# Why AI Assistants Name the Market Leader and Rarely You

> A marketing law, double jeopardy, says smaller brands lose twice: fewer buyers, slightly less loyal. In two months of ChatGPT and Google AI Mode answers, the second loss was far steeper, and the two multiplied into a sales channel that barely reaches the companies it rarely names.

By Vlad Shvets, Partner @ Qvery · Published 2026-03-17 · Updated 2026-11-09 · 10 min read
Canonical: https://qvery.ai/blog/double-jeopardy-in-ai-search/

In their 1990 paper *Double Jeopardy Revisited*, Ehrenberg, Goodhardt, and Barwise describe the pattern: a small brand has fewer buyers, who also buy it less often. The Ehrenberg-Bass Institute, which traces the law to the 1960s and finds it across consumer and business-to-business markets, puts the result plainly: a firm with more market share has many more buyers who are slightly more loyal.

The law's pattern holds, and more steeply than in buying. A company named on fewer of its market's buyer questions is also named less often each time one of them is asked again.

> At Qvery, we help challengers to better-known names spend less staff time turning buyer questions into pages, with custom AI agents that find where a rival is named and draft changes.

The two penalties multiply, limiting how often buyers see your company. These findings cover mainly business-software markets. Smaller here means named less, not fewer employees; a name in an answer is not a sale.

- **62.21%** of all the answers to a market's buyer questions name its most-named company (median market, ChatGPT)
- **0.0435%** name the typical company the answers mention at all, about one answer in 2,299
- **40.42%** of the companies the answers name in a market appear in exactly one answer in two months (ChatGPT)

Source: Qvery's GEO Agent: answers to tracked buyer questions, ChatGPT and Google AI Mode, August to September 2026

## A Handful of Companies Take Most of the Mentions

Start with the first penalty. In the typical market, the five most-named companies took **47.78% of every company mention** on ChatGPT and 36.48% on Google AI Mode. The single most-named company took 13.25% and 12.81%.

These answers come from the GEO Agent's daily records: the same buyer questions asked on ChatGPT and Google AI Mode day after day in August and September 2026, mostly in business-software markets. Typical means the median.

The other end of the list is where the less-named company lives. **Four in ten companies** the answers named in a market appeared in **exactly one answer** in two months: 40.42% on ChatGPT, 43.75% on Google AI Mode. More than half appeared twice or less.

> Four in ten companies named in a single answer over two months is a finding only because the counting holds still: the GEO Agent's daily records ask the same buyer question the same way every day, and ChatGPT and Google AI Mode are counted apart. Before you read your own row, roll every product name up to the company that sells it, or one company splits into three small ones and the long tail looks longer than it is.
>
> Piotr Babel, Partner @ Qvery

## A Smaller Company Named Once Is Usually Not Named Again

For the second penalty, sort companies by penetration (the share of a market's buyer questions on which an answer names the company at least once), then compare their repeat rates (on those questions, the share of each question's answers that name the company, averaged).

**The Fewer Questions Name a Company, the Less Often Each One Names It Again** (% of answers naming the company)

Companies named on under 5% of their market's questions were named in 2.17% of the answers to those questions; companies named on half or more, in 38.06% on ChatGPT and 29.07% on Google AI Mode.

| Label | ChatGPT | Google AI Mode |
|---|---:|---:|
| Under 5% | 2.17% | 2.17% |
| 5% to 10% | 5.67% | 4.71% |
| 10% to 25% | 8.77% | 6.83% |
| 25% to 50% | 16.24% | 13.15% |
| 50% or more | 38.06% | 29.07% |

Median repeat rate (the share of all answers to a question that name the company, on the questions where it is named at all, averaged), by the share of its market's questions on which the company is named. ChatGPT and Google AI Mode, August and September 2026; companies from every market Qvery's GEO Agent tracked. Source: Qvery's GEO Agent: repeated answers to tracked buyer questions, ChatGPT and Google AI Mode, August to September 2026

Inside **every market** we tracked, the companies named on a quarter or more of the questions had a higher repeat rate than the companies named on under 5%.

Part of that gap is arithmetic: companies named in a single answer sit at the bottom of the line by construction.

So we split the two numbers across months: penetration from August's answers, the repeat rate from September's answers to the same questions. The gap held. The typical company named on under 5% of the questions in August had a September repeat rate of **0.00%** on both assistants.

Most of them vanished outright: 69.30% were not named on those questions once in September on ChatGPT, and 73.97% on Google AI Mode. The typical company named on half or more had **39.52%** and 31.11%, and none went unnamed.

The split does not remove the plainest explanation: an assistant names some companies more readily than others, and that habit shows on both counts. In buying, that is the law's explanation too. The penalty comes with the smaller share, not with anyone's intent.

The two measures sit on different scales, so the comparison with buying is one of shape, not of a number.

![Two dot grids, rows are buyer questions and columns are days asked. The leader is filled on four of five rows, about eight dots each; the smaller company on one row, two dots.](https://qvery.ai/blog/double-jeopardy-in-ai-search/illustration-two-penalties.webp)

*Read the rows, then the dots in each row: the smaller company loses on both.*

Read the drawing by rows, then by dots: **fewer rows** for the smaller company, and **fewer dots** on the rows it reaches. A buyer who asks any one of those questions, on any one day, is far more likely to read the leader's name.

> A screenshot of your company in one ChatGPT answer is one coin that landed heads. It cannot tell you whether buyers will see your name tomorrow. Ask the same questions over several weeks before treating that screenshot as reach. The question for whoever sent it is how often it lands heads, and on how many questions.
>
> Vlad Shvets, Partner @ Qvery

## The Two Penalties Multiply for a Buyer Who Asks Once

> 🤔 **But My Buyers Do Not Ask Twice** A buyer asks a question once, reads one answer, and moves on. A repeat rate describes the assistant's habits, not the buyer's choice, and an AI answer is rarely the whole decision anyway.

Gartner's survey of 645 business buyers supports that objection: buyers used **seven sources** on average during a recent purchase. **45%** used generative AI, mostly for vendor and product information, and **69%** preferred to check AI insights with a salesperson.

An answer is one step toward a shortlist. This data says nothing about who signed.

But a buyer who asks once gets one draw from that question's answers. Nobody knows which day's answer they will see.

So the repeat rate is that buyer's **chance of seeing you** on that question, asked the way we asked it.

Across a whole market, the chance is your penetration times your repeat rate. The product is **your share of answers**: the share of the answers to your market's questions that name you, each question counted once.

- **25.79%** of all the answers name the typical company found on half or more of its market's questions
- **1.26%** name the typical company found on 10% to 25% of them
- **0.0377%** name the typical company found on under 5% of them

Source: Qvery's GEO Agent: share of all answers to tracked buyer questions naming the company, ChatGPT, August to September 2026

From the top band to the bottom, the share of all answers falls **684.21 times** on ChatGPT and 507.78 times on Google AI Mode, far more than either penalty alone.

Buyers do not ask every question equally often, so read this as the shape of the channel, not a forecast of your inquiries.

The objection has a second form: a company named on few questions will be named rarely on each of them by arithmetic alone. The law works the same way in buying, where loyalty, in the institute's words, "is largely a predictable function of market share."

That is what makes the law useful. If AI answers behave like buying, the repeat rate will not move on its own, and the bigger lever is the first number: **more of the questions**. The institute's finding for brands that grew is the same, with winning new buyers roughly twice as important as keeping old ones.

If each answer were an independent draw, a company named in one answer in ten would go unnamed across five askings in a row **59.05% of the time**. A person who checks a question now and then cannot tell a low repeat rate from none.

GEO Agent: https://qvery.ai/geo-agent/

## What the Agents Do With the Daily Answers

The agents start from a question list built from your own pipeline (sales emails, inquiry forms, lost-deal notes), approved by your pipeline owner, the person who runs the sales pipeline. They read the daily answers on both assistants, find the questions where a rival appears in most answers and you in none, and draft the page changes. The pipeline owner chooses which pages change and approves publication.

That page work is someone's job. Say two marketers spend about six hours a week each turning sales notes and inquiry forms into page changes, at about $127,800 a year each with benefits: **$38,340 a year**. With the agents taking half, 312 hours and $15,570 a year come back after running costs, and the savings cover each $10,000 of the build in 7.7 months.

(An interactive calculator on the page: roi-question-page-drafting.)

Every draft they write is logged with the answers it rests on, so whoever reviews it sees which rival was named, on which assistant, and on how many of the month's askings.

The Questions Your Rivals Own, Listed With Draft Pages by Qvery: https://calendly.com/qvery-vlad/30min

## Bring Us Your Buyer Questions

We start from your buyer-question list and the rivals' names beside it. Within **two weeks** of access, your pipeline owner has the first worklist: the questions where a rival fills most answers and you are absent, each with its page change drafted.

## Frequently asked questions



### Is being named in an AI answer the same as being cited?

A citation is a page listed as a source; a mention means your company appears in the answer's text. This article counts company names because those are what buyers see, and a cited page is a different thing to measure.

### Who in the company should own this?

The pipeline owner should own the question list, because it comes from sales emails, inquiry forms, and lost-deal notes and affects what the company publishes. Agents can handle the daily asking and counting; that person reads the monthly result and picks the questions worth a page.

### What should we do when our repeat rate is below our band's median?

Read the answers that name your company and the pages they cite before deciding what is wrong. A low repeat rate tells you that mentions are inconsistent within that band; it does not establish the cause or tell you which page change will fix it.

### How many buyer questions should we track?

Enough to cover how your buyers really ask, written from sales emails, inquiry forms and lost-deal notes and frozen for a quarter. A list rewritten every month measures your edits instead of the answers.

### Should we focus on the questions we already win?

Keep them, but spend most of the effort on questions where you are never named. The law in this article says the repeat rate moves slowly, so the larger gain comes from being named on more questions, not more often on the same ones.

### What do the agents change on our site?

Nothing on their own. They draft a page change for each question where a rival is named and you are not, and the pipeline owner decides which pages change; a person edits and publishes every one.

### Will the agents see our customer data?

No. Their work runs on public answers and public pages, plus the sales notes your buyer questions come from, so no customer account has to open for them.

### Why did our company drop out of AI answers this month?

Usually because it was barely there. A company named on few of its market's questions sits one or two answers from zero, and most such companies in the data were not named on those questions at all the next month. Read a drop against your band before treating it as something you caused.

### What does it cost to have agents read the daily answers?

About $300 a month for model usage, hosting and upkeep on one company's question list. That is our estimate, and it grows with the number of questions you track.

### Does a smaller company ever beat the leader in AI answers?

On single questions, yes: a specialist named for a narrow need can appear more often than the leader on that question. Across a whole market the leader's share stays larger, so the useful goal is more questions, not the whole market.

### How do we know a page change worked?

The agents compare the share of answers naming you on that question for the weeks before and after the change, on each engine separately. One answer proves nothing; a change shows up as a new level that holds.

### Which AI assistants should we measure?

The ones your buyers use, usually ChatGPT and Google AI Mode, kept as separate columns. They draw on different pages, so a combined figure hides which one needs which change.

### Does double jeopardy hold in business-to-business markets?

The Ehrenberg-Bass Institute finds the law in business-to-business markets as well as consumer ones, and the answers behind this article come mostly from business-software markets. Smaller there means named less often in answers, not fewer employees.

### Does this work for companies that sell through partners?

Yes. The question list then includes how buyers ask for the partner's product with yours inside it, and the page changes may sit on a partner's site, which a person requests through the partner manager.

### Is this worth doing if we are already the market leader?

Yes, for a different reason: a leader's job is to keep the questions it already wins and catch the ones a new rival starts to take. The same worklist shows both, question by question.
