicky and spicy, and it has a zesty tang from red wine vinegar. I spread it over bread and toast it, or dunk my bread directly into the tub as I stuff my face over the kitchen sink.

Make your voice heard in the State of Marketers survey: Research is good when it helps people. It’s even better if you participate in it too. In last year’s State of Marketers survey, more than 1,000 marketers participated, and the report helped people feel seen and benchmark their experiences. This year’s survey takes about 6 minutes, and the more marketers who contribute, the richer and more useful the final report will be.

Because it’s nice to read good writing: The Atlantic is one of the few print subscriptions I have, and I’ve found that lately, the pieces I’ve been reading there are more in arts & humanities space. It’s just extra refreshing these days to read human-written prose. Enjoy this gift link for “The Night My Marriage Fell Apart.”

A fake skincare funnel with a real marketing lesson: I built a fake-but-very-plausible quiz funnel for Medik8 to show what growth automation can actually look like when it’s not a seven-tool duct-tape nightmare. The bigger point is that small teams can’t afford to lose leads in the gaps between forms, enrichment, email, SMS, CRM, and checkout anymore. Every drop in the bucket has to land in the cup. So I walked through how Typeform can now collapse a lot of that customer journey into one place — from quiz capture to enrichment to follow-up to payment. Listen/watch if you want the nerdy funnel breakdown with a side of skincare obsession.

Medik8 is also the kind of brand that would benefit enormously from a smart quiz funnel.

Skincare is the category for personalization. Nobody walks into a Sephora knowing the exact two products that they need. At best, they go in expecting to buy Too Faced’s Born This Way The Natural Nudes Eye Shadow Palette, but end up walking out with Makeup by Mario’s Ethereal Eyes: The Original palette. (Shoutout to Makeup by Mario for this maturing lady’s makeup needs.) Everyone wants someone to tell them what to use. Brands like Curology and Function of Beauty built entire businesses on this insight — get someone to answer six questions, deliver a personalized recommendation, then convert them into a subscriber.

A Medik8 quiz funnel could be magic for them. So here’s what I’d do if I were running their marketing — and why I think this is a useful exercise for any small DTC brand. Below is my fully fleshed out idea, but you can listen to today’s episode of the Zero Click Marketing podcast for a more conversational version. Details are here in the newsletter, the thought process is in the pod.

What the quiz funnel needs to do

The basic shape is familiar:

A customer lands on medik8.com and sees an opt-in: “Build your Medik8 routine in 60 seconds.” They take the quiz — skin type, primary concern (anti-aging, brightening, breakouts), routine complexity preference, budget — and get a personalized 3-step routine at the end. They drop their email for the recommendation. Then the brand has 30-90 days to convert them into a buyer.

That’s the front-end. The back-end is where most small brands lose the plot. Not that it’s their fault. It’s just that personalized marketing is that hard. Because to do it “right” with the conventional stack, you’d need:

• A quiz tool (Typeform, Octane AI, Outgrow)

• An email service provider for nurture emails (Klaviyo, Mailchimp)

• An SMS platform for high-intent leads (Postscript, Attentive)

• A lead enrichment tool to figure out who’s actually filling out the quiz (Clearbit — RIP — Apollo, Crossbeam)

• A CRM (HubSpot, Pipedrive)

• Zapier to ping someone on Slack when a high-LTV lead shows up

• Stripe / Shopify to take the money

That’s seven tools. Anywhere from $400-$1,500/month at small DTC scale. Seven dashboards, seven APIs, seven places where data can go to die. And critically: seven handoffs where leads get dropped, mis-tagged, or sent the wrong follow-up because the SMS tool didn’t know the email tool already unsubscribed them.

If you’re a small brand, you literally cannot afford this stack to run inefficiently. So most brands compromise — they skip the enrichment, or they skip the SMS, or they let the CRM rot because nobody has time to clean it. The funnel works at a fraction of its potential.

What I’d actually do… now

I’d build the whole thing in Typeform.

Not because I have brand loyalty to Typeform (I don’t, particularly — I just used their forms for years like everyone else). But because they just built out something called Growth Flow that collapses most of the stack above into one place. I’ve been poking at it for the past few weeks and it’s the most surprised I’ve been by a marketing tool in a long time. And yes, this newsletter is sponsored by Typeform, which is why I got early access.

Here’s how the Medik8 quiz funnel maps to it:

Capture. The quiz lives in Typeform — same as it would anywhere. Six questions, pretty, on-brand.

Enrich. This is the part that blew my mind. The moment a customer submits with her email, Typeform pings third-party data providers and pulls back a profile — age range, location, professional details if it’s a work email. Match rates in the 70-90% range. This is what Clearbit used to do as a standalone product before HubSpot bought it and put it down. Sad Amanda has been waiting years for a replacement. It was hiding inside a form tool.

So now Medik8 knows: this lead is a 40-year-old in Los Angeles who works at a tech startup. The quiz only asked her about her skin.

Organize. A contact record auto-creates. Every future form she fills out — CSAT survey, restock notification, referral signup — ties back to the same record. No CSV, no nightly Zap, no manual deduping.

Act. A workflow kicks off automatically. Welcome email with her personalized routine (C-Tetra Luxe + Liquid Peptides in the morning, Advanced Night Ceramide at night — that’s just my routine, but you see how this works). Three-day delay. Social proof email featuring customers in her enrichment-derived demographic. Seven-day delay. Targeted discount code. If she scored as a high-intent lead based on her quiz answers, she gets an SMS instead of waiting for the next email. The Medik8 team gets a Slack ping when a high-LTV lead converts. Amanda gets plump, glowing skin in about eight weeks (hey, beauty takes time).

Close. Checkout via Stripe block inside the email flow. She never has to bounce to a separate ecommerce experience for the first purchase. (Bigger brands will still need full Shopify integration for inventory and subscription management — but for the first conversion off a quiz, this is enough.)

Understand. AI-generated insights tell the marketing team which quiz paths are converting, which segments are dead weight, and what’s trending up.

That’s the entire funnel in one tool.

Why this matters beyond Medik8

The reason I’m writing this isn’t really about Medik8 or Typeform. It’s about a pattern I think is going to define the next few years of marketing tools.

The unbundled SaaS era was a luxury of cheap traffic. When leads were $2 and you had thousands of them per week, it was fine to lose 10% to tool-handoff issues. The math worked. Now leads cost 5x more and arrive at 1/5th the volume, and every drop in the bucket has to land in the cup. Stitching together seven best-of-breed tools made sense when traffic was abundant. In a Zero Click world, the cost of stitching is higher than the cost of compromising on a few features.

This is the same pressure that has Notion turning into a CRM, Linear into a docs tool, and Typeform into a growth stack. Watch what the small teams are switching to. They’re voting with their wallets.

If you’re a founder or a small-team marketer right now, the question is no longer “what’s the best [thing] in [category]?” It’s “what’s the fewest tools I can run my entire customer journey through?”

The answer is probably smaller than you think.

And Medik8, seriously — call me.

This essay is a paid partnership with Typeform. They asked me to play with Growth Flow and write about it honestly. I did. The Medik8 fandom is unsponsored and possibly clinical.

Peep my YouTube channel if you want the video version of this episode of Zero Click Marketing.

🧁 Petits Fours

Four delicious considerations because this newsletter is, after all, called the “Menu.”

🔥 Rand Fishkin wrote a banger on why “make great content” is no longer enough: The argument is spicy but useful: if Google and AI tools can summarize, remix, and commoditize your content before anyone clicks, then the more defensible move is to build something they can’t replace with an answer box — an inimitable product, service, experience, or community. Content still matters, but increasingly as marketing for the thing, not the thing itself. Read it here.

🍋 My favorite vitamin C serum: Look, nobody asked (ugh) but I really do love Medik8 C-Tetra Luxe. That’s not an affiliate or referral link or whatever. It is just my duty as a girl’s person (I dunno, maybe you’re not a girl who’s reading this) to not gatekeep great products. I love their Advanced Night Ceramide too, but I think it can be a little strong for sensitive skin.

💙 My good friend Talia wrote a truly excellent post on optimizing for conversions in a zero-click world: The gist: your buyers are forming opinions about you long before they hit your website — in AI answers, Reddit threads, YouTube reviews, LinkedIn feeds, etc. So your job isn’t just to “show up everywhere.” It’s to understand what people already believe, fear, and need to hear by the time they do land on your site. Talia connects zero-click marketing with emotional targeting in a way that’s very smart, very practical, and very worth your time.

📚 A little book update: The Zero Click Marketing book crossed 500 preorders in two weeks, and I’m feeling very grateful about it. If you’ve been meaning to order, preorder bonuses are still available. And if you’re outside the U.S., consider grabbing the audiobook bundle so you can skip the hefty international shipping fees. You’ll snag your preorder bonus, Rand and I will sell more books, and it’s a win for indie publishers. Big hugs and thank you to everyone who has supported so far.

It’s a fair starting point. But the better discussion to be had is deciding which parts of a piece should be freely available, and which parts are actually valuable enough that someone would trade their email for them.

Sometimes the gated content fails that test. The form itself isn’t the problem; the problem is what marketers keep putting behind it. Gate the tool, not the framework. Gate the deeper benchmarks (if you must), not the headline findings. Gate the template, the calculator, the implementation kit — the thing that helps someone do the work — and let your actual ideas be read and crawled by the internet. This is what I break down in today’s episode of the Zero Click Marketing podcast.

A form’s friction is sometimes fair. If I’m registering for a live session with someone whose time is worth my attention, take my email. If I’m downloading a template that will save me three hours or accessing a real benchmark, fair trade. But if I’m trading my email for a 9-page PDF that tells me to “define my goals, know my audience, and measure my results” — I’m not your lead. I’m a hostage.

This is the bargain that broke. The original deal — give us your email, get something genuinely useful — got stretched into something worse, where buyers were asked to pay for information they could find faster elsewhere. It worked for a while because there wasn’t much elsewhere. Now there is. There’s more content on the internet today than there ever was before, and a lot of it is free (as in, no email trade required). AI search, Reddit, YouTube, podcasts, private communities, and your competitor’s ungated blog post are all sitting between your form and a buyer’s curiosity, and they’re winning that race more often than not.

That changes the job of content. Your best ideas need to be findable. They need to be quotable. They need to show up in the public record around your category — in AI answers, in search results, in the Slack channels where buyers actually decide what they think. If your sharpest argument is hidden behind a form, you’ve protected the asset and weakened its reach at the same time.

You got the email, but did you earn the memory? You got the download, but did you actually change how the buyer thinks? You got the MQL, but did you create demand, or just count it?

Because a form fill is not demand generation. It’s a countable event that may or may not represent intent. Sometimes the person filling it out wants to talk to sales. Sometimes they want the template. Sometimes they’re a competitor snooping around. Sometimes they’re racing against the clock to complete a slide deck, and they just need one stat that helps them make their case.

But we persist anyway with our gated content because downloads, registrations, and cost-per-lead fit into dashboards. Influence doesn’t; it’s mostly invisible. But now, when buyers form opinions long before they fill out a form, influence is what actually matters. It’s someone citing your framework in a meeting three weeks after they read it. It’s your name showing up in an AI answer because the model has seen your ideas most often. It’s the buyer arriving on the sales call already warmed up because your thinking was doing the work before your attribution software could see it.

Before gating the next ebook or “toolkit,” try a better set of questions. Would a smart buyer with options find this gate annoying, or fair? Is the asset actually valuable enough to survive in a world where AI, Reddit, and your competitors will all happily answer the same question for free? And — the one that stings — are we gating this because it serves the buyer, or because we need something countable for the report?

Ungate the argument. Gate the utility. Let your best thinking travel. Then, when you do put something behind a form, make sure it’s worth the email.

Much more over the podcast.

Petits Fours

Four delicious considerations because this newsletter is, after all, called the “Menu.”

Aleyda Solis and the Finchling team launched Trending PR Campaigns: This is a free resource that shows the PR campaigns earning the most media coverage right now, updated daily, ranked by coverage, and browsable by industry. Finchling also explains why each campaign gained traction and what marketers can learn from it, which is especially useful if your current PR brainstorm process is “stare at a calendar and hope a relevant but admittedly obscure awareness day appears.”

Lily Ray built an LLM Content Visibility Scanner: This free tool helps you catch obvious visibility issues, showing what AI crawlers can actually see when they fetch your page’s raw HTML. If your important content depends on JavaScript to render, some LLM crawlers may not see it at all.

HubSpot has an AEO Grader: The tool gives you a one-time diagnostic of how ChatGPT, Perplexity, and Gemini represent your brand, including sentiment, presence quality, brand recognition, share of voice, and competitive positioning. I’d treat any score like this as a starting point; it’s a quick way to see what answer engines think they know about you.

Have you pre-ordered Zero Click Marketing yet? I’m hearing rumors. Alarming ones. Anyway, pre-ordering directly through our publisher gets you the bonuses: SparkToro credit, an Alertmouse discount, and a seat at Zero Click Summer School. If you’re outside the U.S., international shipping is… rude. So if you want the bonuses without the shipping costs, grab the audiobook + ebook bundle instead. We’ll likely open pre-orders through Amazon, Kobo, and other retailers at the end of summer, but the bonuses are tied to publisher pre-orders.

This is the argument behind today’s episode of the Zero Click Marketing podcast: GEO/AEO is real, but it is not magic. The work is part technical, part editorial, part credibility-building, and very much a zero-click problem. Listen here (or watch on YouTube if that’s your thing).

I heard someone recently ask:

“What’s the ideal H2 header length so that it’s optimized for LLMs?”

I get it. We’re all trying to understand how AI systems retrieve, parse, summarize, and cite information. There is a real technical layer here.

But also… shouldn’t a heading be as long as it needs to be while still being concise? Weren’t we supposed to be doing that anyway?

Then I saw someone else say, “LLMs have started to discover video. And it might not be that people are watching these videos — the LLMs are. Now it’s more important than ever to do video.”

Yes, AI systems may increasingly use video transcripts, captions, and metadata as part of the information they retrieve.

But weren’t people already making videos? Weren’t captions already useful? Weren’t transcripts already good for accessibility, search, repurposing, and plain ol’ human comprehension?

This is the part of the GEO/AEO conversation that makes me feel like I’m the one who’s losing my mind here.

Before we go further, a quick terminology caveat — partly for you, partly for me. I’m aware this is a fast-changing topic, and so people often use terms like AEO, GEO, and even AI and LLMs interchangeably. (Present company included.) I think of Answer Engine Optimization (AEO) as one piece of the larger Generative Engine Optimization (GEO) umbrella.

AEO is about making content easy to surface as an answer in chatbots and search. GEO, more broadly, is about making your content, brand, and public evidence easier for generative systems to synthesize, cite, or mention, in an AI-generated response.

Next, I acknowledge that AI search changes things.

Ahrefs’ February 2026 update found that the presence of an AI Overview correlated with a 58% lower average click-through rate for the top-ranking page. So sure, when clicks are dramatically lower, it feels like what you have left is the chance at visibility.

Meanwhile, Cyrus Shepard recently analyzed 54 AI citation experiments, patents, and case studies, then ranked the factors most associated with AI citations. His TL;DR was refreshingly sane: most of the critical factors still align with traditional SEO practices. Or, in his words: “win SEO, win AI citations,” most of the time, with extra steps.

The visibility stakes are changing. The mechanics are changing. But a lot of the “new” advice keeps arriving at suspiciously familiar destinations: clear structure, specific answers, credible evidence, useful formats, and content people can actually understand.

Great. Love this for us.

I just refuse to pretend the pressure washer is groundbreaking when all we had to do was rinse the spoon in the sink.

“AkShUaLLy, There’s a Technical Side Too.”

1/ Retrievability

I know. I don’t disagree with that. In GEO, AEO, LLMO, or whatever acronym we are arguing about this week, of course there is a technical side: crawling, rendering, retrieval, indexing, sitemaps, structured information — all of that matters. In fact, Cyrus’s top-ranked factor was URL accessibility: whether a page is available and crawlable during training or grounding. He notes that this has become more complex as AI companies use different crawlers and as more sites deploy AI-scraper-blocking protections.

In the Zero Click Marketing podcast episode I recorded this week, I refer to this retrievability: can AI-mediated systems access the thing in the first place?

But retrievability is only the first layer.

2/ Extractability

The second layer is extractability: once the system finds your content, can it easily understand and reuse the useful parts? Here’s where GEO starts to look suspiciously like good writing.

Clear definitions. Structured information. Useful headings. Concrete examples. Evidence close to the claim. Comparison language. Original frameworks. Standalone statements. Video when video helps people better understand the thing.

Cyrus also calls out factors like AI-ready structure, factually specific content, explicit phrasing, cited sources, and self-contained passages as characteristics associated with AI citations. In other words: structure the page clearly, make specific claims, show your work, and write important points so they can stand on their own.

Say the thing, explain the thing, and prove the thing. I’ll say… it’s nice (I guess?) that more people are caring about all this. But it’s writing for humans’ comprehension. It’s not really writing for the robots.

3/ Credibility

Once the system finds and understands your content, does it have a reason to trust it?

You know those listicles called like, “Top 20 Marketing Experts to Follow?” and it’s the same 20 people except for the addition of one rando who — surprise! — is the author of the shitty listicle? Or those 2,000-word articles called, “What is Customer Research? A Complete Guide,” and it’s written by someone who has never spoken to a customer, never conducted research, and appears to have learned about the topic by reading eight other articles called — surprise! — “What Is Customer Research? A Complete Guide”?

AI search sees this human slop (we can’t blame AI for this particular sin) and makes the problem weirder. Sometimes it summarizes the lukewarm consensus. Sometimes it pulls in the thing that stands out. And sometimes the thing that stands out is not the thing you wish it had found.

So no, credibility doesn’t mean you can force AI to cite your “complete guide.”

It actually means you need a stronger public record than generic content, stale claims, and whatever random artifact happens to be most distinctive.

Credibility means your content gives people — and search systems — a reason to trust this answer from you and you specifically.

That can mean original research. First-party data. Expert quotes. Clear methodology. Customer examples. Specific experience. Screenshots. Named sources. Author expertise. Transparent limitations.

And — surprise! — this is not just an AI thing. It’s a trust thing.

4/ Public evidence

And finally, public evidence. This is the most zero-click part of GEO.

Because AI systems and buyers do not only learn from your website. They learn from Reddit threads, YouTube videos, LinkedIn posts, reviews, comparison pages, third-party mentions, communities, and many more discovery surfaces.

Unfortunately, those are all the places you don’t own… which is largely what makes them more credible to your audience.

So if the public record around your brand is thin, confusing, outdated, or nonexistent, you have a problem.

• What does the internet seem to know about you?

• What do customers say?

• What comparisons exist?

• What questions come up repeatedly?

• What misconceptions go uncorrected?

• What phrases are associated with your brand?

• What proof exists outside your own website?

Your website matters, yes. But so does your audience’s language, the trail of proof, the conversation around your brand, and whether you and your category are visible enough in credible places that both humans and machines can connect the dots.

The goal is not to follow every hack to trick the robots. It’s to become more trustworthy. First, by people. Then, by robots.

So… now what?

Say it with me, fellow marketers who love nuance: “It depends.”

Because it’s not all about “just good writing,” fretting over your robots.txt or focusing on PR. The answer — surprise! — is somewhere in the middle.

GEO is technical enough that marketers need to understand retrievability. Can the system access the page?

It’s editorial enough that we need to care about extractability. Can your page be understood, summarized, and cited?

It’s evidentiary enough that we need credibility. Can this answer be trusted?

And it’s zero-click enough that we need public evidence. Does proof exist beyond your own website?

Because when search becomes an answer engine, your content has to do more than attract traffic. It has to earn its place in the answer. Achieving this is not that different from what good marketing has always been.

Be findable. Be useful. Be credible. Be memorable. Show up where your audience learns. Say something worth repeating. Create evidence that outlives the click.

And for the love of God, stop washing your spoons with the pressure washer. Just turn on the faucet.

🧁 Petits Fours

Four delicious considerations because this newsletter is, after all, called the “Menu.”

💰 Join this webinar on CRO, hosted by me, led by Talia Wolf: Most CRO advice is tactical noise: tweak the button, rewrite the CTA, move the form, repeat until everyone loses the will to live. Talia Wolf has a smarter take. She’s the author of Emotional Targeting and CEO of GetUplift, and at SparkToro Office Hours, we’ll talk about how to optimize across the full customer journey — not just the website. Save your seat or get the replay here.

🎁 Content Marketing in a Post-SEO World: I’m joining my friend Matt Lerner for a live session on what content marketing looks like when organic traffic is harder to win, social platforms penalize links, and AI answers questions before people reach your site. We’ll talk Zero Click Marketing, measurement, and what founders should do when they have limited time and no content team. Register here.

⭐️ Proofed, a one-day Growclass marketing summit: Sarah Stockdale, CEO of Growclass, is hosting a virtual summit for mid-career marketers who feel like everything is changing too damn fast. The day covers AI, search, social, content strategy, Claude Code, and yes, Zero Click Marketing! I’ll be speaking at 1pm ET on The Zero-Click Playbook: How to Build Demand Before the Website Visit. Register here.

📖 Psst! Pre-order my book? We launched preorders for the Zero Click Marketing book. If you order directly through our publisher, Damn Gravity, you can get bundles that include the hardcover, ebook, audiobook, or even consulting with me — plus special deals from SparkToro and Alertmouse. Outside North America or waiting for Kindle? Amazon and third-party retailer preorders should open later in August, though those won’t include the direct pre-order bonuses.

I can finally say the thing… I am… an author?!

Rand Fishkin and I are so excited to announce the thing we have obsessed over for 15 months:

Zero Click Marketing: How to Build Brands and Earn Customers on a Traffic-Starved Web.

Pre-orders opened today. It comes out this fall.

I first coined the term “zero-click content” back in 2022. I had no idea it would become a thing, let alone define the next era of marketing.

The short version of why this book exists:

Brands and their websites are starving for traffic. But most marketing teams are still marketing like it’s 2018.

Google answers the question on-page for two-thirds of searches. Small publishers lost about 60% of their Google traffic last year. Social platforms actively punish you for linking out. ChatGPT summarizes your blog post before anyone ever opens it. AI referrals? Less than 1% of open-web traffic, despite every headline telling you otherwise.

And your boss still wants a last-click attribution report by Friday 🙄

Zero Click Marketing is the honest, tactical, occasionally unhinged, and aggressively footnoted guide to earning attention, building an audience, and growing a business on a web that has mostly stopped sending clicks.

It’s not a “how to go viral on TikTok” manual. (Good luck out there!) It’s also not one of those More-Successful-Than-Thou airport business books where every chapter should have been a blog post.

It’s the playbook Rand and I have been running ourselves for years. It works. We have the receipts.

Why pre-order today:

🎓 Zero-Click Summer School. A live online workshop series Rand and I are teaching this summer, 🚨EXCLUSIVELY🚨 for pre-order readers. Where we go deeper on the ideas and tactics the book only had space to set up. Pre-order customers only. Miss it and you miss it.

📦 Pre-order only bundles. Grab the hardcover + ebook or the audiobook + ebook, plus a $50 SparkToro credit and 50% off any Alertmouse plan for the year. There’s also a bundle with one with an hour of 1-on-1 consulting with ME (capped at 25 seats, so don’t dawdle).

💡 Pre-orders actually matter. Our publisher uses pre-order sales to convince bookstores to carry the book, so we can reach the normies! We’ll also use early sales to determine the print run size (and I want a mountain of books). Seriously though, pre-ordering really helps us and we will appreciate you forever.

Your options:

• Hardcover + ebook, $31.50

• Audiobook + ebook, $31.50

• Consulting Edition — everything above plus a full hour of 1-on-1 with me (prep, session, follow-up). 25 seats, period. $999. If you’ve got a real ZCM problem you want me inside of, this is the one.

• Team buys:

  • 15% off at 10+ books

  • 20% off at 20+ books

  • Want to buy LOTS more? Send us an email 😊

Some of you have been following my work on Zero Click Marketing for a minute. You’ve watched this whole argument get formed in public… one post, one “but Amanda, how do you measure that?” reply, one conference talk at a time.

None of this would be happening without Damn Gravity, our publisher and partner in turning this sprawling, occasionally unhinged idea into a real book. Huge thanks to Ben Putano and the Damn Gravity team for believing in Zero Click Marketing and helping us bring it into the world.

And if you want the companion rant/audio experience, the newest episode of the Zero Click Marketing podcast also dropped today. It’s a good one to queue up if you want more on why this shift is happening, what marketers should do about it, and why I am apparently incapable of having a normal hobby.

It feels nice to finally be able to hand you the whole thing as a single bound object, with an actual index, and a cover, and everything.

Clicks are dying. But marketing is just getting interesting. 😎

Hugs,

Amanda

I was explaining the deck like, “Here’s how I think marketing works now.”

Matt, who comes from more of a growth marketing background, saw it and basically said, “This is content marketing.”

The funny thing is, I don’t think either of us was wrong.

Because yes, the deck talks a lot about content. Native-to-platform posts. Search behavior. Public record. Reddit. AI answers. But to me, none of that is just “content.”

It’s how buyers understand you. Learn to trust you. Remember you. It’s how they decide you’re worth seeking out later.

That’s not a content problem. That’s a business problem.

For a long time, content marketing was treated like a traffic function. Publish the blog post. Rank for the keyword. Get the click. Capture the lead. Attribute the conversion. Repeat.

That system was never as clean as we pretended it was, but at least the story made sense.

Now the story is breaking.

Search keeps growing, but more searches end without a click. Social platforms punish links and reward native content. AI answers can shape perception before someone ever sees your homepage. Reddit threads, podcasts, creator mentions, comparison pages, and branded searches all influence the buyer journey in ways your analytics will never fully show.

So yes, content is involved.

But the job is bigger than publishing. The job is to create understanding, trust, recall, and preference wherever your buyer already is. That was one of the final ideas in my deck: the new job of digital marketing is not merely to send people to your website. It’s to build those four things across the places where decisions are actually shaped.

This is also why Cyrus Shepard’s recent research on Google winners and Kevin Indig/AirOps’ research on ChatGPT citations felt so timely to me.

Both point toward the same idea that broad content alone is becoming less defensible. The winners are more useful, more focused, more distinct, and harder to replace.

Cyrus’s research looks at which sites are winning in Google. His analysis suggests that the winners are more likely to offer a product or service, allow task completion, own proprietary assets, maintain topical focus, and have a strong brand.

Kevin and AirOps looked at ChatGPT citations and found something different but related: pages that closely match a user’s query seem to have an advantage over pages that try to cover everything.

Put those together, and the classic content marketing instinct starts to look shaky.

The answer is no longer to write the most comprehensive guide — it’s to build a destination worth seeking out that allows the user to fulfill their long-tail query or even better, their transaction.

That’s where the business problem comes in. And by the way, I cover this in more detail in this week’s Zero Click Marketing podcast:

…And this is where I think Matt’s reaction helped me see the tension more clearly.

A content marketer may look at Zero Click Marketing and see an argument about content distribution.

A growth marketer may look at it and see a debate about channels, conversion paths, and measurement.

A founder may look at it and see something else entirely: a question of whether the company is distinct enough to be remembered when the click never comes.

I think all of those readings are valid.

But I also think we’re reaching the point where they are converging.

The “content problem” and the “business problem” are becoming the same problem.

Because when clicks were abundant, you could get away with content that existed mostly to capture demand. You could write the “ultimate guide,” rank well, earn the click, and let the website do the rest.

But when clicks get scarcer, the content has to do more work before the visit ever happens. It has to explain the category, create trust, generate memory, and shape the public record. The content has to help Sales, Customer Success, Comms, Product, and Demand Gen do their jobs better. And if content has to do all this, then… it’s a function of business infrastructure.

This is the part I want marketers to take more seriously.

If we keep framing content as the stuff that fills the blog, we will keep fighting for scraps.

But if we can show that content shapes demand, supports sales, reduces friction, builds trust, strengthens the public record, and gives people a reason to seek us out directly, then we are not talking about a publishing calendar anymore.

We are talking about how the business earns attention in a world where attention is harder to move.

So maybe Matt and I weren’t disagreeing after all.

Maybe we were just looking at the same problem from two sides.

He saw content marketing. I saw business strategy. And increasingly, I think those are becoming the same thing.

And you know what? I don’t totally know where our conversation is going to go.

I mean, I know what we’re going to talk about. I just don’t know what we’re going to talk about.

If you want to see that play out live — and by “live,” I also mean “sign up now and get the recording later” — join us for the webinar.

🧁 Petits Fours:

Four discrete, bite-sized thoughts because this newsletter, after all, is called the “Menu.”

Content as a Service (CaaS): The Framework That Aligns Marketing with the Business: You know how several paragraphs ago I said, “The content has to help Sales, Customer Success, Comms, Product, and Demand Gen do their jobs better?” Well. I have more on that! Check out this past blog post of mine where I teach you how to treat Content as a Service. The best content has a job to do and a client to serve. (I also talk about this in my keynote.)

“Dad, will you generate me a story?”: That’s the punchline of one of my favorite marketing destinations — Marketoonist. I got to meet creator Tom Fishburne at this week’s Experience Inbound, and I’ve loved his work for years. I was really stoked to get the firsthand experience of what a lovely human he is. You probably already read his comics but if not… then yay! Ya learned something new.

HubSpot’s Field Notes: Ok, now this is funny. I also met Kyle Denhoff this week (another lovely human!) and after swapping funny stories about our kids, we nerded out about making things, and we talked about our podcasts. Embarrassingly, I didn’t know the name of his show and I laughed when he told me it was “Field Notes” — because, I said, “I swear! I didn’t copy you!” I have a recurring segment called Field Notes in the Zero Click Marketing podcast. He teasingly asked if Claude came up with that name and I shook my head, “Sadly, that is all me. I guess we marketers aren’t always as clever as we think.” Anyway, now I’m nerding out in my hotel room watching clips of his show. Check it out!

And look, ABSOLUTELY NO ONE ASKED: But here was my keynote outfit this week. A vest and trousers from Cider — a fast fashion-y brand but you know what, their clothes generally fit me better than most other brands and the styles and silhouettes are so much more interesting that what you typically find at retailers. Although one sneaky thing I did was tailor my own vest by hand. I cut it at the shoulders, trimmed off 4cm, and sewed it back together. You’d never know… until I put my hair up and you see the Frankenstein-esque stitches. Appalling.

Haters will say I wish I were a fashion influencer. And you know what? The haters are right. But you don’t choose the B2B lyfe. It chooses you.

See ya next week, besties!

They can give you a baseline, help you avoid obvious mistakes, and keep you from totally eating shit. But best practices are usually just generalized averages. They’re what tends to work across many audiences, many contexts, and many companies. So when you apply them without personalization or a point of view, you’re not really making a strategic decision. You are, quite literally, making a statistically average one. It’s beige marketing, and beige is an even bigger problem now than it used to be.

Everybody has access to the same AI tools, the same LinkedIn takes, the same recycled “playbooks,” and the same tired list of things we are supposedly supposed to do. Post more often. Start a YouTube channel. Start a podcast. Send the pitch. Shoot your shot. Don’t get left behind, bro.

We’re not asking ourselves often enough: “What does this audience actually respond to?”

That’s where audience research comes in. And this is what we’re talking about on today’s episode of the Zero Click Marketing podcast.

Good marketers have always done some version of this

One reason I’m so passionate about audience research is that I realized, in hindsight, that it was basically how I had always done marketing. I just didn’t have a clean term for it yet.

Twelve years ago, I was one of the early B2B marketing hires at Fitbit. We had a very different challenge from the one a lot of B2B brands have today. We were already a known and beloved consumer brand. We didn’t need to increase awareness among B2B buyers. What we needed to do was build trust in that segment, and building trust had very little to do with ranking for generic keywords like “what is corporate wellness” and a lot more to do with proving why our offering was better for businesses. That meant case studies, proof, and thought leadership that showed we understood where wellness and benefits were headed.

One of the most useful inputs I got at the time came from an unlikely place: a PR agency pitching for our business. In their deck, they spent several slides showing us what our audience paid attention to. They told us our audience was reading magazines like Psychology Today and Entrepreneur, and that they cared about social psychology. Just these few insights helped answer questions like where we might advertise, what ideas should shape our next thought-leadership pieces, and how we should start talking to that audience in a way that felt fluent instead of generic.

This was over a decade ago and it really just shows that audience research is not new. Good marketers have been doing some version of it for forever, even when they didn’t call it that.

Audience research makes practices… actually best

When you do audience research, you stop asking, “What should I do?” and start asking, “What does this audience actually care about? What language do they use? What do they trust? What do they ignore? What would feel useful to them?” That shift changes almost every subsequent decision. It helps you get more specific.

If you market to marketers, best practices might tell you to write about SEO tactics. Fine. But audience research might tell you that what they really want is a more specific angle, like what click signals to Google Search may actually affect visibility. If you market a skincare product, best practices might tell you to publish the evergreen “7-step nighttime skincare routine” post. Audience research might tell you that your audience is already fluent in skincare basics and are looking for a more specific skin goal, like achieving dewy skin. Actually, I don’t have to hypothesize this. It’s true, per SparkToro data:

Audience research is the difference between “best practices” and your best practices.

What audience research actually looks like

The good news is that audience research does not have to be expensive or elaborate.

You can talk to five people in your audience. Seriously. Five. Ask them to complain about a problem so you get their real pain-point language. Ask about their most recent purchase and why they chose it. Ask where they spend time online, which accounts they follow, and what kinds of content they pay attention to. Keep it to 20 minutes. Your script already lays out that exact lightweight process, and your earlier blog post makes the same case that even five conversations can uncover better language and better clues.

You can also go lurk in the communities where they already hang out. Subreddits. Product reviews. G2. Sephora. Yelp. Slack groups if you have access. Read how people talk when they are delighted, confused, skeptical, disappointed, or ready to buy. That kind of language is gold, and both the script and your earlier post make that point clearly. (Or yes, you can use a tool like SparkToro. I mean… you should.)

Best practices can get you on the field

Best practices can keep you from making catastrophic mistakes. They can get you on the field. But they can’t tell you what will feel resonant to the right audience in the right context.

Audience research is what helps you actually win.

You can learn more about how to do audience research and and more on how to not be beige, in the latest episode of Zero Click Marketing:

🧁 Petits Fours: for Builders and Networkers

Four discrete bite-sized thoughts because this newsletter, after all, is called the “Menu.”

✨ Consultation and resources for early-stage builders: My friend Carrie Olsen made a free guide called “3 Steps to Building Something That’s Yours (Even if You’re Terrified of What Others Will Think).” It’s for people who have a real, buildable, needed, and practical thing, but have fallen into the habit of letting other people’s imagined opinions (they’ll hate it, they’ll think it’s dumb, I’ll look like a beginner) dictate whether they start or not. I love how it paints being responsible as a good thing, but also reframes the, “I just need to learn one more thing, get a little more clarity, wait for the right moment” narrative as something that can actually be irresponsible. Carrie’s guide is short, helpful, and free. And best of all, it comes with a free consultation. So if you’re a builder who’s in their early stages, it’s the perfect opportunity.

🥃 🍹 An SEO shot and a chaser: New research from Cyrus Shepard suggests Google is moving beyond rewarding “good content” and increasingly favoring what AI can’t easily replicate. And today, Rand Fishkin explained what that means for your business. A must-read and a must-watch. I should probably record a ZCM Field Notes episode for the pod. Like right now. 🫠

🎤 Keynoting at Experience Inbound: I’ll be in Wisconsin next week — Milwaukee and Green Bay, specifically — keynoting at Experience Inbound. If you’ll be there, come find me. I always love meeting internet friends in real life. :)

🤝 A little MicroConf afterglow, plus some networking advice: One thing I really appreciated about last week’s MicroConf conference was how easy it felt to have real conversations. Not fake-networking conversations. Actual human ones. It was a good group and people seemed to genuinely want to help and get to know each other. (And this was before my talk!) The next time you’re at a networking event, I challenge you to do what I did: walk up to a group of 2–3 people and just say, “Hi! Can I join you all?” It will feel awkward, but I promise it’ll go well.

Today’s Zero Click Marketing episode gives you a playbook for strengthening that public record.

I’ll tell you two stories about why this matters:

First: Wil Reynolds and Nick Haigler at Seer Interactive found one old negative review that kept showing up across a meaningful share of branded AI prompts about their company. It wasn’t even representative of Seer as a company. But because the feedback exists, publicly and online, LLMs found it. Then those LLMs repeated it. When Seer published a post with more specific, current information about their company, the AI outputs changed quickly, and better, more honest information about the company came up to the surface. The lesson: if you don’t help strengthen your public record, weaker claims may do the job for you.

The other story is much smaller (and probably more related to positioning), but I think it matters. Last month, while building the Zero Click Marketing website, I asked Claude Code to recommend a tool for my contact form. It suggested Formspree. I skimmed their product page, saw that it could do exactly what I needed, and signed up.

I became a customer without experiencing the brand through any kind of marketing funnel. I never even saw Formspree’s homepage — which, by the way, is geared towards developers, and had I seen that, I never would have converted in spite of it being the perfect solution for me:

Admittedly, these two stories are AI-specific. But the broader point I want to emphasize is that they are also stories about influence where the brands themselves weren’t the ones controlling the message. We’ve reached a point where customers are increasingly making decisions based on what other people, platforms, and systems say about your brand — not just what your brand says about itself. On one hand, that’s… kinda scary. But it doesn’t mean we just surrender to whatever platforms, reviewers, communities, and AI systems happen to surface. We still have a role to play in strengthening that public record.

So how do you play that role?

In this week’s episode, I walk through a practical framework for strengthening your public record on purpose. It starts with auditing the story the internet is already telling about you: your branded search results, the snippets and citations that appear first, the comparison pages, reviews, Reddit threads, and AI answers that shape perception before anyone ever reaches your site. From there, the job is to figure out which weak but visible claims are winning, which strong facts you can actually support, and how to publish more of the stuff only you can credibly say.

The playbook, in short:

1. audit the current public record,

2. treat third-party surfaces like part of your brand footprint,

3. identify the weak claims that are winning,

4. publish better facts that only you can publish,

5. structure them so they’re easy to retrieve and cite,

6. refresh the record continuously,

7. and measure whether the story the internet tells about you is getting better.

If you want the full framework — and how to actually put it to work — watch (or listen to) the 12-minute episode:

🧁 Petits Fours: Elephants, Robots, Chickens, Shoes, Oh my!

Four discrete bite-sized thoughts because this newsletter, after all, is called the "Menu.”

🐘 Jason Cohen and the elephant curve: I’m at MicroConf this week and got to meet Jason Cohen, the founder of WP Engine and someone I’ve admired for a long time. He gave a killer presentation which was partly about “the elephant curve,” which describes most successful marketing programs. The tactic performs really really well, eventually reaches a point of saturation, and then slowly declines. That’s a really good essay and well-worth the read. And by the way, he has the most-viewed MicroConf talk at over 500,000 views! A must-watch for those looking to built bootstrapped businesses.

🤖 On the AI vs. Human debate: My friend Melanie Deziel wrote this essay, and weeks later I’m still thinking about it. A key line: “It doesn’t matter if AI can actually compete with human creators if decision-makers believe it can.” Because she’s right. Plus… in this AI era, I worry about the increasingly-loud paranoid people who speculate that just about everything they now see is AI. It’s happened to me and to my peers where something we’ve written was accused of being AI-generated. And nobody can prove anything to confirm or dispel, so there won’t be any satisfaction. But if we’re supposed to treat everything we consume like it might be fake, without any real way to know for sure, what’s the point of consuming anything at all?

🐓 On that note of AI vs. Human, here’s a human whose work you should really care about: Lianna Patch. She has a consultancy called Punchline Copy where she offers custom services for copywriting. She can rewrite your website, advise on your email strategy, coach you on your writing, or pretty much whatever else you need better words for. She’s currently emceeing MicroConf and doing a killer job, so I just really wanted to shout out her work because she is fantastic and I would hire her if I could. (Why the chicken emoji here? Because chickens are a theme on her website and it’s cute.)

👠 Literally nobody asked, but let me tell you about my shoes: I have several speaking engagements this year, so one thing on my mind (and on my credit card balance) is conference outfits. I have a couple fast fashion-y tops, Ann Taylor trousers, and my real favorite: Ally shoes — specifically, their bold block heel. It’s a bit of a challenge to walk longer distances (but that’s only because we generally shouldn’t be walking a mile in heels), but for the most part, I can wear these shoes for 6-8 hours with no problems. They’re that comfortable. And I wouldn’t dare gatekeep. Admittedly, they are pricey, but they’re the only heels I ever wear.

Also, thanks for being here.

Since last week, 41 of you joined this newsletter. Welcome! You can expect all things zero-click marketing, yes, but you can also expect occasional weird detours — usually via my Petits Fours which are really just four random things I’m noodling on, found, or felt I needed to amplify because I think they’re genuinely interesting or helpful. I am painfully aware there are a million newsletters out there and I really want to make sure that this is a good use of your time. I’m always open to feedback and topic ideas, too!

See ya next week.