Picture a chef who has spent years perfecting a recipe. Every dish leaves the kitchen the same way, and he can taste it, adjust it and watch the reaction at the table. Then the restaurant switches to a buffet. A hundred guests build a hundred different plates, and he no longer controls what lands in front of each one.
That is the analogy SEO strategist Gert Mellak, founder of SEO Leverage and a Counsel Cast collaborator, used in a recent newsletter to describe what AI search has done to content. For law firms, it fits uncomfortably well. A traditional search engine returned a results page of ten blue links that most users saw in much the same order. Today, AI powered search engines such as Google’s AI Mode and AI Overviews, ChatGPT, Perplexity and Microsoft Copilot assemble AI generated answers for each person, shaped by search history, location, device and context. That is real time personalization, and because AI models are probabilistic, the same question asked twice can produce two different responses.
The consequence, as Mellak puts it: “You can rank #1 and get cited in zero AI responses.” Rankings, impressions and click through rates still tell you something, but they are now only half the picture. A law firm can hold the top spot on Google search and still be invisible in the AI generated summaries that a growing share of users read instead of clicking.
In this, we post looks at how AI search works under the hood, why the old scoreboard is breaking, and which inputs a law firm can actually control. We draw on Mellak’s Search Everywhere Optimization pyramid, his thinking on input metrics, and his conversation with Karin Conroy on Counsel Cast, “Search Everywhere Optimization for Law Firms: Why Your Content Strategy Is Upside Down (And How to Fix It)”.
How AI Search Actually Works
To decide what to publish, it helps to understand what happens after someone types a question into a search bar or speaks it through voice search.
✅ Traditional Search Engines
Traditional search was largely about matching keywords: the engine looked for pages containing the words in the query and ranked them using links and other signals.
✅ AI Search
AI search adds several layers on top. Natural language processing lets the system interpret conversational queries the way people actually phrase them (e.g. “what happens if my business partner wants out of our LLC?”) rather than as strings of just keywords.
✅ Semantic Search
Semantic search then looks for meaning instead of exact wording, so a page that never repeats the user’s phrasing can still be a relevant result if it matches the underlying user intent.
✅ Vector Search
Behind that sits vector search where content is converted into numerical representations, stored in a vector database, and compared with the meaning of the query. Large language models trained on massive datasets supply the semantic understanding, including the ability to handle queries in multiple languages.
✅ Retrieval Augmented Generation
The final ingredient is retrieval augmented generation, or RAG. AWS describes it as “the process of optimizing the output of a large language model so it references an authoritative knowledge base outside of its training data before generating a response.” In practice, the system pulls up to date information from the web, gathers passages from multiple sources, and uses generative AI to write direct answers, often with direct links and suggested follow up questions attached.
Generative AI does the writing. Retrieval decides which sources it has to work with, and that is where your firm’s content either shows up or does not.
From Public Search to Private Data: The Same Engine Everywhere
The same architecture now runs inside firms. Microsoft describes Azure AI Search as a service that connects your data to AI so that AI agents and large language models can produce grounded answers, and it integrates with Azure OpenAI. For a law firm building tools over its own data, such as a knowledge base of precedents and other enterprise data, features such as document-level access control matter, because sensitive data must stay in the right hands. Public tools follow the same retrieve-then-generate logic, just over the open web.
There is no shortage of lists ranking the best AI search engines. For a law firm, the more useful observation is what they share. Whether it is Google’s AI Mode, ChatGPT, Perplexity or Microsoft Copilot, the key features are the same: retrieve from many places, synthesize, and answer directly.
The Buffet Has Many Tables: Search Everywhere Optimization
The evidence on what this does to traffic is starting to arrive. A randomized field experiment of 1,065 U.S. users, run by Saharsh Agarwal of the Indian School of Business and Ananya Sen of Carnegie Mellon, found that AI Overviews reduced outbound organic clicks by 38 percent on the queries where they appeared, and that zero-click searches rose from 54 percent to 72 percent. The paper is an unreviewed draft and focused on desktop Chrome users, so treat the figures as directional, but the direction is hard to miss. Referral traffic from the results page is being squeezed while more people turn to AI for legal questions. A UK survey of more than 3,200 people by the charity Justice and the Administrative Fairness Lab found that 26 percent of 18 to 24 year olds had used AI when facing a legal problem.
Where AI Looks: The Directories Behind the Answers
Where AI looks matters too. A 2026 report from 5W Public Relations and Haute Lawyer Network found that a handful of ranking directories, including Chambers, Legal 500, Super Lawyers, Best Lawyers, Martindale, Avvo and Justia, dominate the sources AI platforms cite for legal queries. In other words, AI visibility increasingly depends on being present in the places the AI trusts, not only on your own site.
Why Your Blog Sits at the Top of the Pyramid, Not the Base
This is the problem Mellak’s Search Everywhere Optimization pyramid is designed to solve. Its five layers are built from the base up:
- Audience Platform Research: find out where your prospects research, compare and decide.
- Alert System: monitor conversations about competitors, industry challenges and key topics so you can join them in time.
- Industry Publications: earn third-party credibility through bylines in established trade publications.
- Distribution: amplify everything on the right platforms, to the right people, at the right time.
- Your Own Publications: create original content for the platforms identified in layer one.
The insight that matters most is where your blog sits.
In Mellak’s words, “your own publications sit at the top of the pyramid, not at the base.” He adds that “each layer has a specific role. Skip one, and the layers above it won’t hold.” On his Counsel Cast episode, he put it even more directly: “If your law firm has been publishing blog posts consistently but not seeing leads, the problem isn’t your content. It’s the sequence.” Clients now research across platforms such as Reddit, LinkedIn and YouTube (which Mellak calls the second largest search engine after Google) and AI tools long before they reach a firm’s website. The job is to connect content across those places so that the firm is already familiar when the decision is made.

Focus On What You Can Control: Inputs, Not Outputs
If a buffet plate cannot be read from the kitchen, what can a firm actually manage? Mellak’s answer is the inputs. In his input metrics article, he writes that “the only metrics worth obsessing over are the ones you can change,” and that “you cannot control whether a piece of content ranks on page one. You can control whether it gets written, edited, published, and distributed properly.”
His newsletter lists the inputs AI systems weigh:
- Content that is the best answer to one specific question
- A technically clean site
- Brand mentions across the web
- Visible E-E-A-T, and
- Consistent publishing.
None of this is new. As he puts it, AI has simply made it impossible to hide behind output-side vanity metrics.
5 Habits That Make Your Firm Easy for AI to Find and Trust
For a law firm, these inputs translate into five habits:
- Answer first. Our guide to writing law firm content that ranks in AI search recommends opening with a direct answer in plain language and targeting real client questions rather than just keywords. AI generated summaries reward pages that give quick answers to complex questions, and clients ask those questions the way they would ask a colleague.
- Make the site easy for machines to read. A fast, crawlable site with clean structured data helps search engines and AI systems interpret your practice pages, attorney bios and FAQs. Think of your practice pages as a product catalog: each service clearly described, consistently labeled and easy to retrieve.
- Earn confirmation from places you do not control. Our piece on how law firms build authority AI search can see argues that reviews, media mentions, directory listings and guest articles count for more than a firm’s own claims. That lines up with the directory findings above.
- Make expertise visible. The E-E-A-T framework still applies: author credentials, real client experience and original expert insights. Keep your state bar’s advertising rules in mind, and see our ethics and compliance guide to using AI in legal marketing.
- Publish consistently, in more than one format. Because AI tools retrieve up to date information and deliver real time answers, fresh, regular content signals relevance. Formats such as podcasts and video travel across platforms, a theme Counsel Cast explores in Why Podcasts Win in the AI Era with Rachel Allen.
This is also why an input dashboard beats a ranking report. Instead of watching click through rates alone, track what goes in: pieces published, distribution steps completed, third-party mentions earned and questions answered. Or, in Mellak’s words, “Track what goes in. Optimize what goes in. The outputs will follow.”
A Practical Audit To Follow.
Mellak’s suggested first step is deliberately small: audit your inputs, not your rankings. Pick one practice area your firm owns and work through it.
- Test the search experience your clients have. Ask Google’s AI Mode, ChatGPT, Perplexity and Microsoft Copilot the questions a prospective client would ask about that practice area, then ask a few follow up questions. Note which sources appear, whether your firm is named, and whether the direct links point to you, a directory or a competitor. Keep sensitive data and client details out of these prompts.
- Check whether your content is the clearest answer available. Does the opening of each page answer the query in plain language? Is the advice specific to one question, or broadly useful and therefore forgettable?
- Look for mentions outside your own site. Reviews, directory profiles, podcasts, trade publications, LinkedIn and guest articles all count. Where the pyramid’s lower layers are empty, start there, and our own article on authority AI can see is a good place to begin.
- Review the technical basics. Confirm that pages can be crawled and indexed, load quickly and carry accurate structured data. If your firm serves clients in multiple languages, make sure the content exists in those languages too.
- Set up an input dashboard. Track what you can control each month, and let rankings and referral traffic become supporting evidence rather than the scoreboard.
SEO Still Matters, Just in the Right Order
Of course, this does not replace SEO. It puts SEO inside a bigger system. As we argue in Why “More SEO” Isn’t the Answer to Your Law Firm’s Marketing Problem, search helps clients find a firm, but positioning, client-focused content and proof decide whether they hire it. And in our article, How AI Is Changing SEO for Law Firms, we explain why verifiable expertise now beats keyword tricks. Mellak’s pyramid adds one more lesson about effort: “it takes much more effort to distribute something properly than to create it in the first place.” A firm that treats distribution as part of the job, not an afterthought, is the one that AI agents and human users keep meeting in different places.
Build Inputs Worth Retrieving
Artificial intelligence has not made search disappear. Instead, it has moved the audience. Traditional search engines still matter, but their search results now share the stage with conversational interfaces that blend web search, advanced algorithms and generative AI into a single response. Google keeps adding AI features to its results, and each one decides what to surface from sources it can read and trust. Retailers have long structured their product catalogs so that machines can parse them. Law firms need to make relevant results about their own services just as easy to retrieve.
To return to our ‘buffet analogy, the firms that get found will not be the ones trying to read the plates from the kitchen. They will be the ones doing the input work consistently: a clear point of view, expert insights shared where clients actually research, and third-party confirmation that connects the dots.
How Conroy Creative Counsel Can Help
At Conroy Creative Counsel, we are proud to be an award-winning law firm marketing agency that helps law firms build the inputs AI search rewards. Our team combines strategy, branding, law firm website design and SEO and PPC so that your firm is easy to find, easy to understand and easy to trust.
If you want to become the source AI tools recommend, The Authority Tour is built to help you position your expertise across the platforms that matter, in a way both clients and AI can see. And if you would like to hear more from experts such as Gert Mellak, Counsel Cast, our podcast for law firm leaders, covers strategy, positioning and AI search in every episode.
Connect With Us Today
If you are looking to strengthen your online presence and position your firm as a trusted authority in an evolving digital landscape, contact us today to discuss whether we are the right strategic legal marketing partner to help elevate your visibility, authority and long-term influence.




