You probably think of an AEO score as something a tool spits out — a number between 0 and 100, a green checkmark, maybe a list of missing schema types. However, a score only means something if the rubric behind it is stable, inspectable, and versioned — and until now, nearly every AEO score on the market has come from a black box. Today we are removing ours.
The AEO Standard — the 100-point rubric we score every page against before it ships — is now public, versioned, and licensed for anyone to adopt.
This is the same system that runs behind our free AEO scanner, the same one wired into the CI pipelines of every site we operate, and the same one our content pipeline enforces before a single post goes live. Nothing was softened for publication. What follows is what is in it, why we opened it, and how to put it to work on your own site this afternoon.
What is the AEO Standard?
The AEO Standard is a 100-point, seven-section scoring system for Answer Engine Optimization, published openly by iSimplifyMe under a CC BY 4.0 license. It measures both whether AI engines can extract a page and whether the page says anything worth citing, with hard gating rules that override the score entirely.
Why Publish The Rubric At All?
The honest answer has two halves. First, Answer Engine Optimization is young enough that most of what gets sold under the name is unverifiable — scores without published criteria, audits without stable thresholds, promises without version numbers. A discipline matures when its measurements become public and contestable. Someone was going to publish the first open standard; we decided it should be the team that has been enforcing one in production for years.
Second — and we say this plainly because the standard itself demands it — publication is an authority play. The rubric's own heaviest section scores originality and demonstrable expertise, and a published, versioned methodology is exactly that kind of artifact. We are, quite literally, practicing what we score.
What are the gating rules in the AEO Standard?
The AEO Standard applies two gates before any points are scored: hidden machine-only content and fabricated authority. Content hidden from human readers — clipped, off-screen, or screen-reader-only body text — and invented credentials, statistics, or citations each trigger an automatic REJECT verdict, regardless of how well the page scores otherwise.
What The 100 Points Actually Measure
The rubric divides into seven sections, and the weighting tells you what we believe matters. Here is the full breakdown:
- Substance & Originality — 25 points. The largest section, deliberately. Does the page make a claim the generic top-ten results do not? Would a one-line LLM prompt reproduce it? Commodity content fails here no matter how clean its markup is.
- RAG / Retrieval Readiness — 20 points. Front-loaded facts, short paragraphs that chunk cleanly, consistent entity names, attributed statistics, and definitive statements. Retrieval pipelines embed chunks — hedged, rambling prose embeds weakly.
- Atomic Answer Blocks — 15 points. Visible, self-contained, 40–60-word answers to distinct questions, each under a question-phrased heading. The blocks you see in this post are the pattern.
- Structured Data & Schema — 10 points. FAQPage, Article, BreadcrumbList, and page-type schema that validates cleanly. BreadcrumbList is the single most commonly missing schema we encounter — including, until this week, on our own homepage.
- Semantic HTML & Headings — 10 points. One H1, properly nested subheadings, questions as headings where natural.
- Internal Linking & Fan-Out — 10 points. Whether the page's cluster covers the adjacent sub-queries a model expands a question into — the least understood check in the rubric, and one of the most decisive.
- SEO Meta & Technical — 10 points. Title and description length bands, canonical, alt text, OG tags, load time. Table stakes, but measured.
All of these add up to a verdict: 95 or above publishes, 85 to 94 revises, 70 to 84 rewrites, and anything below 70 — or any gating violation — is rejected. Every page we ship clears 95 before it goes live, and now you can hold us to that.
What is the difference between mechanical and judgment checks?
Every check in the AEO Standard is marked mechanical or judgment. Mechanical checks — schema validation, length bands, heading structure — can be scored by software. Judgment checks, including all 25 points of Substance and Originality, require a human reviewer or an LLM pass, because whether content is worth citing cannot be measured by regex.
Why A Score Is A Floor, Not A Forecast
Keep in mind that answer engines are not deterministic. A 98-point page can go uncited; a mediocre one occasionally gets lucky. What the score establishes is a floor — the page is extractable, the substance is real, the hygiene is clean — not a forecast of citations. We wrote that disclaimer into the standard itself, in bold, because the alternative is a market of tools selling certainty that does not exist.
That honesty extends to the weights. The point values encode several years of operating a seventeen-site production network against this rubric. They are experiential, not the output of a controlled study — and a standard that overstated its own evidence would fail its own first section.
What does "a floor, not a forecast" mean in AEO scoring?
A floor means the score certifies readiness, not results: a 95+ page is extractable, substantive, and technically clean. It does not predict citations, because AI answer engines select sources non-deterministically. Any AEO tool or consultant guaranteeing citation outcomes from a structural score is promising something the underlying systems cannot deliver.
How To Score Your Own Site Today
There are four ways in, depending on how you like to work. First, the free AEO scanner scores any URL in the browser — no account, no install. Second, for the terminal-inclined, an early preview of the CLI is live on npm: npx aeo-scan <url> checks the core mechanical signals in a few seconds, with the full standard-conformant CLI in development.
Third, the methodology itself lives at the AEO Standard's canonical home and in the public reference repository — readable, forkable, and citable. And finally, the white-paper PDF edition is available for teams that want the standard on paper.
Remember that the judgment sections are where most content actually fails. If your pages score well mechanically and citations still are not coming, the question to ask is the uncomfortable one: would a generic model have written this page? For teams that want the full rubric applied — judgment sections included — that is exactly what our AEO infrastructure work and the Nexus platform do continuously.
How can I check a page against the AEO Standard?
Three ways: the free web scanner at isimplifyme.com/tools/aeo-scanner scores any URL in the browser; the aeo-scan CLI preview on npm checks core mechanical signals from the terminal; and the full methodology is published at isimplifyme.com/labs/aeo-standard under CC BY 4.0, so any team can score content against it directly.
What Happens To The Standard From Here
The standard is versioned like software. Point releases clarify wording, minor versions adjust thresholds, and major versions restructure sections — every change recorded in a public changelog, so a score always references a specific revision. If you find an ambiguity, a contradiction, or evidence that a weight is wrong, the repository accepts issues; this is a standards document, and standards improve under scrutiny.
What we will not do is chase the discourse. The rubric changed slowly in private for years, and it will change slowly in public. After all, a measurement that moves every week is not a standard — it is a mood.
Frequently Asked Questions
Can I use the AEO Standard for my own clients or products?
Yes. The standard is licensed CC BY 4.0 — use it, adapt it, and score against it commercially, with attribution to iSimplifyMe and a link to the canonical home at isimplifyme.com/labs/aeo-standard.
Doesn't publishing the rubric help your competitors?
It hands them a checklist, not a capability. The rubric tells you what a citable page looks like; producing one still requires original substance, and applying the standard across hundreds of pages still requires infrastructure. We would rather compete on execution against a public standard than on secrecy around a private one.
Will the point values change over time?
Probably, and openly. The weights are experiential — refined across a seventeen-site production network — and as evidence accumulates, minor versions will adjust them. Every revision is versioned and logged, so any score can be traced to the exact rubric that produced it.
Does the AEO Standard replace SEO auditing?
No. Roughly a third of the rubric — semantic HTML, schema, meta hygiene — overlaps with technical SEO, and clearing it helps both. The rest measures what SEO tooling ignores: substance, atomic answer structure, retrieval-ready chunking, and query fan-out coverage across a content cluster.
Is there a tool that scores all 100 points automatically?
Not yet, by design. The mechanical checks are automatable and the aeo-scan CLI is being built against them. The judgment checks — 25 points of substance alone — require a human or an LLM pass, and the full CLI will offer exactly that as an optional scoring mode.
Score Something. See Where You Stand.
The fastest way to understand the standard is to watch it grade a page you care about. Run your site through the free AEO scanner, read the full standard, and if the gap between your score and 95 looks like infrastructure rather than copyediting — that is the work we do.