FIG. 2 — CLASSIFICATION CRITERIA
Methodology
Most SEO tools sit on a spectrum from real infrastructure to a thin AI wrapper — the honest answer to "could I build this myself?" is "it depends which tool." SEOWrapper sorts tools into three tiers based on observable technical signals, not opinion.
FIG. 1 — EXPLODED VIEW · SEO SOFTWARE · THREE STRUCTURAL CLASSES
01INTERFACEReports, filters, exports
02QUERY LAYERSearches what's already been collected
03STORAGE / INDEXA searchable record of links, going back in time
04CRAWL PIPELINEDedupes, scores, and saves each page
05OWN CRAWLERCrawls the web non-stop, years of history
SPECIMEN →
Part count is a description, not a score. A two-part tool that saves you an hour a week is worth paying for. Knowing which one you bought is the point.
Data Moat
Runs its own large-scale web crawler and/or maintains years of historical backlink, ranking, or keyword data, often paired with a licensed panel of real users' browsing data it doesn't collect itself. AI may sit on top as a feature layer, but it isn't the product. Rebuilding the underlying data asset isn't something an individual can do with a prompt.
API Stitch
Real, if modest, engineering — login and account handling, scheduling, data storage, the interface — built on top of third-party APIs (e.g. Google Search Console, GA4) rather than a proprietary data moat. A capable non-developer using an AI coding tool could plausibly rebuild a serviceable version.
Pure Wrapper
Typically a prompt — sometimes with a scrape — wrapped in a UI and a subscription. Rebuildable in an afternoon with an AI coding tool.
How we classify
We prefer letting an observable technical signal do the work — Does the tool run its own crawler? Does it call a third-party AI model? Is there a public API or tech blog describing its infrastructure? — rather than editorializing. Verdicts are about buildability, not integrity, value, or legitimacy: a wrapper being simple to rebuild is not the same as it being a bad product. Packaging, the interface, and good prompt-writing are real work even where it doesn't cost much extra to build on top of an existing API or model.
How we write: plain language, then the evidence
Most people reading this are marketers, not developers, even though every entry is a technical teardown. So we split the writing in two. The opening description and the verdict — can you just connect an API and build it yourself? — are written to stand on their own: read only those and you have the full picture, no unexplained jargon, no developer background required. The build-your-own section and the small footer underneath it (pricing, sources, an FAQ) are more technical on purpose — for anyone who wants to check our work, or actually build the thing, rather than just take the verdict on faith.
Glossary
A few terms that show up in the technical sections. SEO-native terms you already know — backlink, SERP, keyword — aren't included here on purpose.
CRAWLERA program that visits web pages on its own, automatically, to collect what's on them — the thing that builds a search index or a backlink database.
CORPUSThe body of data a tool has collected and stored — its database, in effect.
APIA defined way for one piece of software to request data or actions from another, over the internet — how most tools without their own crawler get their data.
LLMAn AI model that generates text — what's behind ChatGPT, Claude, and most "AI writing" features.
ENDPOINTA specific address a piece of software calls to get one kind of data — one small piece of a larger API.
INDEXA pre-organized, searchable copy of data — built once so it can be queried instantly later, instead of gathering everything from scratch each time.
Corrections
If you're part of a team behind a tool listed here and think we've got something wrong, see the About page for how to reach us.