Sample Gap Audit · fictional company

What does a Gap Audit tell a leadership team?

This is a sample Gap Audit for a fictional company, Brightwater Systems, which sells medical device quality management software in Ireland. It follows the structure of a real Audit: a 10-minute executive summary, the evidence, and what to do. Every company and number is invented; the changes in AI marked “Sourced” are real.

Part 1 · Executive summary · a 10-minute read · Gap Audit · Brightwater Systems · Ireland · October 2026

Brightwater Systems is 41 points behind Veltrix when Irish buyers ask AI who to consider.

Winning in AI search is not a climb to 100%. It is a contest against the companies your buyers compare you with: being put forward to people who have never heard of you, and being described well to those who have.

Buyers who don't know you

19%

put Brightwater Systems on the list · Veltrix 60%

The people who sign off

4%

CFO and CEO · Veltrix 60%

Buyers who know the name

11 of 15

told about the right company

Winning new business

“When buyers who don't know us ask for options, are we on the list?”

Rarely. Brightwater Systems is put forward in 19% of answers; Veltrix in 60%. With the finance leaders who approve the budget, Brightwater Systems appears in about 4 answers in 100.

Protecting business in motion

“When buyers who already know us ask about us, what are they told?”

Mostly the right story. AI describes Brightwater Systems correctly in 11 of 15 answers, but confuses it with a water-treatment business in the rest, and picks Veltrix in 11 of 15 head-to-head comparisons.

Where Brightwater Systems is strong

  • With the Head of Quality and the IT and validation lead, Brightwater Systems appears in about 1 answer in 3.
  • AI credits it ahead of Veltrix for small quality teams.
  • Gemini and Perplexity read and quote its website directly.

Where Brightwater Systems is exposed

  • The CFO, CEO and procurement almost never hear its name.
  • On ChatGPT and Claude, Veltrix is named five to seven times as often.
  • Two AI-native newcomers are already level with or ahead of Brightwater Systems.

Why it is happening

None of this is a product problem, and none of it needs a rebrand. Three quarters of the Gap comes from two causes: AI choosing a competitor that answers the buyer's question more directly, and finding more independent proof for Veltrix. Both can be fixed with work Brightwater Systems already knows how to do.

What the Gap is costing

For every 100 Irish buyers who start with AI, about 60 hear about Veltrix and about 19 hear about Brightwater Systems.

A buyer who never sees your name in AI arrives at the sales conversation leaning towards someone else, if they arrive at all. Nothing in the CRM records it, so the cost has to be estimated. Using Brightwater Systems's own figures, agreed at the closing workshop:

Illustrative business-impact model (assumptions agreed with the client)
Average first-year deal value€85,000
Win rate once on the shortlist25%
Extra buyers who see Brightwater Systems for each 10 points of Gap closed, per 100 AI-led buyers10
Pipeline value of closing 10 points, per 100 AI-led buyers€212,500

A planning estimate, not a forecast: it assumes buyers who see Brightwater Systems convert at its normal rate. Its purpose is to size the decision, and each Gap Pulse tests it against the pipeline Brightwater Systems actually sees.

The board question

Closing half the Gap in 12 months would put Brightwater Systems in front of about 20 more of every 100 buyers who start with AI. At Brightwater Systems's own conversion, that is the commercial case for the plan in Part 3.

The ground moves every month

What changed in AI, and what it means for Brightwater Systems

AI assistants change where they look and what they trust without notice. Every Audit and Pulse records the changes that could affect the numbers, so a movement is never misread as success or failure.

  1. Sourced

    ChatGPT cut how often it cites Reddit

    Reddit's share of ChatGPT Search citations fell from 3.83% to 0.52% between late July and mid-August 2026, in two steps on 8 and 14 August, according to provisional tracking data reported by Relevant Audience on 20 August 2026.

    What it means for Brightwater Systems: Brightwater's community presence is small, so the direct effect is limited; Veltrix's forum mentions now count for less on ChatGPT, which narrows one of its advantages.

  2. Sourced

    Adding schema did not raise AI citations

    Ahrefs tracked 1,885 pages that added structured data against matched controls and found no meaningful uplift in citations on Google AI Mode or ChatGPT over 30 days (Ahrefs, 11 May 2026).

    What it means for Brightwater Systems: Brightwater's web agency has proposed a schema retainer. We recommend a one-off fix instead, re-checked each quarter.

  3. Illustrative

    A platform changed its model mid-measurement

    One platform moved to a newer model between runs 2 and 3. Answers became shorter and named fewer companies, lowering every vendor's figure on that platform.

    What it means for Brightwater Systems: Recorded on every answer and adjusted for, so the change is not mistaken for movement caused by Brightwater or its competitors.

  4. Illustrative

    Two AI-native newcomers entered the shortlist

    Qualia AI and Auditly, neither on Brightwater's competitor list, are already named in 23% and 14% of answers, mainly through comparison guides they publish themselves.

    What it means for Brightwater Systems: Added to the tracked set, so each Gap Pulse shows whether they are taking ground.

Sources: “Reddit's ChatGPT citation share fell 86 percent and nobody can explain the timing”, Antonio Fernandez, Relevant Audience, 20 August 2026. “We Tracked 1,885 Pages Adding Schema. AI Citations Barely Moved.”, Louise Linehan, Ahrefs, 11 May 2026.

What we recommend

Four moves, in order, and one thing not to spend money on.

  1. Settle who Brightwater Systems is, everywhere · first month
    One agreed description, used word for word on the website, LinkedIn and every directory, and the LinkedIn industry corrected. The quickest way to stop AI confusing Brightwater Systems with another business.
  2. Answer the finance buyer's questions · months 2–3
    Clear public pages on cost, return and payback, and on EU MDR readiness, so AI has something to say when the CFO asks.
  3. Be present where AI looks · months 1–9
    Reviews from the mid-sized device makers Brightwater Systems serves best, honest comparison pages, and a place in the comparison articles AI cites. This is where most of the Gap sits.
  4. Measure the same way every quarter · ongoing
    A Gap Pulse each quarter, so every change can be traced to the work and new competitors are seen early.

What not to spend money on

A retainer for structured data (“schema”). It is one-off housekeeping for the web partner, measured in days and re-checked each quarter. Independent research found adding it made no meaningful difference to how often AI cited pages.

What we ask of the board

  • Name one owner for the programme, with the CEO as sponsor.
  • Approve the first quarter. Moves 1 and 2 are internal time, not new spend.
  • Judge it on evidence: what AI actually tells buyers each quarter, not activity.

Part 2 · The evidence · Where you stand

The Gap and the peer group

Visibility on its own says nothing about whether you are winning. 19% sounds poor until you know the leader's figure. The Gap shows how far behind the company your buyers compare you with you start, and it is the one figure that tells the board whether the investment is working.

Share of 900 core answers naming each company · 47 companies named in total, top 10 shown · margin of error about ±3 points

  1. 1VeltrixDirect peer60%
  2. 2QuorraDirect peer49%
  3. 3MeditraceDirect peer41%
  4. 4Halden QMSEnterprise suite33%
  5. 5CirrionDirect peer27%
  6. 6Qualia AIAI-native newcomer23%
  7. 7Brightwater Systems youYou19%
  8. 8ArcwellEnterprise suite17%
  9. 9AuditlyAI-native newcomer14%
  10. 10NorvaleOther category11%

Who you are really competing with

58% of AI's recommendations went to direct peers, 21% to enterprise quality suites built for global manufacturers, 13% to AI-native newcomers and 8% to other categories. When AI frames Brightwater Systems beside an enterprise suite, buyers judge it on scale, where a specialist loses, rather than on speed to validate, where it wins.

Where you stand

Platform by platform: the Gap is widest on ChatGPT and Claude.

Different buyers use different assistants, and each chooses its sources differently. The pattern shows which kind of evidence is missing: Brightwater Systems's website is read by assistants that search the live web, but the reviews and comparisons that ChatGPT and Claude weigh mention Veltrix far more.

PlatformBrightwater SystemsVeltrixGapWhat the pattern says
ChatGPT12%71%59Widest Gap; weighs reviews and comparison articles
Claude9%66%57Rarely names Brightwater; common with enterprise buyers
Google AI Mode22%57%35Reads brightwater.ie, but comparison pages favour Veltrix
Gemini26%54%28Quotes Brightwater's own website directly
Perplexity27%52%25Narrowest Gap; cites recent articles and the website

Each platform is pinned to the model version most buyers get, recorded on every answer.

Where you stand

The buying committee: visible to the people who use the product, invisible to the people who sign off.

These six personas were defined with Brightwater Systems's team and weighted by how often each one decides or sponsors deals in the CRM, so the Gap is read where the revenue is.

PersonaDeal weightBrightwater SystemsVeltrixGap
Head of Quality and Regulatory
Functional owner
30%31%55%24
CFO
Economic buyer
20%3%58%55
CEO / Managing Director
Economic buyer
15%5%62%57
Operations Director
Functional owner
15%24%57%33
IT and Validation Lead
Technical evaluator
10%34%61%27
Procurement
Risk and procurement
10%7%71%64

Each persona rests on 150 answers, so its margin of error is about ±7 points.

Where you stand

Buying criteria: AI doesn't yet credit Brightwater Systems with most of its real strengths.

Before a buyer meets a salesperson, AI has already told them what to look for and who is good at it. For every question touching each criterion, this shows how often AI credited each company.

CriterionBrightwater SystemsVeltrixHead to head
ISO 13485 document control34%61%Both named in 9 of 30; Veltrix alone in 12
Validation effort (21 CFR Part 11, Annex 11)41%52%Brightwater's real strength, close to level
EU MDR readiness12%58%Brightwater rarely credited, despite its MDR module
CAPA and complaints workflow22%49%Credited mainly on Gemini and Perplexity
Total cost of ownership4%46%Almost never credited; no public pricing or ROI pages
Fit for small quality teams38%29%Brightwater ahead: its clearest advantage in AI's eyes

Why

Why answers are lost: three quarters of the Gap comes from two causes.

“Low visibility” is a symptom. Every answer that left Brightwater Systems out was traced to the point where it dropped out, because each cause has a different fix and a different owner.

  1. AI found you, but chose someone else

    40% of the Gap

    Comparison pages, category language and finance-buyer questions answered by competitors, not by Brightwater.

    Usual owner: Marketing lead

  2. AI knew you, but found less proof

    35% of the Gap

    Veltrix has four times the reviews and appears in 7 of the 10 comparison articles AI cites most; Brightwater appears in 2.

    Usual owner: Communications, sales and customer success

  3. AI confused you with another business

    15% of the Gap

    Brightwater is confused with a water-treatment company of a similar name, and its LinkedIn record lists the wrong industry.

    Usual owner: Marketing, with whoever manages listings

  4. AI never found you

    10% of the Gap

    Two product pages render their text with JavaScript, and one AI crawler is blocked at the firewall.

    Usual owner: Website or digital lead

Why

The sources AI trusts: 70% of citations come from websites neither you nor your competitors own.

AI does not invent its recommendations; it repeats the sources it trusts. 1,640 citations from 312 websites were read and classified.

Source typeOwnerShare of citationsOf the 10 most-cited of this type
Comparison and “best of” articlesIndependent29%2 of 10 name Brightwater Systems; 7 of 10 name Veltrix
Review platformsIndependent18%3 of 10 name Brightwater Systems; 9 of 10 name Veltrix
Competitors' own websitesCompetitor21%—
brightwater.ieOwn site9%—
Trade press and associationsIndependent12%1 of 10 name Brightwater Systems; 4 of 10 name Veltrix
Directories and databasesIndependent11%2 of 10 name Brightwater Systems; 5 of 10 name Veltrix

The commercial read

Brightwater Systems has a review profile on one platform with 11 reviews; Veltrix has 46 across three. The comparison articles that shape most answers name Veltrix in 7 of 10 cases and Brightwater Systems in 2. That is where most of the Gap sits.

Why

AMBER When buyers know your name: right company in 11 of 15 answers.

Referrals, event leads and renewals all check Brightwater Systems in AI before they call. These questions name Brightwater Systems, so they never count in the Gap; they show whether it is believed once it is known.

  • “What is Brightwater Systems?” right company11 of 15
  • Confused withA water-treatment company of a similar name
  • With a category hint, right company15 of 15
  • Head to head with Veltrix, Brightwater Systems recommended4 of 15
  • Tone when correctly describedPositive or neutral in 96%

What is missing: outcomes. AI almost never describes Brightwater Systems in the language senior buyers use (time to validate, audit findings avoided, cost per site). It describes Veltrix in those terms in a third of its mentions.

Your own assets · scanned October 2026

Can AI read your website, and does every channel tell the same story?

These are hygiene checks: a one-off job for the web partner, re-checked each quarter. None gets a company recommended on its own, but each failure is a cheap risk to remove.

CheckBrightwater SystemsVeltrixQuorra
AI crawlers allowed (robots.txt and firewall)Partly: one blocked at the firewallYesYes
Words visible without JavaScriptNo on 2 product pagesYesYes
Organisation markup with profile linksYes, 3 profiles linkedYesPartly
One consistent company descriptionNo: 3 versionsYesPartly
Pages answering finance-buyer questionsNone4 pages2 pages

Owned channels: 3 of 8 facts consistent

AI checks what a company says about itself in several places. Where those places disagree, AI hedges, and a hedged answer rarely makes a shortlist.

  • What the company doesPartly consistent
  • Who it is forInconsistent
  • Industry classification (LinkedIn)Inconsistent
  • Headquarters and marketsConsistent
  • Products listedPartly consistent
  • Founding yearConsistent
  • Customer proofPartly consistent
  • Leadership namesConsistent

Part 3 · What to do

The Action Register: what to fix, in order, by whom.

Every action is tied to the cause it fixes and its share of the Gap, so effort goes where the return is. Owners and timings are agreed with Brightwater Systems at the closing workshop, and every Gap Pulse reports each action's status against the measure it should move.

#What needs fixingCause · shareOwnerEffortDone whenMeasured byTiming
1Publish one agreed description of Brightwater, word for word on the website, LinkedIn and every directory; correct the LinkedIn industryConfused · 15%MarketingDays8 of 8 profiles match the website“What is Brightwater Systems?” answered correctlyMonth 1
2Fix the two JavaScript-rendered product pages and the firewall rule blocking one AI crawlerNever found · 10%Web partnerDaysAll pages readable without JavaScript; crawler allowedLive-web platforms cite product pagesMonth 1
3Publish pages on cost, return and payback, and an EU MDR readiness page, written for the CFO and the Head of QualityChose someone else · 20%Marketing with salesWeeksPages live and cited by AIVisibility with the CFO and CEOMonths 2–3
4Publish honest comparison pages: Brightwater against Veltrix, Quorra and enterprise suitesChose someone else · 20%MarketingWeeksThree comparison pages liveHead to head with VeltrixMonths 2–4
5Run a review programme with mid-sized device makers, the customers Brightwater serves bestLess proof · 20%Customer successMonths40 new reviews across two platformsShare of answers citing review sitesMonths 1–6
6Earn a place in the comparison and “best of” articles AI cites mostLess proof · 15%CommunicationsMonths5 of the top 10 sources name BrightwaterShare of answers citing those sourcesMonths 2–9

Sequencing and targets

Fastest proof first, then the compounding work.

  1. Month 1

    Settle the identity

    Agreed description, LinkedIn corrected, website fixes.

  2. Months 1–3

    Switch on the engine

    Finance-buyer and MDR pages; review programme starts.

  3. Months 4–6

    Corroborate

    Comparison pages; first placements in the articles AI cites.

  4. Months 7–12

    Compound

    Steady reviews; ChatGPT and Claude catch up.

What success looks like: targets agreed at the closing workshop, re-measured every quarter
MeasureTodayIn 6 monthsIn 12 months
The Gap to Veltrix41 pointsunder 30under 20
Visibility with buyers who don't know Brightwater19%30%40%
CFO and CEO who see Brightwater on the list4 in 1001 in 101 in 5
“What is Brightwater Systems?” answered correctly11 of 1514 of 1515 of 15
Head to head with Veltrix, Brightwater recommended4 of 156 of 158 of 15

Live-web platforms (Gemini, Perplexity, Google AI Mode) can show change within weeks; training-led platforms (ChatGPT, Claude) typically catch up over several months. Expect early proof in the first quarter and the larger shifts after that.

Appendix · Method and engagement

How this Audit was measured

Engagement: seven days, fixed fee. A persona and question session with Brightwater Systems's team, using the CRM to choose and weight the buying committee; three measurement runs; analysis and diagnosis; this report; a one-day closing workshop to agree owners, order and targets.

Questions that don't name Brightwater Systems (the Gap): 6 personas × 10 questions (4 shortlist, 3 evaluate, 3 decide), written from the category alone: never the company's name, a vendor, its website wording or a country.

Questions that do: 2 per persona plus 3 identity checks, reported separately and never counted in the Gap.

Platforms and market: ChatGPT, Gemini, Perplexity, Claude and Google AI Mode, each with its own web search on, searching from Ireland. Each question asked as a first-time buyer would ask it: no login, no history. Three runs, on days 1, 4 and 8: 900 core answers. Visibility is the share naming the company (±3 points); the Gap is the leading direct peer's visibility minus Brightwater Systems's.

Sample with fictional data throughout, except the changes in AI marked “Sourced”. In a real Audit, every figure opens the answers behind it in an evidence browser, downloadable for the project team.

Where would your Gap Audit start?

With a Gap Check report: send your website and the market you want measured, and we send it, at no charge and with no meeting.