Loss Recovery Growth Model: Answer Engine Optimization

Occam 13.01.2026 09:20 With the slow and steady evolution from keyword searching to resolution questions typed into Answer Engines, you are going to lose traffic somewhere in the rage of -18% to -64% during the course of the next calendar year. Today, our challenge is three-fold: A. How can you identify the size of your loss? B. What can you do to recover some losses? C. What actions can you take to take advantage of this shift and grow revenues? Every smart company is building a forecasting model to estimate these three life-critical dimensions. Let me make that important exercise a little easier for you. My model will help you estimate your losses, identify opportunities to recover, and share actions, and by how much each can drive growth. You can use the model to have strategic conversations with your CEO, CFO, and help your CMO create a clear, prioritized, list of actions . This blog post was originally published as edition #482 of my newsletter, TMAI Premium. Each week, the newsletter shares strategic frameworks and practical here's how to stay at the very bleeding edge of CFO-proof Marketing and Analytics. Sign up for TMAI Premium to accelerate your career trajectory. The Librarian to Grad Assistant Search Transformation. Past: When the User asked a question, traditional Google was a librarian. It did not know answers, it pointed us to a section of the library – four plus ten blue links – and said the answer is highly likely to be in these books, good luck. As a business, our job: Be the most attractive, relevant, authoritative book on the shelf, and/or pay to put our book on the special display by the front door of the library. Present: When the User asks a question, the new Answer Engine is the ruthlessly efficient grad research assistant. It goes to the library. It reads the 14 relevant books from a billion books, synthesizes the information, writes a perfect one-page summary directly answering the question. Most businesses lose the clicks, one or two get the click now – and even that we might not get as the Answer Engine absorbs ecommerce, and as AEs become Agentic eliminating the human step altogether. ChatGPT, DeepSeek, Claude are examples of Answer Engines. AI Mode in Google, now out in 40 countries, is also an example of an Answer Engine experience. ChatGPT Agent, Rufus from Amazon are examples of AE experiences that get close to Agentic search – humans need only express a wish; the agent does everything else. Learning, shopping, homework, will never be the same again. My recent AEO newsletter series, six editions , covered the changes underway, the actions you need to take on your digital experience, your organic and paid strategies, new measurement to embrace, and how you can prepare to live in an Agentic “Search” world. If you’ve not read that almost mini-book, please do. It is critical to understand the nuance and detail of this conversation. Here’s a picture I sketched for a recent keynote on AEO, outlining the transformation underway… And, the implications… The completed forecast model will cover items 1 through 5 in the picture above, which are raised from these three questions for every business: 1. How much traffic will we lose? 2. Given the librarian shift, how much of this loss can we recover? 3. Can I take advantage of this shift and grow new traffic? Let’s answer. The Business Losses Are Here . The change in user behavior above is driving a change in the user experience . Ex: Answers take up most of the real estate, with the answer often reducing the need to go to a downstream site. Ex: Paid ads might come, but for now there are either no ads or few ads . Both create losses. My expectation is that the losses will accelerate in 2026. Important note: Losses, Recover & Grow will have different answers for different companies. Ex: For many publishers , the loss in traffic is already large, and permanent. These entities will need a phoenix type rebirth. I am going to focus on normal businesses: Ecommerce-type entities, B2B & B2C where the earning of revenue is a short, medium, long-term objective. 1. Factors driving losses in SEO Traffic. Informational queries are at the highest risk. These are the what is, how to, and best of queries. They represent, what some people call, top of the funnel. AEs are designed to directly answer these. Commercial queries are close to the highest risk. These are the vs queries, comparisons. Lenovo ThinkPad Snapdragon vs. Asus Zenbook A14. These are very high value queries, AEs will not generate an on-the-fly custom comparison table and even tell you to buy the Zenbook . Non-brand High-Intent Transactional queries are at moderate risk. These are the user typing “buy men’s waterproof hiking shoes for an Alaska trip.” The current UX is that the AE will provide a handful of recommendations, with a summary of why. The user will click fewer times, and your organic category page will be ineffective. Branded & Navigational queries are at low risk. These are “Kate Spade Black Friday Sale.” Or “Coach Tabby in loved leather.” The AE will deliver the traffic to you, as that is the intent of the human. In future iterations, there might be insertion of things like “site links,” which might be a small risk. So, what’s the potential loss through the year 2026 of no AEO action by your company? Let us assume you currently get 5,000,000 visits from Organic Search, and the Revenue Per Visit is $2.5. Based on my research, conversations with the top LLMs, competitive intelligence tool builders, and my judgment, here’s the section of the model that proposes the potential losses due to SEO : From an anticipated Revenue of $12.5m, a potential loss of $4m is coming your way. Your to-do is to go to your Digital Analytics tool and take your current Organic Search traffic and compute the size of your Informational, Commercial, Non-Brand Transactional, and Branded & Navigational. Then, update column 2 . The rest of the cells will update and you’ll see your real losses. My assessment of Potential Loss in each row is conservative, and for “strong brand ecommerce” type business. If your business is different, please invest in research and update that column. With so much changing all at the same time, it is difficult to predict user behavior shifts with 100% accuracy. Consider the loss estimates as the best educated markers. They could happen much faster, could take a little bit of time, but if you stretch over the year 2026, you will broadly be in this range. Hence… Time. For. Action! 4. Factors driving losses in PPC Traffic. The threat to Paid Search is different, and just as scary. It is defined by two words: Transformation, and displacement. Remember, for now, none of the LLMs have advertising as a business model . That’s displacement. Google, with the largest Search business to protect, will surely have ads in AI Mode, but for now they will look different and seem to be below the fold. Bing is trying new ad formats in its version of AI Mode. This is Transformation. Translation: Fewer options for you to pay to get clicks. Let's use the same categorization as SEO to reflect on risks. Informational queries represent the same high risk of loss. Setting aside the no ads and below the fold ads issues, the entire purpose of Answer Engines is to answer. Hence, budget spent on informational questions will likely be completely wasted. The User is likely to ignore the ad, and certainly not scroll for ads below the fold. Commercial queries hold a giant point of friction. Our paid ads for, say, “best shoes for hiking,” will compete with the Answer Engine’s AI-analyzed answer for… best hiking shoes! Unless by the grace of God your brand was the AI’s assessment of the right answer, your text ad will lose the fight for attention. Non-Brand High-Intent Transactional queries are likely the bulk of your current ad spend. Ex: buy waterproof hiking shoes for a trip to Alaska. The risk here is not elimination, rather it is format shift. Ex: Text ad slots with breadcrumbs are being replaced with one AI-powered carousels. The four ad slots were clearly visible; in a carousel format that might not be the case. AEs are likely to have new approaches to Ad Quality Score , this might make it harder for your ad to show up. Other as yet unpredictable ad “innovations.” Branded & Navigational queries, as with SEO, are likely to have a lower risk of reduction. to measure success, identify new opportunities. Here's a helpful sketch, by our friend Gemini, of my Big 5 recommendations: The post Loss Recovery Growth Model: Answer Engine Optimization appeared first on Occam's Razor by Avinash Kaushik.

Číst celý článek >>