The expert system
Trained as a procurement assessor, not a chatbot
The hardest question buyers ask us is a fair one: why trust an AI to judge a bid? Here's the honest answer — what the system does, how it's kept honest, and where we draw the line. (We're open about how it behaves; the internal calibration stays under the hood — that's the part that makes it work.)
Six things a general AI tool can't do
It knows which rulebook applies
Before it judges anything, the system classifies the tender’s regime — Procurement Act 2023 above or below threshold, light-touch, NHS Provider Selection Regime, or the Scottish, Welsh and Northern Irish rules — because what counts as a strong answer changes with the regime. A general chatbot has no idea which game is being played.
It scores like an evaluation panel
It rewards what real evaluators reward — specific, evidenced, on-question answers — and marks down what they mark down: generic prose, unsupported claims, and waffle. It’s calibrated to behave like the panel marking your bid, not like an assistant trying to please you.
It reads the actual scoresheet
Upload the ITT and it works from that tender’s real criteria, weightings and word limits — telling you where the marks are, which strengths to lead with, and where bidders typically lose points. Guidance grounded in the document in front of you, not generic best-practice.
It’s tested before it ships
Every change to the scoring is regression-tested against a private library of expert-built bid scenarios with known right answers. If a change makes the judgement worse on any of them, it doesn’t ship. This is the discipline a general-purpose tool simply doesn’t have.
It never ghostwrites
It is deliberately pointed away from writing your bid for you. It surfaces gaps, predicts marks and tells you what to strengthen — then you write the winning version in your own voice. Because templated, machine-written answers are exactly what panels penalise.
You stay in control
Every score comes with its reasoning, so you can see why a tender ranked the way it did and overrule it. It’s decision support, not a decision — you always make the call, and you always verify against the official notice.
Where we draw the line on transparency
We'll always show you what the system concluded and why — every score carries its reasoning. We don't publish the internal scoring rubric, the prompts, or the library of expert test scenarios. That calibration is months of procurement expertise, and it's precisely what a copycat would need. Keeping it under the hood is what lets us keep the judgement sharp and honest for you.
Questions, answered
Is this just ChatGPT with a logo?
No. A general AI tool has read the internet but has never sat on an evaluation panel. BidSquirrel’s expert system is purpose-built for UK procurement: it applies the correct regime to each tender, scores with an evaluator’s discipline, and is regression-tested against expert bid scenarios before any change ships. It is also pointed at a different job — sharpening your answers, not ghostwriting them.
Will it write my bid for me?
No, on purpose. Generic, machine-written prose is precisely what evaluators mark down, and it buries the track record, named people and local knowledge that win. The system reviews and scores your draft, flags gaps and unsupported claims, and tells you what to strengthen — you write it.
How do you stop it making things up?
Its guidance is grounded in the documents you give it — your service line and the tender’s own ITT — rather than invented from thin air, and it’s built to flag claims your profile doesn’t support rather than assert them. Every output shows its reasoning so you can check it, and you should always verify an opportunity against the official notice before bidding. It is decision support, not gospel.
Do you use my bids or documents to train your AI?
No. Your documents are used only to produce your own results. We don’t sell your data and we don’t use your bids or uploads to train models. See our Privacy Policy for the detail.
Why won’t you publish exactly how the scoring works?
We’re open about how it behaves and how we keep it honest — regime-aware, evaluator-calibrated, regression-tested, never ghostwriting. We don’t publish the internal scoring rubric, prompts or test scenarios themselves, because that calibration is the work that makes the judgement trustworthy. What matters to you is the result and the reasoning behind it, and both are shown on every score.
See it score your next bid.
The expert system, on every tender that matches your business. From £49.99/mo. Why it beats a bid writer →
Get started