Your keyword rankings are a vanity metric now. That is the uncomfortable claim, and it is worth sitting with before you argue, because the thing your rankings measure is drifting away from the thing that drives your business. You can hold the top three positions for every keyword you care about and still be completely absent from the AI answer that a growing share of your buyers see instead of the results page. The ranking is real. It is just increasingly disconnected from whether you get chosen. Prompt visibility is the metric that closed that gap, and if you are still reporting only on rankings, you are measuring a race that fewer people are running.
What is prompt visibility? It is how often and how prominently your brand shows up in the answers AI engines write for the prompts your buyers actually ask. When someone types “best project management tool for agencies” into ChatGPT and gets a paragraph naming four tools, prompt visibility is whether you are one of the four and where you sit among them. It is ranking, relocated from a list of links to the inside of a generated answer, and it behaves differently enough that the old playbook does not carry over cleanly.
Why rankings and prompt visibility diverge
A search ranking is a position in a list the engine retrieves. A prompt answer is a piece of text the engine composes. Those are different operations, and they draw on the brand’s presence in different ways. Retrieval asks which pages best match the query. Composition asks which brands the model has learned to associate with the topic strongly enough to name them when it writes about it. You can win the first and lose the second, because ranking a page and being named in prose are not the same achievement.
Consider what the engine does when it composes an answer. It is not just copying your top-ranked page. It is drawing on a picture of your category assembled from everything it read across the web, and it names the brands that picture makes salient. If your presence in that broader picture is thin, the answer skips you even while your page ranks, because the page rank does not force the model to mention you in prose it writes about the category. This is why brands with strong traditional SEO keep discovering they are invisible in AI answers. The two systems reward overlapping but distinct things, and prompt visibility rewards breadth of credible mention more than page-level optimization does.
The divergence is going to widen, not close, as more queries get answered by a generated response instead of a list. Every query that shifts from ten links to one answer is a query where your ranking matters less and your prompt visibility matters more. The brands treating prompt visibility as a curiosity while they optimize rankings are optimizing the shrinking surface. The ones building prompt visibility now are moving to where the attention is going. That is the case for treating this as a primary metric rather than a secondary one, even though it is harder to measure and slower to move.
The prompt ladder
Prompt visibility is not one number, and treating it as one hides where you actually stand. The framework I use is the prompt ladder, which sorts the prompts in your category by how specific they are and recognizes that visibility on each rung is a different competition. At the bottom sit the broad prompts, “best CRM,” “top PR agencies,” the ones with the most volume and the fiercest competition, where dozens of established brands fight to be named. At the top sit the narrow prompts, “best CRM for solo real estate agents in Texas,” which almost nobody has optimized for and where being named is far more winnable.
The mistake most brands make is aiming only at the bottom rung, the broad high-volume prompts, and getting crushed because those answers are dominated by incumbents with years of accumulated mention gravity. Climbing the ladder from the top down is the smarter move. Win the narrow, specific prompts first, where the field is thin and your relevance is high, then use the credibility and traffic from those wins to build toward the broader rungs. A brand that owns two hundred specific prompts has more real prompt visibility, and more qualified buyers reaching it, than a brand fighting hopelessly for three broad ones.
The prompt ladder also tells you where to look for wins you are already close to. Somewhere on the ladder are prompts where you are named in second or third position and could reach first with modest work, and prompts where you appear inconsistently and could lock in with more mention consistency. Those near-misses are the highest-return targets, and you only see them if you have mapped your prompt visibility rung by rung instead of collapsing it into a single average. Map the ladder, find the rungs where you are one push from winning, and push there first.
How to measure prompt visibility
You measure prompt visibility by asking the engines directly, because there is no rankings tool that shows it for you yet in the way a rank tracker shows positions. Build the real list of prompts your buyers use, the questions they actually type when they are trying to solve the problem you solve, not the keywords your SEO tool suggests. Then run those prompts across the major engines and record, for each one, whether your brand is named, in what position within the answer, and which competitors appear alongside you. That record is your prompt visibility baseline, and it is the only honest view of where you stand in AI answers.
Do it on a schedule, because the value is in the trend, not the snapshot. A single run tells you today’s state. Monthly runs tell you whether your work is moving the needle, which prompts are improving, which competitors are gaining, and where you are slipping. The brands that take this seriously turn it into a tracked metric with the same regularity they once gave to rank tracking, because a metric you check once is a curiosity and a metric you track over time is a management tool. The discipline of measuring monthly is what turns prompt visibility from an interesting idea into something you can actually improve on purpose.
Watch position and context, not just presence. Being named last in an answer, or named with a hedge, is worth less than being named first as the clear recommendation, and the raw yes-or-no of whether you appear hides that difference. Record where in the answer you land and how you are described, because moving from a grudging third mention to a confident first is real progress that a presence-only measure would miss. The texture of how the engine names you is where a lot of the improvement lives, and you only capture it if your measurement looks past the binary.
What actually moves prompt visibility
Consistent, credible mentions across the sources engines read are the engine of prompt visibility, because the answer is assembled from those sources. When credible third-party publications, industry sites, and respected references name your brand in connection with your category, they feed the model’s picture of who belongs in answers about that category. This is where earned coverage and machine visibility meet: the mentions you earn across trustworthy sources become the mentions the engine reaches for when it writes. Nothing you do on your own site substitutes for being named across the web, because the engine trusts the web’s consensus about you more than it trusts your own claims.
Consistency of identity turns scattered mentions into a coherent signal. If your brand is named differently across sources, tied to different categories, or described in conflicting ways, the engine cannot build a confident association, and prompt visibility suffers even when your raw mention count is high. Keeping your name, your category, and your description uniform everywhere lets each mention reinforce the last, so the model’s picture of you sharpens instead of blurring. Two hundred consistent mentions produce more prompt visibility than four hundred contradictory ones, because the engine can act on a clear signal and hesitates over a muddy one.
Answer-shaped content on your own site does the supporting work. Pages that clearly state what you do, who you serve, and how you compare give engines clean material to read and cite, and they reinforce the category association the third-party mentions establish. This is the part you fully control, and it matters most when it aligns with the external signal rather than fighting it. The brands that win prompt visibility run both tracks together: a growing body of consistent external mentions, and a site that tells the same clear story the mentions tell. Build that alignment, track your prompt ladder monthly, and the answers start naming you where it counts.