Vendor selection
Vendor selection, scored together.
Lynctera turns your RFP into a scoring rubric, reads every proposal against it, and scores alongside your evaluation panel. Your panel's ratings set the ranking, and every total is calculated.
Security and data protection, weighted 20%
Scored against the rubric your team approved
How it was scored
- All client data stored and processed in the United Kingdom.RFP, requirement S2, must have
- “Client data is hosted in our EU data centres in Dublin and Frankfurt.”Kestrelline proposal, word for word
- The AI's rating: 3 of 5. Partial on a must-have: EU hosting today, a UK region planned for 2027.
See how it works.
Every proposal read against every requirement.
- A rubric you approve firstYour RFP and evaluation sheet become weighted criteria, with every requirement tagged must-have or nice-to-have. Your team approves the rubric before any vendor is scored.
- No quote, no creditEach vendor gets met, partial or gap on every requirement, with its own words quoted. A must-have it misses caps that criterion's rating.
- Scored by your panel and the AIThe AI rates each criterion from the evidence. Each evaluator agrees or gives their own rating, with a reason when they differ. Once your panel has scored, its ratings set the ranking.
Where it helps
- Supplier selection
- Technology and platform RFPs
- Outsourcing reviews
- Re-tendering an incumbent
- Evaluation panels
Every score shows its working.
- A quote on every verdictEach met or partial verdict carries the vendor's own words. No quote, no credit.
- Totals calculated, not generatedWeighted totals and the ranking come from a fixed calculation, never from the AI.
- People on the recordEvery agreement and override is saved under the evaluator's name, and an override needs a reason.
- Your data kept apartHeld in its own database, separate from every other customer. UK data. AI hosted on Microsoft Azure in the EU.
Common questions.
What do we need to start?
Your RFP, your evaluation sheet or scoring criteria, and the vendors' responses: Word, PDF and spreadsheets, however many files each vendor sent. Each file is sorted to its vendor, and you confirm before anything is scored.
Does the AI pick the winner?
No. The AI is one evaluator. Each member of your panel sees its rating and the evidence behind it, then agrees or gives their own. Once your panel has scored every criterion, its ratings set the ranking. Until then the ranking rests on the AI's evidence-based score, and a single override cannot swing it.
How do we know a score is right?
Every met or partial verdict quotes the vendor's own proposal, every criterion rating shows the requirements beneath it, and the totals are calculated, not generated. If a vendor was not fully assessed, the scorecard says so.
How do we start?
Send us an RFP you have already decided, with the vendors' responses. We will show you how Lynctera scores it, and why. Then a two-week pilot on your own work, for a fixed fee.
More of what you get back.
Due diligence, read and cited
Data rooms read against your term sheet and scope, every finding cited to its clause.
Decisions without the meeting
The committee votes independently on the same pack. You meet only on the disagreements.
Briefed before every meeting
Any company, supplier or counterparty researched from its own site and public sources, every point cited.
What was agreed
Answers from your own email and documents, with the source.
Exceptions, already checked
Breaks and mismatches found, with the likely cause attached.
Your own process
Not on the list? Your steps, rules and sign-offs, configured rather than coded.
Try it on an RFP you have already decided.
Send us one, with the vendors' responses. We will show you how Lynctera scores it, and why. Then a two-week pilot on your own work, for a fixed fee.