AI Search and the Law of Shitty Click-Throughs

This is the LinkedIn edition of AI Native, the newsletter for marketers who want to win the AI Age. Subscribe on Substack to get cutting-edge strategies, AI Search insights, and top marketing trends in your inbox.

Welcome to AI Native — the newsletter for marketers who want to dominate AI Search, supercharge their creativity, and build AI-Native teams.

Wait, didn’t this newsletter used to be called Co-Creative? Yes! We rebranded it because we’re marketers and can’t go three months without rebranding something. It fills the empty space in our souls.

But there was also a practical reason: The focus of this newsletter is changing because marketing is changing.

Whenever I’ve talked to marketing leaders over the past few months, they’ve all said some version of the same thing: They’re under immense pressure to build savvy, AI-native marketing teams, but there’s no blueprint. AI Search is a black box, getting ChatGPT Enterprise past procurement is a living hell, and they have no idea how their team is using AI today.

If that sounds familiar, this newsletter is for you.

I’m Joe Lazer, fractional CMO at Pepper. I spent eight years building Contently, wrote a best-selling book called The Storytelling Edge, and currently write a popular Substack newsletter about content strategy and business storytelling. This newsletter is my excuse to nerd out on AI and marketing. Each week, Pepper Managing Editor Steph Walden and I will bring you the latest and greatest insights on how to win the AI Age — starting with a deep focus on AI Search.

Why AI Search Is a Generational Opportunity for CMOs

Why AI Search? (Or, take your pick of acronym: GEO, LLMO, AIO, EIEIO.)

Because if I’ve learned one thing as a CMO, it’s that the Law of Shitty Click-Through Rates reigns supreme.

The Law of Shitty Click-Throughs says that the effectiveness of every marketing channel or tactic declines over time. The longer you wait to adopt a new channel, the harder it becomes to break through.

The first banner ad had a click-through rate of 78%. Today, the average banner CTR hovers at 0.1% — and most of those clicks are accidental. The same thing has happened with email marketing, Facebook ads, LinkedIn ads, PR, webinars, e-books, TikTok… you name it.

All of these channels and tactics worked brilliantly for early movers, but have delivered diminishing returns over time. By the time a marketing best practice is a best practice, it doesn’t really work anymore.

But if you get in early? That’s where the magic happens.

If you got in early on content marketing and newsletters, you have a powerful organic engine.

If you got in early on TikTok, you have a huge advantage over your competitors.

If you got in early on content marketing and SEO, you’re printing money.

We only get a truly high-upside new channel once every “marketing generation,” which I’d define as every 4-7 years. A few past examples:

• 2007: Social (when Facebook first invited brands into the fold)
• 2011: Content marketing for B2B (when social started driving traffic and Google’s Panda algorithm started rewarding high-quality content)
• 2012: Mobile for B2C (when smartphone ownership hit 50% and Instagram took off)
• 2018: When TikTok first gained traction in the US

It’s been seven years, and we now have another generational opportunity: AI Search.

In May, Google rolled out AI Mode to all users in the US, providing a ChatGPT-like experience for search queries. It’s now on Google’s homepage. Sometime in the next year, they’re likely going to make it the default search experience for all users.

Since Google owns 90% of the search market, that will change marketing overnight.

Combine that with ChatGPT’s fast-climbing adoption, and marketers have an opportunity unlike anything I’ve seen in my 15 years in the industry. If you can master AI Search and make your brand a top recommendation of LLMs, you will have a ridiculously unfair advantage over your competition.

The only logical move is to go all-in and crack it first.

Luckily, Pepper has long been an SEO wizard, driving massive traffic gains for enterprise brands. We’ve been building GenAI products since before ChatGPT exploded onto the scene. That expertise has translated well, and our team has been hard at work in the lab decoding the AI Search algorithms. We’re literally discovering new tactics every day.

For a quick primer, grab our latest guide here, and stay tuned for more insights each week.

NEW AND NOTEWORTHY

The AI news cycle moves fast. We curate the chaos so you don’t have to. Here are the top stories on our radar this week:

• News publications are in an existential crisis after an analysis found that CTRs are dropping up to 80% for stories that appear below an AI overview.
• Google is beta testing Web Guide, a new AI Search feature that organizes results into topic-based clusters instead of that classic list of blue links.
• One in four users now prefer ChatGPT to Google for search, according to a new Adobe study — with young people leading the charge away from the search giant that raised them.
• Still, Google is performing better than expected against ChatGPT, with its new AI features driving up search volume and usage. First-mover advantage FTW.
• Over 100 million people are reportedly using Google's AI Mode, another indicator that the search titan will not go gentle into that good night. Its ad revenue also looks strong, with better-than-expected Q2 earnings suggesting reports of Google's advertising death are exaggerated.
• ChatGPT 5.0 release rumors are swirling, with murmurs that OpenAI's next flagship model could drop in August.
• Anthropic uncovered possible "subliminal learning" capabilities in large language models, suggesting AI might be learning new skills without explicit training.
• Google launched Opal, a vibe-coding tool that lets developers create mini applications by describing what they want rather than coding from scratch.

AI SEARCH DECODED

AI Summaries Are Cannibalizing Clicks

The Pew Research Center delivered what may be the most comprehensive look yet at how AI summaries are reshaping search behavior.

Their study of nearly 69,000 Google searches found that users click on website links just 8% of the time when AI summaries appear, compared to 15% for traditional search results. Even more telling, only 1% of users actually clicked on the source citations within those AI summaries, meaning the websites providing the information that powers these summaries see almost no traffic benefit. The data also revealed that users are 63% more likely to end their browsing session entirely when they encounter an AI summary.

Google, unsurprisingly, is having none of it. The company dismissed the research as using "a flawed methodology and skewed query-set that is not representative of Search traffic," while insisting they haven't seen the traffic drops the study purports.

Google's defense centers on the idea that AI features create "new opportunities for people to connect with websites,” though they're notably light on specifics about what those opportunities look like.

MEME OF THE WEEK

This is the LinkedIn edition of AI Native, the newsletter for marketers who want to win the AI Age. Subscribe on Substack to follow along with the latest AI Search insights and strategies.