The web is drowning in generic AI text, and Google and the answer engines are learning to ignore it. Our flagship service builds pages from real data — so each one says something no one else can, survives AI-content penalties, and earns citations. This is the moat.
Generic AI content is a commodity and, increasingly, a liability. It costs almost nothing to produce, so everyone has it — and because everyone has it, an engine gains nothing by ranking or quoting yours over the next identical draft. Google's helpful-content and AI-content signals are built to spot exactly this pattern, and the answer engines skip past it because there is nothing distinctive to cite.
Content grounded in real, specific data is differentiated by definition. If a page reports facts, statistics and specifics that exist nowhere else, no volume of generic output can substitute for it. That is what survives Google's AI-content penalties, and that is what earns AEO citations — because answer engines quote sources that hand them concrete, checkable specifics.
Put plainly: real data per page = not slop. The words can be machine-written; the data underneath cannot be faked. The data is the defensible part, and the data is what we build every page around.
The differentiator is identical across all three — real data per page. They differ only in where the data comes from and how that part is billed.
You already hold the dataset — exports, a product feed, survey results, an internal spreadsheet. We analyse it and generate unique pages from it. Billed as usage credits for the analysis and generation; no scrape, no subscription.
You have no dataset yet, so we build one for you — and you own it. This is the one hybrid: billed as a retained subscription for the ongoing scrape plus usage credits for the content generated on top. We show that split explicitly.
We generate from our own in-house data platform — the datasets we grow out of real content jobs. You get differentiated pages without sourcing anything. Billed as usage credits for generation.
Generation automates cleanly; publishing to your property is the only part that needs access — and even that is scoped and revocable.
| Mode | How it's delivered | Access needed | Who publishes | Priced on |
|---|---|---|---|---|
| ① DIY | we generate the pages → you download & publish | none | you (or your AI) | cheapest |
| ② Done-for-you | you grant scoped access → our agents publish → you revoke | OAuth / app-password / temp invite | us (agents) | value, not tokens |
| ③ MCP | your AI calls content generation as a tool | none | your AI, via our tool | credits per call |
Two billing models, kept deliberately separate. All non-free pricing is enquiry-based — we quote against your real scope, never a package number.
Analysis and page generation are bought and burned as credits. This covers BYOD in full, the generation half of scrape-for-you, and our-data. Pay for the pages you actually build, nothing standing.
The only part billed as a retained subscription is the scrape-for-you data feed — an ongoing scrape that keeps your owned dataset fresh. If you're not on scrape-for-you, there's no subscription.
Every page is built from real, specific data — not generic prose an engine has seen a thousand times. That's the whole point: real data per page is differentiated by definition, so it survives Google's AI-content penalties and hands the answer engines concrete facts and statistics worth citing.
Bring-your-own-data — you supply the dataset, we analyse and write from it. Scrape-for-you — we scrape the data and you own it. Our-data — we build on our own in-house datasets. All three share the same differentiator — real data per page — and differ only in where the data comes from and how it's billed.
You do. In the scrape-for-you model the dataset we build is yours. That model is billed as a retained subscription for the ongoing scrape plus usage credits for the content you generate on top of it — the one place the two billing models combine, and we show that split explicitly.
Answer engines cite sources that contain quotable specifics — numbers, comparisons, dated facts. A page grounded in real data is full of exactly that, which makes it citation-worthy, whereas generic AI text gives an engine nothing distinctive to quote. That's why this is our flagship AEO play.
Generation is billed as usage-based credits. The scrape-for-you data feed is the only part billed as a retained subscription. Everything is enquiry-based against your real scope — we don't publish package prices, because the work depends on the data and the number of pages.
The free audit shows your content gaps versus competitors — the pages where real data would win you rank and citations.
Run my free audit →