An Ahrefs alternative

→ The link index. Referring domains, link gap analysis, lost-and-found link history, broken-link building, disavow work. All of it depends on owning a continuously refreshed crawl of the whole web. Theirs is among the best that exists. Ours does not exist.

→ Years of stored history. Our trend lines begin the day your domain is added. A tool that has been recording since the last decade can show you the last decade. We will not draw a chart of a past we did not observe.

→ Daily rank tracking, site-wide crawl auditing at their depth, content explorers, PPC tooling. Not built, not sold, not implied anywhere below.

That list is not modesty, it is the price of admission. Everything past this line is something the product does today — which is the only standard this page is held to.

Search did not stop mattering; it stopped being the only place people ask. The common setup today is a deep SEO suite for the first channel and an add-on, or nothing, for the second.

The audit here scores one domain on both at once — classic SEO basics and AI-citation readiness in the same pass, with the exact markup to paste for whatever is missing. One run, one picture, nothing to reconcile between two tools that disagree.

See AI SEO for how the two jobs differ, and What is AEO? for the definition.

AI visibility is mostly sold as mention counting: how often your brand name turns up across AI answers. That is a real signal, it is worth having, and a rising line is genuinely good news.

It is a different measurement from the one that decides whether the work paid. Being named is not the same as being the source. For a fixed question you care about, we record whether your domain was cited — and when it was not, which domains were instead. Same questions every run, so the series means something.

We keep that separate from the readiness score on purpose, because the two are allowed to disagree and usually do at first. A site can pass every checklist item that exists — schema, llms.txt, crawler access, clean server-rendered HTML — and still be cited by nothing. Readiness says you are quotable. Only the citation check says you were quoted.

Their free product hands you data about your own site, and that instinct is exactly right: prove the domain is yours, then get what is yours to see. Credit where it is due — plenty of tools would rather sell you a login than ask you to prove anything.

Our claim flow makes the same gesture and then keeps going. The proof does not stop at unlocking a dashboard; it becomes an artifact published on your own domain — a v=AEO1 record at _aeo.yourdomain.com, signed, pointing at a fact catalog where every fact carries its source and fetch time.

The difference is who the proof is for. Ownership verification inside a product is for the product. A published record is for the engines — and it works whether or not anyone is paying us, because the specification is open.

Keyword tools report one number: how often a phrase is typed into a search engine. We report two — that one, and an estimate of how often the same phrase appears in prompts to AI assistants.

They diverge, often violently. We have measured phrases with almost no Google volume carrying more than ten times that demand in prompts, and category terms with thousands of monthly searches and effectively none. Judged on search volume alone, the first kind looks like nothing worth writing about — which is precisely why so little has been written for them.

Both columns are modelled estimates and are labelled as such everywhere they appear, including in the free audit, where a blank means no estimate is available for that term rather than demand of zero.

A competitor's ranking footprint, the terms they surface for, and whether an answer engine names them are all observable from outside. No access to their site required.

→ Verified — 3 tracked competitors, snapshots monthly.

→ Grounded — 10 tracked competitors, snapshots weekly.

A snapshot describes ranking footprint. It is not traffic, it is never presented as traffic, and a snapshot not yet captured is shown as pending rather than as a zero.

An audit that ends in a list of problems has handed you a second job. Ours ends with the fix in a form something can act on.

→ A copy-paste prompt. One block of text for any AI assistant — Claude, ChatGPT, Cursor, Gemini — carrying the exact files, markup and DNS record, written to merge with what you already have rather than overwrite it.

→ An MCP endpoint. Remote, free, nothing to install. An agent audits the domain, applies the fixes with its own access, then re-runs the audit to confirm the score moved.

→ A report to send upward. Scores and deltas, owner-versus-competitor trends, citation share, the AI-versus-Google demand table, and an explicit note of what is not measured.

Ours is annual, theirs is monthly, so compare the year. Grounded AEO is $99, $499 or $999 per year — Listed, Verified, Grounded. The audit is free and needs no account.

As of August 2026, ahrefs.com/pricing lists Starter at $29/mo, Lite $129/mo, Standard $249/mo, Advanced $449/mo and Enterprise $1,499/mo, with up to 17% off for annual payment. Their AI-visibility product, Brand Radar, is sold separately from $199/mo, with custom prompt packages at $50, $100 or $250/mo.

Do the year on that. The AI-visibility add-on alone lists at about $2,388 — more than twenty times our Listed tier, and over twice our top tier, before any SEO plan sits under it. Standard plus Brand Radar at list is roughly $5,376 a year, about five times Grounded.

Different products, and not a feature-for-feature comparison — they are buying you a link index we have already conceded. The narrower point: the answer-layer half of the problem does not have to arrive as a four-figure annual add-on.