# Get Your Competitor Alternatives Page Cited by AI Engines

> Find the best AEO tools lists AI engines cite, learn what editors need before they list your software, and track whether a placement changes AI answers.

- Author: Sheetal Shakshi, Founder's Office, Nurdd
- Published: 2026-11-08
- Publisher: Noma by Nurdd (https://nomahq.app)
- Original page: https://nomahq.app/blog/best-aeo-tools-lists-ai-engines-cite

The best AEO tools lists AI engines cite are mostly ordinary 'best X' and 'X alternatives' articles, and a vendor earns a fair place on them in three moves: find which lists each engine actually quotes, give the editor facts they can check in minutes, and measure whether answers change. In [Ahrefs, December 2025](https://ahrefs.com/blog/best-lists-research/), 'best X' blog lists were 43.8% of all page types cited by ChatGPT.

This guide is for a marketing, SEO or growth lead who has noticed that the 'best AEO tools' and 'alternatives' results are held by other companies and wants to be on them honestly. It does not rely on link-volume figures, because none are published for this category. It gives you a check to run today, a triage table, an editor submission pack and a tracking method.

## Key takeaways

- Do not guess which lists matter. Run your buyer prompts on each engine and write down every list page it cites. The set differs by engine, so one engine's list is not another's (see [why engines name different top brands](https://nomahq.app/blog/ai-engines-disagree-on-top-brand)).
- Most lists are written by companies that sell in the category. Expect a rival's page to be a source and plan for the minority of independent editors who have no stake.
- Send editors a one-page fact sheet with pricing, engines or features, limits, a free way to test and a dated change log. Do not send a press release.
- Never pay for a ranking or swap links. Google's spam policies name both, and an engine that quotes a purchased list is quoting something you cannot defend.
- Being cited is not being recommended. Track mentions and citations separately, on a fixed prompt set, across several weeks, because [cited URLs change from run to run](https://nomahq.app/blog/ai-citation-half-life).

## Which sources do AI engines cite for best AEO tools?

They cite a mix of vendor-published rankings, review directories and independent articles, and the mix depends on the engine and the question. A search for this exact topic on 7 November 2026 returns ranked lists such as [Frase's best AEO tools list](https://www.frase.io/blog/best-aeo-tools-2026), [Conductor's best AEO tools list](https://www.conductor.com/academy/best-aeo-geo-tools/) and [AirOps' AEO tools list](https://www.airops.com/blog/answer-engine-optimization-tools). Each is published by a company that sells software in or next to the category. Those are the pages written to help a buyer choose, so they are also the pages an engine reads when someone asks for the best AEO tools.

Two published studies show how the citation mix is built. Neither measures AEO software specifically, so treat them as patterns in similar commercial questions, not as measurements of this category.

| Study | What it counted | Finding | Limit |
| --- | --- | --- | --- |
| [Ahrefs, December 2025](https://ahrefs.com/blog/best-lists-research/) | 26,283 source URLs across 750 ChatGPT prompts in software, product and agency categories | 'Best X' blog lists were 43.8% of all page types. 79.1% of the 1,100 cited lists it dated had been updated in 2025. 35% of cited lists sat on low-authority domains | ChatGPT only; a snapshot in time; Ahrefs sells SEO software |
| [Onely, October 2026](https://www.onely.com/blog/llm-friendly-content/) | 963 pages cited by ChatGPT, Perplexity, Gemini and Google AI Mode across 31 'best agencies' runs, after removing Onely's own pages | 650 pages (67.5%) were lists of providers. The provider's own page was 24.1% of citations, and 60.5% of ChatGPT's 210 | Agency questions, not software; Onely is itself an agency and publishes the study |

Two readings follow. First, freshness is part of the pattern: in the Ahrefs sample the cited lists were mostly recent, so a list that is not maintained is a weaker target. Second, engines differ. Onely reports that ChatGPT leaned on the provider's own site for follow-up questions, while Perplexity, Gemini and Google AI Mode leaned on lists for the first question. For you, that means the third-party list is the route for the shortlist question, and your own pages carry the follow-up.

> "Any public website can appear in ChatGPT search." [OpenAI Help Centre, as checked November 2026](https://help.openai.com/en/articles/12627856)

OpenAI's publisher FAQ goes on to say the site must not block OAI-SearchBot. That applies to you and to the editor's site, but you only control yours. Check your own crawler access first with the free [AI crawler checker](https://nomahq.app/tools/ai-crawler-checker), which shows whether GPTBot, OAI-SearchBot, ClaudeBot and PerplexityBot can reach your pages. Background on how that engine picks sources is in [ChatGPT search: how it works](https://nomahq.app/blog/chatgpt-search).

## How do I check which lists each AI engine quotes?

Run a fixed set of buyer prompts on each engine, record every cited URL, and rank the list pages by how many engines and prompts cite them. This takes an afternoon by hand for one category and gives you a target list built from evidence, not from whichever list ranks first on Google.

1. **Write 15 to 25 unbranded prompts.** Use the questions a buyer types before they know a vendor: 'best AEO tools for a mid-size B2B team', 'alternatives to [named rival]', 'which AI visibility tool tracks Gemini'. Keep branded prompts separate. The [free AI prompt finder](https://nomahq.app/tools/ai-prompt-finder) suggests questions by intent, and [which prompts to track](https://nomahq.app/blog/which-prompts-to-track) explains how to choose.
2. **Run each prompt on every engine you care about.** Use a fresh session each time and turn personalisation off where you can. Run each prompt three times, because one answer is one draw.
3. **Log every cited URL.** Record the prompt, engine, date, the brands named in order and each source link. Note whether your brand was named, cited, or neither.
4. **Classify each list page.** Mark the publisher type: a company that sells in the category, a review directory, an independent publication, a community thread.
5. **Count.** A list cited on three engines for six prompts matters more than one cited once.

A worked example of the arithmetic: 20 prompts on 4 engines, three runs each, gives 240 answers. If one list page appears in 36 of them, it is cited in 15% of answers. That is the number you compare week to week, and it is illustrative arithmetic, not a measured result for any brand.

Mentions and citations are different events. An engine can cite a list and then name a different brand from it, which is why [citation versus brand mention](https://nomahq.app/blog/citation-vs-brand-mention) should be two columns in your log.

## Which lists are worth pursuing? A triage table

Not every cited list deserves outreach. Use this table once your log is filled in. It is an original framework built from the patterns above, not a published standard.

| List type | Who owns it | What to do | What to avoid |
| --- | --- | --- | --- |
| Independent article or publication | An editor with no product in the category | Send the fact sheet and a free way to test. Offer corrections if a fact is wrong | Asking for a ranking position |
| Review directory or marketplace | A platform with its own listing rules | Complete your profile, keep it current, and follow its rules for collecting reviews | Incentivised or fake reviews |
| Competitor's 'best tools' or 'alternatives' page | A vendor who sells a rival product | Do not pay or trade links. If a fact about you is wrong, ask for a correction with evidence | Link swaps, guest posts that exist only to place a link |
| Community thread or video | Individuals and moderators | Take part openly within the community's rules and disclose your connection (see [community and creator signals without spam](https://nomahq.app/resources/community-and-creators-aeo)) | Posing as a customer |
| Stale list, last updated long ago | Anyone | Ask whether a refresh is planned before investing time | Assuming old pages keep getting cited |

The competitor row is the one teams get wrong. A rival's alternatives page is an advertisement for the rival. The fair move is to make sure the facts it states about you are right, and to build the pages and independent coverage that let engines find a more balanced source. [How to get on the vendor shortlists AI writes](https://nomahq.app/blog/get-on-ai-vendor-shortlists) covers that second half.

## What do editors need before they list your software?

Editors list products they can verify quickly and describe in a sentence. A submission that makes that easy gets read. One that asks for a placement does not. Build a single page, hosted on your own site, that an editor can link to and an engine can read.

### The one-page fact sheet

- **What it is, in one sentence.** Category, buyer and job, in the words buyers use.
- **Pricing as published,** with the date it was last checked and a link to the pricing page.
- **What it measures or does,** and what it does not. Say plainly which engines or features are not covered.
- **Limits.** Prompt counts, seats, projects, regions.
- **A way to test it.** A trial or a free tool, with the exact terms.
- **A dated change log.** Editors and engines both favour pages that are visibly maintained.
- **A contact for corrections.** A named inbox, with a promise to answer within days.

### The pitch email

Keep it under 120 words. Name the specific list, say which section your product fits, and give one reason: a use case the list does not yet cover, or a fact that is out of date. Link the fact sheet. Offer a free account so the editor can check claims. Do not ask for a position, do not offer payment, and disclose that you sell the product.

Why disclosure matters is stated in Google's rules. [Google Search Central, as checked November 2026](https://developers.google.com/search/docs/essentials/spam-policies) lists exchanging money, goods or services for links as link spam, including sending a product in exchange for a write-up with a link. A free trial offered so an editor can verify a claim is different from a free product offered for a link, so keep the offer unconditional and in writing.

### Make the page itself quotable

Engines quote passages. Give the editor, and the engine, sentences that stand alone: the product named in full, what it does, who it suits, with a date. Google's own guidance says that [Google Search Central, May 2025](https://developers.google.com/search/blog/2025/05/succeeding-in-ai-search) treats the technical requirements for search as covering its AI formats too, so there is no separate markup to add for lists. The [free AI SEO checker](https://nomahq.app/tools/ai-seo-checker) reads your fact sheet the way ChatGPT, Claude, Perplexity and Google AI Mode would and flags what stops it being read or quoted. Run it on the page before you send it to anyone.

## Should I pay to be on a 'best tools' list?

No, not for a ranking. Paid placements carry two risks. They can break Google's link policies if the links are meant to influence rankings, and they put you on pages an editor can drop the moment the arrangement ends. The Ahrefs study adds a third: 35% of the cited best lists it examined were on low-authority domains, and it called many of them questionable. A placement there may be cited today and be a liability later.

Ahrefs also reported that brands which placed higher on a list were more likely to appear in the answer that cited it. That is a correlation in 750 hand-checked lists, and it should not be read as a reason to buy the top slot. It is a reason to be accurate and complete on lists where you legitimately belong.

Self-published lists are a related choice. Publishing your own 'best tools' page is allowed, but a list where you rank first and rivals get thin descriptions is an advertisement, and readers can see that. If you publish one, state that you sell one of the tools, use the same criteria for every entry, and date the facts. The checks in [how to choose AEO tools](https://nomahq.app/blog/aeo-tools-how-to-choose) are a fair template for criteria.

## How do I track whether a placement changed AI answers?

Hold the prompt set constant, record a baseline before the placement goes live, and compare the same prompts on each engine over at least four weeks. A single before-and-after screenshot proves nothing, because answers vary between runs.

| Metric | How to calculate it | What it tells you |
| --- | --- | --- |
| Mention rate | Answers naming your brand divided by answers run, per engine | Whether engines name you at all |
| List citation rate | Answers citing the target list divided by answers run | Whether the list is an active source |
| Mention given list cited | Answers that cite the list and name you divided by answers that cite the list | Whether the placement carries into the answer |
| Description accuracy | Answers with correct facts divided by answers naming you | Whether the editor's wording is being repeated |
| Referral visits | Sessions from AI referrers in analytics | Whether any of it sends visitors (OpenAI notes ChatGPT adds a utm_source=chatgpt.com parameter, per its [OpenAI Help Centre, as checked November 2026](https://help.openai.com/en/articles/12627856)) |

Report each engine separately. A rise on one engine and a fall on another can cancel out in a blended figure. How to build the per-engine split and judge sample size is in [how to measure AI share of voice](https://nomahq.app/blog/measure-ai-share-of-voice), and the manual logging steps are in [tracking brand mentions in ChatGPT](https://nomahq.app/blog/track-brand-mentions-in-chatgpt-manual-and-tool).

Read the result honestly. If the list is cited more often but the brand is not named more often, the placement is not doing the job and the description on the list may need work. If you are named but the facts are wrong, treat it as a correction task for the list's editor and for your own fact sheet.

## Where does Noma fit?

Noma by Nurdd is an AEO and AI visibility platform that asks 13 AI engines, including ChatGPT, Gemini, Perplexity and Claude, the questions your buyers ask, every day. For this problem it shows where your brand is named, where competitors are named instead, and which pages each answer cites, so you can see which 'best tools' lists the engines rely on and whether a placement changed anything.

[Book a demo](https://nomahq.app/demo) to see it on your own brand, or [start the 3-day free trial](https://platform.nomahq.app/signup). A card is taken at signup so billing can start on day 3; cancel before then and nothing is charged. Plans and prices are on the [pricing page](https://nomahq.app/pricing).

## Common mistakes to avoid

- **Starting with outreach, not evidence.** You pitch lists no engine cites.
- **Checking one engine.** The Onely figures show citation behaviour differs by engine.
- **Treating one run as a result.** Run three times and compare rates.
- **Counting citations as recommendations.** Log mentions separately.
- **Offering value for a link.** Keep trials and free tools unconditional.
- **Letting the fact sheet go stale.** Recently updated lists dominated the Ahrefs sample, and a stale fact sheet undermines the editor.

## How we researched this

We reviewed the pages ranking for the primary keyword, two published studies of the sources AI engines cite (Ahrefs and Onely), OpenAI's publisher documentation, and Google Search Central's spam policies and AI search guidance. Facts, figures and page contents were checked on 2026-11-07. No Noma customer data or results are used. The studies cover ChatGPT, agency questions and a specific sample, so their numbers describe patterns, not your category.

## What to do next

This week, run your 20 prompts on each engine, log the cited lists and fill in the triage table. Publish the fact sheet and test it with the free [AI SEO checker](https://nomahq.app/tools/ai-seo-checker). Then contact the three independent editors with the most citations. To run the check daily across engines and keep the baseline, [book a demo](https://nomahq.app/demo) or [start the 3-day free trial](https://platform.nomahq.app/signup).