Most SEO tools are vague about this, and the vagueness costs users real money — decisions get made on numbers whose origin nobody checked. So this page lists every source we use, labelled by how much weight it deserves.
Three labels are used throughout the product and on this page. Measured means we or Google looked at the thing itself. Third-party means a data provider told us, and their coverage is their own. AI estimate means it was generated by a model and should be treated as a starting point, not a fact.
Every source, itemised
Rankings and search performance
MeasuredGoogle Search Console (your own account)
Clicks, impressions, CTR and average position come straight from your verified Search Console property, per country where you ask for it. This is Google reporting on your own site, not a model of it. Position is a 28-day average over queries you already receive impressions for — it cannot tell you where you rank for a keyword you do not rank for yet.
Core Web Vitals and page speed
MeasuredGoogle PageSpeed Insights API
LCP, CLS, INP and the mobile/desktop scores are fetched live from Google’s own API, so they are exactly as fresh as Google’s. No cached third-party index sits in between.
Site audit — technical issues, broken pages, internal links
MeasuredOur own crawler
We fetch your pages ourselves and read what is actually there: titles, meta descriptions, headings, canonicals, schema, images, redirects, status codes and in-content links. Nothing here is inferred from an index — if we report a missing H1, we loaded the page and there was no H1.
Keyword volume, CPC and competition
Third-partyGoogle Ads data from a licensed keyword data provider
Search volume and CPC are Google Ads data from a licensed keyword data provider, for one country at a time (a worldwide search uses the United States and says so). In the Keywords workspace, when that is unavailable the column shows “—” rather than a stand-in. Some free public tools fall back to a Bing result-count proxy, which is a rough popularity signal rather than a search volume — it is labelled differently in the data so the two are never mixed up. Ad competition (how many advertisers bid) is shown as its own number and never as keyword difficulty.
Keyword difficulty and competitors’ keywords (keyword gap)
Third-partyA licensed keyword data provider
Keyword difficulty (0–100) is a score based on how strong the pages ranking in Google’s top results are. The keyword gap uses an index of which keywords each competitor ranks for in Google, fetched when you press Refresh and kept for 30 days. Your own positions come from the same source after a refresh, otherwise from your Search Console. When this data is not available we show nothing in its place.
Bing search counts and related keywords
MeasuredBing Webmaster Tools (your own account)
When you connect Bing Webmaster Tools, the Keywords workspace shows Bing’s own count of how often a query was searched on Bing, per market, and Bing’s related keywords. These are Bing’s numbers, not Google’s — they are labelled as Bing everywhere they appear.
Topic gap
AI estimateCompetitors’ public sitemaps and page titles, read by our own fetcher
We read the sitemaps your competitors publish and a sample of their page titles to list the topics they have pages about. A topic counts as covered when one of your pages, Search Console queries, articles or saved keywords contains its words — a word match, not a judgement of quality. It says nothing about search volume or rankings. Sites that publish no sitemap, or block bots, are shown as such.
Backlinks
Third-partyA third-party backlink index, plus Bing Webmaster Tools where connected
We do not run our own web-wide link crawler, and we will not pretend otherwise — building one is a nine-figure undertaking and the tools that have done it say so loudly. Backlink data comes from a third-party backlink index, and from Bing Webmaster Tools for sites you have verified there. Expect fewer links than Ahrefs or Semrush report.
Domain authority scores
Third-partyA third-party authority score
A third-party authority score, not Moz’s Domain Authority and not Ahrefs’ Domain Rating. Different providers score the same domain differently; use it to compare domains against each other within Flux N Pro, not against a number you read in another tool.
Link risk flags
AI estimateOur heuristics over the link data we hold
We flag links worth reviewing based on the source domain, anchor text, rel attribute, authority and how many links come from one domain. This is not a spam score from a link-graph provider. It flags for review, always shows its reasons, and never disavows anything — because disavowing a good link costs you ranking you cannot easily get back.
Competitor social media (Social Spy)
MeasuredYouTube Data API, Bluesky and Mastodon public APIs
Followers, posts, likes, comments, shares and views are read from each platform’s own public API and snapshotted daily, so growth is measured, not modelled. Engagement rate and share of voice are calculated from those numbers with the formula shown on screen. Instagram, Facebook, LinkedIn, X and TikTok do not let apps read other accounts’ posts, so Social Spy lists them as not available rather than estimating them.
AI content, summaries and recommendations
AI estimateAI language models
Article drafts, content briefs, audit fix suggestions and report summaries are generated by AI from your own project data. They are drafts for a person to review, which is why approval workflow is built into the platform rather than optional.
Local SEO analysis
AI estimateAI estimate from the business details you enter
NAP consistency scoring, citation suggestions and local ranking notes are generated by AI from the business information you provide. There is no Google Business Profile or Places API behind it and no live citation lookup — the tool says so on screen too. Treat it as a starting checklist, not a measurement.
What we don't have
A data-sources page that only lists strengths is marketing. These are the gaps, so you can decide whether they matter for your work before you pay us anything.
Our own backlink index
We use third-party link data. Ahrefs and Semrush have spent years and enormous sums building their own crawlers; we are honest that we have not.
Google Analytics 4
Not connected yet. Traffic figures in Flux N Pro come from Search Console, not GA4.
Traffic estimates for sites you do not own
We do not publish estimated-traffic numbers for arbitrary domains. The industry’s estimates diverge from real analytics by large multiples, and we would rather show nothing than a confident wrong number.
Live SERP rank checking
Rankings come from Search Console. Live SERP checks require a paid SERP provider and are only available when one is configured.
How we label estimates in the product
Where a figure is generated rather than measured, the tool says so on the screen showing it — not in a footnote. Local SEO analysis and keyword cannibalisation grouping both carry an on-screen notice that they are AI estimates, and link risk flags always show the reasons behind them.
When a data source is unavailable, we say the check could not run. We do not record a zero, and we do not let a failed lookup appear as a real result — a fabricated dip in your chart is worse than a gap in it.
Found something on this page that looks wrong?
Tell us and we will correct it. This page is only worth publishing if it stays accurate, and we would rather hear it from you than have you quietly stop trusting the numbers.
Get in touch