Answers scattered.
The answer to one requirement can sit across a method statement, an appendix and a CV.
Does this bid meet every requirement?
Every bid read against every requirement in the tender, so evaluators start from the same evidence and spend their time on judgement, not searching.
See it worked throughLong bids.
Tight deadlines.
A tender can draw bids hundreds of pages long. Evaluators must check each against every requirement, fairly and in the same way.
Scores drift.
Two evaluators can read the same answer differently, and moderation takes days.
Challenges follow.
A losing bidder can ask why. The reason for every score has to be clear.
What the AI reads.
Every bid read against the tender’s requirements, scoring guidance and mandatory conditions.
In
- Tender requirements
- Scoring guidance
- Bid responses and appendices
- Certificates and accounts
- Clarification questions and answers
- Evaluator notes
At work
- Sense Reads every bid document, from method statements to scanned certificates.
- Understand Finds the answer to each requirement, wherever it sits in the bid.
- Check Tests each answer against the requirement and links the finding to its page.
- Operate Prepares a findings sheet for each evaluator and flags where scores disagree.
Out
One findings sheet per bid. Every finding linked to its page.
Every requirement checked.
Every finding sourced.
Evaluators see where each bid answers each requirement, what is missing, and the page that shows it.
6 requirements · 4 met · 1 partly met · 1 missing
| Req. | Requirement | Finding | Detail | Source |
|---|---|---|---|---|
| R1 | Public liability insurance of £10m | Met | Certificate valid to next June | App. C, p. 2 |
| R2 | Two years’ audited accounts | Met | Both years included and signed | App. D, pp. 1–24 |
| R3 | Plan for winter gritting | Partly met | Covers roads but not footpaths named in the tender | Method, p. 14 |
| R4 | Named site supervisor with a valid licence | Met | Supervisor named; licence number given | CVs, p. 3 |
| R5 | Carbon reduction plan | Missing | No plan found in the bid or its appendices | Not found |
| R6 | Price within the stated budget | Met | Annual price inside the published range | Pricing, p. 1 |
Missing
No plan found in the bid or its appendices.
R5 · Carbon reduction planPartly met
Covers roads but not footpaths named in the tender.
R3 · Plan for winter gritting
Illustrative example. Tender, bidders, requirements and amounts are fictional.
Built for fair procurement.
The AI finds the evidence. Evaluators score and the panel decides.
People score every bid.
Findings support the evaluators. Nothing is scored or excluded automatically.
How Operate works ↗Every finding cites its page.
Each finding links to the page of the bid it came from, so every score can be explained.
How Check works ↗Runs in your environment.
Bids stay confidential inside your boundary, including air-gapped.
Deployment options ↗Same records.
Other questions.
The documents and data behind this work answer other questions too.
Did they deliver what the contract says?
Contract compliance
See whether obligations were actually met.
See how →Which grant claims don’t add up?
Grant claims
Check claims against the rules and the evidence, and flag the ones that need a closer look.
See how →What does this new rule change for us?
Regulatory changes
Map new rules to what the business must change.
See how →Where is the evidence for the audit?
Audit evidence
Pull the proof together from every system.
See how →Bring us a closed tender and its bids.
We’ll show you where each bid meets each requirement, and where your evaluators’ readings differed.
Talk to us →LondonDiscovery AI Limited · United Kingdom
HyderabadReDscvry Technology Private Limited · India
ChennaiReDscvry Technology Private Limited · India