How we work
The question behind every content purchase in this vertical is whether the writer made the numbers up. Here is our answer, in enough detail that you can hold us to it.
What happens between your brief and your inbox
Six steps. The fourth is the one most suppliers skip, and it is the reason the rest of this page exists.
-
We read what you already have
Your existing pages, your brand facts, your keyword set, your previous briefs. A supplier who starts from scratch every time produces work that contradicts what is already on your site, and that inconsistency is what stops a search engine trusting it.
-
Research, with the SERP ahead of the rulebook
We check what actually ranks for the term before applying a convention. Standard practice often does not hold in these markets, and where we depart from it we say so and give the evidence that justified it.
-
Draft, with a source tracked per claim
Every number gets a source attached as it is written, including sources that will never be published. A draft that arrives without them cannot be checked afterwards, which is why this is not a step we can add at the end.
-
An independent check, blind to the drafting
A separate pass goes through the draft against those sources, without having seen the conversation that produced it. It looks for unsourced figures, invented first-hand experience, claims that drifted from their source, and anything that needs a legal review.
-
Mechanical checks
Title and meta pixel widths, duplicate titles across the site, keyword placement, heading structure, word counts per section, internal link targets. Things with one correct answer get computed rather than eyeballed.
-
Delivery in a format you can use
One document per page, with the sources. Spreadsheets returned in your own format with columns added. Live-page edits as find-and-replace pairs. Nothing that needs interpreting before someone can act on it.
The accuracy rules
These do not bend for a deadline
They are about being right rather than about ranking, which is why they are the only rules here with no exceptions.
- Never invent a number. Bonus terms, wagering requirements, fees, limits, RTPs, payout percentages. Unsourced means it does not ship.
- Never fabricate testing or first-hand experience. No "we tested", no "our team deposited", no "in our experience", unless you did it and sent the evidence.
- Track sources during drafting, including ones that stay unpublished, so the check at the end has something to check against.
- Never copy a competitor's structure or claims. Research tells us what intent a term rewards. It does not license reproducing somebody's page.
- Watch for pages competing with each other across the whole site, and across brands where a client runs several.
- Flag compliance, never resolve it. We are not your compliance function, and a contractor guessing at a jurisdiction's rules produces confident wrong answers.
Why the same fact has to read the same on every page
A site that states its minimum withdrawal three different ways on three pages has a trust problem that no amount of link building fixes.
It used to be a tidiness issue. It is now a ranking one. AI answer engines have to decide whether to repeat what your page says, and contradictory facts across a site are exactly what stops one citing it confidently. A human reader forgives an inconsistency. A model treats it as a reason to quote somebody else.
So brand facts live in one versioned record per client: official names, spellings, product names, licence details, the figures that have to match everywhere. Every page is written from that record rather than re-derived each time. When fresh evidence contradicts it, the record gets corrected and we tell you, rather than quietly working around it on one page and leaving the other forty wrong.
The same discipline runs across a batch. Before a set of pages ships, any claim of the form "the only", "the first", "the largest" or "the cheapest" is checked against every other page in the batch, because that is the claim most likely to be true on page three and false on page thirty.
While researching this site we read five iGaming content agencies. Three of them contradict their own published numbers: one claims more than thirty languages and lists twenty-four on the same page, one gives two different team sizes and three different language counts across its own pages and blog, and one states two different amounts of iGaming experience on two pages.
None of that is dishonesty. It is what happens when nobody owns the facts.
Common questions
What happens if you cannot source something?
You get told, and the page says less. We do not fill a gap with a figure that looks right.
In practice this comes back to you as a short list of questions: what is the actual wagering requirement, what is the real withdrawal window, which licence number should appear. Answering takes a few minutes and it is the difference between a page you can defend and one you cannot.
Who does the checking?
It runs as a separate pass with no sight of the drafting, against the sources attached to each claim. That matters more than it sounds. A check made by whoever wrote the draft already believes every claim in it and waves them through, which makes the check theatre.
Do you use AI to write?
We use it as a tool, the way any studio now does, and we are not going to pretend otherwise. What matters is what happens around it: research from real sources, a tracked source per claim, an independent check before delivery, and a read-through for the patterns that make text feel machine-made.
If you need content to pass a specific AI detector at a specific threshold, say which one and what score. We will check against it and tell you honestly whether we are hitting it.
How long does a batch take?
It depends on the research load, and we quote the timeline with the price rather than publishing one. A catalogue page built from a supplied data sheet and a market guide that needs regulatory sourcing are different jobs.
What we will not do is quote a 24-hour turnaround on work that needs sourcing. That number is a promise about typing speed, not about accuracy.
What does the handover look like?
A document per page, with the body, the SEO title, the meta description, the H1 and the heading structure separated out, plus the source list. Spreadsheets come back in your format with columns added, never reformatted into ours.
For edits to live pages, you get the exact original sentence and its replacement so somebody can find it and swap it without interpreting anything.