The Gap

What is the Gap, and how do we use it to measure AI visibility?

The Gap is how many points your company trails the competitor AI recommends most, across the questions your buyers ask. It is the measure at the heart of Genivista's investigation: not how visible you are, but how far you are from the company taking your place.

How is the Gap calculated?

The Gap is the leading competitor's visibility minus yours, in points. Visibility is the share of AI answers to your buyers' questions that name a company, recorded live on named AI platforms.

You name your top three competitors for that market; the leading competitor is whichever of them AI names most. AI is not asked to decide who your peers are, so a five-person specialist is never measured against a global group it does not compete with. Everyone else AI names is shown as it was named.

A worked example

Illustrative figures The numbers below are an example, not a client result.

A company's buyers ask AI a set of questions about its category. AI names the leading competitor in 72% of the answers and the company in 18%. The Gap is 72 minus 18: 54 points.

A Gap Check is a first reading from a single run, so it shows no ranges. In a Gap Audit and each Gap Pulse the full question set is run more than once, and small movements are never presented as progress.

Leading competitor72%Your company18%The Gap: 54 pointsShare of AI answers that name each company · Illustrative
IllustrativeThe Gap is the distance to the most-named of your top three competitors: here it names the leader in 72% of answers and your company in 18%, a Gap of 54 points.
Illustrative data behind the chart
CompanyShare of AI answers naming it
Leading competitor72%
Your company18%
The Gap54 points

What are the four possible explanations for a Gap?

We investigate four possible explanations for every lost answer. They are not mutually exclusive: a business can have a name collision, thin third-party evidence and an unclear website at once. Knowing which apply matters, because each needs a different fix and a different owner.

  1. AI never found you

    AI could not read your site, or never came across you in the sources it searches when a buyer asks.

    Usual owner: website or digital lead

  2. AI found you but chose someone else

    AI knew you existed, but a competitor answered the buyer's question more clearly, more specifically or more often.

    Usual owner: marketing lead

  3. AI confused you with another business

    AI mixed you up with a similarly named company, an out-of-date description or the wrong category.

    Usual owner: marketing, with whoever manages your listings and profiles

  4. AI knew you but found less proof

    AI found more independent evidence for the competitor: reviews, case studies, press coverage and listings it trusts.

    Usual owner: communications, with sales and customer success

Why not use a score out of 100?

Because a score out of 100 cannot tell you whether you are winning. Two companies can share the same score while one leads its market and the other trails it badly.

Winning is a contest against the companies your buyers compare you with. In a crowded category, a score of 40 may put you ahead. In a category with one dominant name, the same 40 may leave you 40 points behind.

The Gap removes that ambiguity. It is always measured against the leading competitor you name, so the number means the same thing in every market. It shows how far your share of AI answers trails the competitor named most often. It does not, on its own, show lost revenue or market share.

Company A · scores 40 out of 100You40Leading peer3010 points ahead: winningCompany B · scores 40 out of 100You40Leading peer8040 points behind: losingShare of AI answers naming each company · Illustrative
IllustrativeTwo companies with the same score of 40 out of 100: one leads its peers by 10 points and is winning, the other is 40 points behind. The score alone cannot tell them apart.
Illustrative data behind the chart
ScoreLeading peerThe Gap
Company A403010 points ahead
Company B408040 points behind

What earns a place on the AI shortlist?

Proof and presence: clear descriptions of what you do, evidence buyers trust, and the third-party sources AI draws on. A Gap Check shows which of the four possible explanations are causes is costing you most.

Platform by platform

Which AI assistants do your buyers use?

Different buyers use different assistants, and each chooses its sources differently. A Gap that is narrow on one platform and wide on another shows which kind of evidence is missing. The descriptions below are general patterns, not measurements of your buyers; a Gap Audit tests the assistants that matter to yours.

ChatGPT

Widely used by professionals

For many B2B categories, a likely place for a buyer to ask who to work with.

Measured in: Initial Review, Diagnostic

Gemini

Google's assistant

Searches the live web and often quotes company websites directly, so changes can sometimes show up within weeks.

Measured in: Initial Review, Diagnostic

Perplexity

The researcher's assistant

Shows its sources, which makes it a natural fit for research-led buyers such as analysts and consultants.

Measured in: Initial Review, Diagnostic

Claude

Used by many professionals and organisations

If you sell to enterprise buyers, it is worth testing how Claude describes you; a Gap Audit does.

Measured in: Diagnostic

Google AI Mode

AI answers inside Google Search

Reaches anyone who searches on Google, often before they see a single link or advert.

Measured in: Diagnostic

For the sceptic

There are hundreds of free AI-visibility graders. Why is this different?

Four reasons: it is measured against your peers, from your buyers' questions, with real answers you can read, by a firm with nothing to sell you but the finding.

Measured against peers, not out of 100

A score out of 100 cannot tell you whether you are winning. The Gap compares you with the most-named of your top three competitors, because that is who takes your place on the shortlist.

Your buyers' questions, not your keywords

Questions are written blind, without your name or wording, for each person on a buying committee and in your market's own terms. Questions built from your own website tend to flatter you; these cannot.

Real answers, recorded live

Answers are recorded live, as a first-time buyer would see them, and kept unedited. In a Gap Check or a Gap Audit, each finding opens the answers behind it, and Neil offers to walk you through them.

Nothing to sell but the finding

You choose the market and your competitors; the questions and scoring are set by the method and cannot be edited for a better result. Genivista sells no content, fixes or retainers.

Why it needs measuring every quarter

Does AI change where it looks for answers?

Yes, often and without notice. A position measured once is out of date within months.

ChatGPT stopped citing Reddit, without notice

Reddit's share of ChatGPT Search citations fell from 3.83% to 0.52% between late July and mid-August 2026, a drop of about 86%, in two steps on 8 and 14 August. The tracker describes the figure as provisional.

Source: “Reddit's ChatGPT citation share fell 86 percent and nobody can explain the timing”, Antonio Fernandez, Relevant Audience, , reporting Promptwatch data. Read the source

Questions buyers ask

The Gap: common questions

What is the Gap in AI visibility?

The Gap is how many points a company trails the competitor AI recommends most, across the questions its buyers ask. It is the leading competitor's visibility minus the company's own, measured from live AI answers.

How do you measure AI visibility against competitors?

Questions are written blind from the category, for each role on a buying committee, and put live to named AI platforms. Visibility is the share of answers that name a company, and the Gap is the distance between you and the most-named of your top three competitors for that market. AI is not asked to decide who your peers are; the answers decide everyone else.

How is the Gap different from a free AI visibility score?

A free score usually rates you out of 100, which cannot say whether you are winning. The Gap is measured against your leading peer, from questions you did not choose, with the real answers available to read.

How precise is the Gap?

A Gap Check is a first reading from a single run: it shows the direction of travel and where the Gap appears, without ranges or margin figures. A Gap Audit runs the full question set three times over eight days, so one unusual day does not set the result. Visibility is the share of its 900 core answers that name a company, and the Gap is the difference between you and the most-named of your named competitors, in points.

How is a Gap Audit priced?

As a fixed fee, agreed before any work begins, with no retainer and no implementation upsell. The Gap Check that usually comes before it is at no charge, and Neil offers a 30-minute Q&A to discuss it.

What is your Gap?

A Gap Check measures it against your leading competitor. Send your website and market; we send the report at no charge, and Neil offers to walk you through it.