How it works - 4 min read
We prepare, your adviser lodges: how our process works end to end
A walkthrough of the whole process: from connecting a repository, through the free preview and the guided interview, to a substantiated package your registered adviser reviews and lodges. What you do, what we do, what they do.
If you are weighing this up, the fair question is: what actually happens, step by step, and who is responsible for what? This is the whole process end to end, with no hand-waving about the parts that matter, including where you pay and where your adviser takes over.
The one-line version: we prepare and substantiate your claim from your GitHub history; your registered tax adviser reviews it and lodges it. Here is the long version.
This is general information, not tax advice. Your registered tax adviser determines and lodges your actual claim.
Step 1: Connect a repository (free)
You start by connecting one or more GitHub repositories through a per-company GitHub App install, scoped to the repos you choose. Nothing leaves your control that you have not granted, and you can connect more than one repo for a single company.
There is no card required to get this far.
Step 2: We ingest and triage (free)
We pull your pull requests (or your commit history, for repos that do not use them), the contributors behind them, and the timestamps. Then we run a deterministic first pass that sets aside the obviously routine and mechanical work: dependency bumps, lockfile-only changes, docs-only changes, bot-authored commits. This is cheap, fast, and it focuses everything that follows on the work that might actually be R&D.
You see the result: how many PRs we found, how many we set aside as routine, and why.
Step 3: See where your R&D likely sits (free)
We build a map of your codebase, classifying each module by its apparent purpose and flagging which parts look like core product versus internal or routine work, with the signals behind each call. This is a genuine preview of value at no cost: before you pay anything, you can see where your potentially eligible work is likely to sit, ahead of confirming it in the interview.
This free side, connect, ingest, triage, and map, is deliberately the end of the no-cost portion. It is also where you decide whether to go further.
Step 4: Start the guided interview (this is the paywall)
Starting the interview is where you pay. We are upfront about why the line sits here: everything before it is cheap and bounded, and the interview is the expensive, open-ended work that is the heart of the product. You pay a flat fee banded by the size of your estimated claim, never a percentage of your refund.
The interview itself asks your engineers specific questions grounded in their actual code changes, "you reworked this three times with two reverts, what did you not know going in?", and maps each genuine experiment against the four-gate statutory test. You confirm what is genuinely R&D and scope out what is not.
Step 5: The skeptic pass
Before anything reaches your package, a separate adversarial pass challenges each candidate experiment and scores it (genuine R&D, weak, or not R&D). Its job is to remove unsupportable claims, not pad them. This is the de-risking step, and it runs on every claim because precision matters more than volume.
Step 6: Hours and cost
You estimate effort on a small, stratified sample of pull requests, and we extrapolate to a defensible per-contributor figure, deducting routine work and non-PR overhead. A named reviewer attests the numbers. Every eligible hour ties back to specific pull requests and the experiment they belong to.
Step 7: Your package, and your adviser takes over
You get a reviewable package: the confirmed experiments, their four-gate substantiation, the evidence each one links to, the hours and cost per contributor, and an estimated dollar offset. This is the deliverable.
From here, your registered tax adviser reviews it, applies their judgment on eligibility, handles the tax treatment, and lodges the claim. They remain the backstop and the only party who lodges. We do not lodge, calculate your final lodged figure, or give tax advice, by design.
Who is responsible for what
To be completely clear about the division:
- You confirm what is genuinely R&D, attest the technical facts and the hours, and own the scope.
- We read your history, surface candidates, structure the substantiation against the test, and produce the package. We prepare and substantiate; we do not lodge or advise.
- Your adviser reviews, determines eligibility, and lodges, and is the professional backstop.
The whole thing is designed to take about an hour of your time, and to produce something built to be defended if the claim is ever questioned, not just lodged.
This article is general information, not tax advice, and not a determination of eligibility. Speak to a registered tax adviser before lodging an R&D Tax Incentive claim.