Kevin Indig's Recent LinkedIn Posts

If you think writing the "best answer" for a single keyword will get you cited by ChatGPT, the data says otherwise. After analyzing 1.2 million ChatGPT responses, the reality of how AI picks its sources is much more consolidated than traditional search. Here is what the data actually shows: 1. Across verticals, the top 30 domains capture 67% of all AI citations for a given topic. 2. The pages that get cited repeatedly aren’t highly specific, single-intent pages. They are broad, category-level guides that answer dozens of related questions at once. 3. 58% of cited URLs appear in only one prompt and then vanish. The top 5% of pages, however, answer 10+ unique prompts. The "one keyword, one page" model is... not good. To earn a seat at the table, you want to build topical hubs that address query clusters. Read the full breakdown in this week's Growth Memo.

AI SEO transformation fails at the alignment layer, not the tactics layer. Right now, teams are panic-testing short-term tactics rather than reorienting the whole ship for better long-term outcomes. The proof? In the recent Growth Memo reader survey, 84% of practitioners said they feel their current LLM visibility measurement approach is inaccurate. You can see this misalignment happening in real-time when: 1. Executives ask for increased traffic in a growing zero-click environment. 2. One stakeholder wants "rankings in ChatGPT," another wants brand mentions, and no one agrees on how to measure value. 3. Different teams (SEO, content, PR, product) run experiments in their silos because there is no explicit ownership. AI SEO is not just an algorithm update; it is a search product change and a user behavior movement. Because of this, our roles have to evolve. We have to elevate from being channel executors to organizational translators. If you can't explain your AI SEO mandate to leadership in one sentence, you aren't ready to execute. Read the full breakdown in this week's Growth Memo.

Growth Memo crossed 25,000 subscribers. This one feels different from 20k because I know what it took. The AI overview usability study. The 1.2 million search result analysis. The AI Mode study. The backlink research. None of that is cheap or fast. 600+ of you fund that research directly. That's what makes it possible. On to 50k. 🙏

Location within a document is only one variable in LLM retrieval. My analysis of 11,022 citations identified specific linguistic characteristics that distinguish cited text from skipped text. The data highlights three primary factors that increase citation likelihood. 1) Entity Density Standard English text typically contains 5-8% proper nouns. Text cited by ChatGPT averages an entity density of 20.6%. The model prioritizes sentences containing specific "anchors" (brands, tools, people) because they reduce perplexity. Generic advice is frequently skipped in favor of verifiable nouns. 2) Definitive Phrasing Cited content is nearly twice as likely (36.2% vs 20.2%) to use definitive verbs like "is defined as" or "refers to". Vector databases prioritize strong bridges between a subject and its definition. A direct "X is Y" structure allows the model to resolve a query in a single sentence. 3) Readability Scores Complexity correlates negatively with citation. "Winning" content averaged a Flesch-Kincaid score of 16 (college level), while skipped content averaged 19.1 (academic/PhD level). The algorithms favor simple subject-verb-object structures over complex sentence architectures. Writers seeking AI visibility should prioritize high information gain per sentence rather than narrative complexity. More details in this week's Growth Memo.

Why do so many teams spend quarters executing AI search tactics and get absolutely nowhere? Most AI SEO strategies are actually just tactic lists. We see teams chasing citations in ChatGPT without understanding if that solves an actual business problem. If your starting point is simply seeing that AI search is growing and deciding you need to rank in it, you are reacting instead of strategizing. A real strategy starts by identifying a specific revenue pressure. Are you dealing with brand visibility erosion where AI answers branded queries without attribution? Or are you facing a pipeline protection issue where qualified traffic is shifting to AI mode, but your brand is invisible? Once you define the exact business problem, you can concentrate your resources on the surfaces where your buyers actually research. I cover the framework for building an AI SEO strategy from scratch in this week's Growth Memo.

Review platforms gain share in ChatGPT citations as buyer intent increases. I looked at ~35,000 ChatGPT citations (US-only, Dec 2025) to understand the presence of review platforms along the software buyer journey. Review-platform citations: • Discovery: 7.4% • Evaluation: 13.2% (~1.8x of discovery) Overall, review platforms are cited 8.6% vs 4.0% for publishers. Inside the review category, G2 + Capterra + GetApp + Software Advice account for 84% of citations. If you sell software, your review presence is part of your bottom-funnel surface area now. Full analysis in the comments.

Why do so many teams spend quarters executing AI search tactics and get absolutely nowhere? Most AI SEO strategies are actually just tactic lists. We see teams chasing citations in ChatGPT without understanding if that solves an actual business problem. If your starting point is simply seeing that AI search is growing and deciding you need to rank in it, you are reacting instead of strategizing. A real strategy starts by identifying a specific revenue pressure. Are you dealing with brand visibility erosion where AI answers branded queries without attribution? Or are you facing a pipeline protection issue where qualified traffic is shifting to AI mode, but your brand is invisible? Once you define the exact business problem, you can concentrate your resources on the surfaces where your buyers actually research. I cover the framework for building an AI SEO strategy from scratch in this week's Growth Memo.

If your e-commerce SEO team is still measuring success by classic blue-link rankings, you might be tracking a vanity metric. We all know SERP features have been eating organic clicks for years. But there is a much larger shift happening right now with Product Grids that is fundamentally changing who actually gets the click. I recently analyzed over 4,000 keywords and nearly 40,000 product grids to understand the real impact of Google's transformation into a shopping marketplace. The data revealed a massive, industry-wide blind spot. The brands completely dominating traditional top-3 organic rankings are often practically invisible where buyers actually look. Conversely, there are brands bypassing traditional SEO entirely and still capturing the lion's share of the visual SERP. How is that possible? Because organic rank and product grid presence run on entirely independent systems, and if you appear well in one, you don't necessarily appear well in the other. In tomorrow's edition of the Growth Memo, I break down the exact mechanics behind this divide, why your primary SEO metrics might be misleading you, and the entirely different playbook required to win Google's real shopping surface.

E-commerce SEO has split into two games, and most teams are only playing one of them. Organic rank and grid presence are independent systems. The laptop query space is a great example of this new divide. Discount Computer Depot is the traditional SEO powerhouse, holding over 87% of standard organic rankings in the top 3 positions. Yet, their product grid presence is virtually non-existent, at just 2.4%. Back Market, on the other hand, has a mere 1.7% of top 3 rankings but owns 59% of the visual product grids. They saw massive growth in grid placements, jumping from 745 in May 2025 to 1,960 in February 2026. Back Market didn't win by playing the old game better. They won by recognizing the game had changed. I break down the exact mechanics behind this gap and how the two systems operate independently in the newest edition of the Growth Memo.

As Google's AI Overviews begin to impact overall click volume, advertisers are finding themselves paying a premium for the clicks that remain. This shrinking inventory naturally concentrates bidding, which steadily drives up the cost per click across the board. While this happens, user discovery has been diversifying to platforms where the usual paid search brand tax doesn't apply. Search behavior expanded to dozens of platforms, with alternative engines and communities like Reddit, Amazon, and YouTube taking a much larger share of the initial discovery phase. If a company continues to allocate the vast majority of its paid budget solely to Google (brand), it risks over-optimizing for a single channel. Defending branded keywords on one search engine does little to build visibility on emerging AI surfaces or alternative platforms. Ultimately, investing in brand visibility and trust earlier in the user journey makes sound financial sense. Building that familiarity before the click often produces much better economics than paying increasingly high acquisition rates at the very end of the funnel. I explore the financial case for shifting focus toward AI SEO and alternative search surfaces in this week's Growth Memo.

Is copying your competitors' AI search tactics the fastest way to waste a quarter? Sustainable AI visibility requires a foundation built on advantages only your brand has. When you build an approach, it needs to rely on your unique organizational capabilities. Do you have proprietary data that no competitor can access? If so, your approach is product-led content. Do you have recognized experts on staff? Your play is authority multiplication by getting those experts published on third-party sites. Do you have an active customer community? In that case, you amplify community signals because AI models weigh user-generated evidence heavily. If your action list does not directly serve one of your distinct advantages, you are just working through a generic checklist. I detail how to match your brand's unique assets to a focused execution plan in the latest Growth Memo.

See you at Sydney SEO Conference 2026 on March 20. I’m talking about visibility layer, trust stack, and how to win when AI is the interface. Bonus: all profits go to Chris O'Brien Lifehouse. Only ~50 tickets left. 🎟️ Event details: https://lnkd.in/d8Jw8kar

AI visibility is a funnel. Diagnose the leak.

Niche first, scale later

When marketing teams report blended ROAS without separating branded and non-branded search, it often masks the true cost of acquisition. High returns on brand campaigns usually reflect demand that was already created elsewhere - whether through word of mouth, social media, or other upstream efforts. A user hears about your company, searches your exact name, and converts. In the dashboard, the ad platform gets the attribution credit, which naturally justifies spending even more on those same ads. But this is demand capture, not true acquisition. Once you start splitting out those branded terms, the underlying economics become much clearer, revealing what you are actually paying to bring in net-new customers. In the latest edition of the Growth Memo, I share a concrete framework for separating real acquisition from demand capture, including how to test if your branded spend is actually incremental.

A new paper shows how easily AI shopping recommendations can be manipulated.

I mapped exactly where on a page ChatGPT extracts its citations across 7 different verticals. The result is what I call the "Ski Ramp." Across the board, citations peak at the 10-20% mark of a page and drop into a "dead zone" by the bottom 10%. But the data shows the steepness of that ramp is highly vertical-specific: 1. Finance has the steepest drop-off of any vertical we analyzed (43.7% concentration at the very top). If you are writing for YMYL transactional queries, the LLM wants the definitive answer immediately. If it's not at the top, it doesn't exist. 2. Healthcare has the flattest ramp (32.4%). The LLM is demonstrably more willing to extract citations from further down the page, likely looking for deeper context, symptoms, and structural depth. 3. Regardless of the vertical, the bottom 10% of your page earns roughly 3-4x fewer citations than the top 20%. The takeaway: Put your most citable claims, data points, and definitive statements in the first 30% of your page. But tune your page structure to the specific extraction habits of your vertical. Dive into the full analysis of how ChatGPT reads structure, including the breakdown for Travel, SaaS, and Crypto, in the latest Growth Memo!

Most e-commerce SEO teams still measure success against classic search results. But product grids aren't just another SERP feature to track in your dashboard. They represent Google's final transformation from search engine to shopping marketplace. I used Audience Key to analyze 4,000+ keywords and almost 40,000 product grids over 9 months. The data reveals a massive shift: 1. Product grid placements grew 82% in 9 months. 2. 96% of all SERPs in this dataset show product grids. 3. AWR data shows product grids cut the CTR on organic results in half. Product grids push traditional blue links so far down the page that they become secondary navigation options, not primary discovery mechanisms. I dive deep into this transformation and what it means for your e-commerce strategy in the latest edition of the Growth Memo.

Last year, ad costs rose 30% while conversion rates fell 5.1%. The economics of performance marketing are deteriorating: Paid search bounces at 59%, meaning more than half of every paid search dollar produces a visitor who leaves without seeing a second page. And yet, many dashboards still show paid search as the top-performing channel. This happens because branded search is doing the heavy lifting: In B2B accounts, branded campaigns returned 1,299% ROAS versus 68% for non-branded. Branded search is demand capture, not acquisition. Read this week's Growth Memo to see the full breakdown of how branded search inflates the metrics teams use to justify ad budgets.

When marketing teams report blended ROAS without separating branded and non-branded search, it often masks the true cost of acquisition. High returns on brand campaigns usually reflect demand that was already created elsewhere - whether through word of mouth, social media, or other upstream efforts. A user hears about your company, searches your exact name, and converts. In the dashboard, the ad platform gets the attribution credit, which naturally justifies spending even more on those same ads. But this is demand capture, not true acquisition. Once you start splitting out those branded terms, the underlying economics become much clearer, revealing what you are actually paying to bring in net-new customers. In the latest edition of the Growth Memo, I share a concrete framework for separating real acquisition from demand capture, including how to test if your branded spend is actually incremental.

We know how AI reads a page. But why do certain pages get cited repeatedly while others vanish? I just audited nearly 98,000 AI citation rows across 7 verticals to map the structural DNA of a successful source. In tomorrow's Growth Memo, I'm breaking down exactly what the engines evaluate. You'll learn how AI assesses: 1. Page Architecture: The true impact of word count, list structures, and heading depth. 2. Extraction Zones: The specific coordinates on a page where AI hunts for data and what it ignores. 3.Linguistic Signals: How your tone and phrasing influence LLM trust. 4. Industry Nuances: Why universal "best practices" don't exist in AEO, and which tactics will actively backfire depending on your vertical. Full breakdown coming to Growth Memo tomorrow!

Track AI visibility by scenarios that matter to your customers.

Want an 8x success multiplier on your AI SEO strategy? It doesn’t come from a new tool. It comes from how you manage the transition. Too many teams treat AI search as a binary switch: either you have it completely figured out, or you’re failing. But the reality is that AI-SEO adoption isn't just a tactical shift. It’s an educational one. According to Prosci’s benchmarking study, projects with excellent change management are six to eight times more likely to succeed. But getting there doesn't require perfect, company-wide alignment on Day 1. The most successful growth leaders right now are acting as change agents, guiding their organizations through the uncertainty using three principles: 1. Honesty over confidence: They openly admit what we don't know yet (like the exact pipeline value of a ChatGPT mention), while acting on what we do know (that your brand missing from these answers is a risk). 2. Progress over perfection: They don't try to overhaul the entire marketing machine at once. They build momentum in phases: Pilot → Expand → Embed. 3. Translation over broadcasting: They stop reporting on outdated tactical metrics and start translating visibility into the language executives actually care about (Influence → Brand Recall → Pipeline). You don’t need a perfectly aligned organization to start adapting to AI search. You just need to stop mandating tactics and start guiding the transition. In this week's Growth Memo, I break down exactly how to step into this "change agent" role for your team.

How do you get leadership buy-in for an AI search initiative when the metrics they expect systematically decline? Traffic forecasts are tough in the current AI search environment. - When you pitch an AI SEO strategy to the executive team, you have to use scenario planning instead of promising arbitrary click volumes. - The conversation should frame capacity allocation against expected citation increases and how those influence assisted conversions. - You present conservative, moderate, and aggressive scenarios to show exactly what resources each requires and what outcomes they produce. This creates clear stage gates and makes the investment reversible. Executives are far more likely to approve structured experiments with defined decision points than open commitments based on unreliable models. The breakdown for presenting this data to leadership is available in the new Growth Memo.

How to show up in AI search

Traffic can flatten while signups grow. That gap is where AI search is headed.

If you want to win Google's visual shopping surface, the rules of traditional SEO no longer apply. Here is what separates product grid winners from traditional SEO winners: 1/ Feed quality over content quality. Your Merchant Center feed quality matters more than your product descriptions. Clean, complete, structured data beats beautifully written prose. 2/ Visual assets over backlinks. A single high-quality product image on a white background can generate more clicks than a dozen referring domains. Backlinks are invisible in product grids. 3/ Price competitiveness over domain authority. When users can compare prices at a glance, your 15-year-old domain and DR 70 profile mean nothing. 4/ Merchant Center optimization over on-page SEO. Product title templates, GTIN accuracy, and feed error rates are the new meta descriptions and header tags. For the full playbook on how to adapt your e-commerce strategy to win in product grids, check out the premium section of the latest Growth Memo.