Bid analysis software that prices what every bid left out, and cites the page it came from
Hidden costs priced before award, and every score backed by the page it came from.
See one case, screen by screen ↓Bid analysis software for sourcing teams. Eight agents read every supplier bid, rebuild it into the RFP's own cost lines over the contract life, price the exclusions, currency, payment terms and indexation, check the mandatory requirements and draft the clarification questions. The panel scores quality itself from cited evidence with prices sealed, and the procurement director signs the award.
How to choose AI solutions for back-office and enterprise work →
Updated
Suppliers never price the same work the same way. One quotes an all-inclusive annual fee, one unit rates per site, one a lump sum in pounds, one a rate per square metre. The sourcing manager rebuilds them by hand in a levelling spreadsheet: lump sums split by guesswork, unit rates multiplied by estimated volumes, a foreign-currency bid converted at whatever rate is to hand.
The real cost hides in the appendices: weekend call-outs "at schedule of rates", mobilisation "quoted separately", consumables "at cost plus", indexation above the cap, a surcharge for 60-day payment. A capped liability or an expired certificate slips through. Panel scores arrive without evidence, the consensus is an average nobody can explain, and a panel member who once worked for a bidder scores that bidder. When a losing bidder asks how the award was reached, the file cannot say.
Estimated split for a four-bid services RFP, by hand and with the solution. Days are working days for the sourcing team; evaluators still read and score quality themselves.
Eight agents read, level, check and challenge every bid; the panel scores and the director signs.
Kestrel & Vane Group's facilities RFP for seven sites in the Netherlands, Belgium and Germany — 184,600 m², three years from 1 Feb 2027 — closed on Fri 9 Oct with four bids, and no two priced the work the same way. Here is how Lena Hartmann, the sourcing manager, works through it with the agents, screen by screen, in the working solution.
Lena opens RFP-2026-FM-014 and the agents have done the levelling. Each bid sits in the RFP's cost lines over three years, every cell with a confidence and a "from p.N" tag, and the exclusions are written under each column. Orion Facility Services is lowest on the headline at €1,840.0k; normalised, Meridian Workplace Partners is lowest at €2,038.7k and Orion falls to third.
“Orion, lowest on the headline, moves to third at €2,210.0k: +€370.0k it left out, €171.3k more than Meridian. All four bids reconcile to their own totals.”
She clicks the tag on Orion's out-of-hours line, +€214.6k. Page 14 of Orion's commercial bid opens — Appendix C, the schedule of rates — with the sentence that excludes weekend and public-holiday call-outs from the fixed fee highlighted, and the rates: €95 a call-out, €50 an hour. The normaliser priced the RFP baseline of 1,480 call-outs at Orion's own rates.
“Sealed bid. This view is logged for Lena Hartmann at 09:40.”
The table starts from the headlines as submitted and the normaliser replays each step with its source: Orion's indexation at CPI + 3.5% from year 2 against the cap of CPI, maximum 3% (+€66.2k, page 31), mobilisation quoted separately (+€43.2k), the 2.5% surcharge for 60-day payment (+€46.0k), Brightwater's pounds converted at the bid-close rate of 1.1494. A cross-check then re-adds each rebuilt bid to its own total: 4 of 4 reconcile.
The normaliser never prices an exclusion without a rate from the bid or the RFP; with no rate, it asks for a question.
Eight agents, each with its plan and steps, the sources it read, its tools, a confidence and the person it hands to. The bid reader rebuilt 212 pages; when Taskline's price workbook was locked, it asked for the password through the portal instead of guessing. The normaliser could not price Brightwater's payment terms and left them as question Q5. Three agents are waiting for a person.
“95% · below 85% the agent hands to a person.”
The compliance checker tested 18 mandatory requirements against every bid's certificates and terms. Meridian meets all 18. Orion's clause 14.2 caps liability at €1,000,000 per event against the €5,000,000 required; Taskline's ISO 45001 certificate expired on 30 Jun 2026 and its bid is valid 60 days, not 120. Lena opens Orion's M05, reads the clause beside the contract template and chooses DV-02, deviation to clarify before award — a reason is required.
“The compliance checker proposes; you decide.”
The challenge agent drafted a clarification for every gap the agents could not close, each linked to its cell or requirement. Q4, on whether the 1,480 call-out baseline includes public holidays, is about the RFP itself, so it goes to all four bidders without naming who raised it. Replies are due Mon 19 Oct, 17:00 CET, through the sourcing portal.
“Nothing is sent until you press Send.” The challenge agent may not invite a change of price.
Lena sends Q1 to Orion. At 09:53 Orion confirms that the €95 call-out and €50 hourly rates apply at all seven sites for the full term. The challenge agent reads the reply, confirms the rates the normaliser used, and the out-of-hours line stays at +€214.6k with its confidence raised from 92% to 99%.
Ingrid scores service delivery 0–5 against the descriptors, with the evidence gatherer's passages for each bid beside her — 142 in all, each with its page. She sees no price, cost line or price score. Her Taskline score cannot be submitted until she writes a rationale and ticks at least one passage she relied on. Jonas Weber, who declared that he worked for Brightwater FM until March 2023, cannot open or score Brightwater.
“Price is sealed. Panel members do not see any price, cost line or price score until the quality scores are locked.”
The moderation checker flagged Orion's service delivery at 5 / 5 / 3: Pieter Claes's rationale cites no passage, and the staffing plan on page 44 names a mobile team of six engineers for weekend cover. Tomás, who chairs and does not score, records a consensus of 5 with the reason, then Meridian's sustainability at 4, and locks moderation. Only then does the price envelope open to the panel.
“No average is offered: the panel agrees one score.”
The award writer recommends Meridian Workplace Partners at €2,038.7k over three years, 88.00 of 100: €247.3k below the incumbent baseline and €171.3k below Orion's true cost. Meridian stays first for any price weight from 20% to 60%. Lena approves the shortlist and sends it for signature; Catherine signs AW-01, most advantageous bid on the published criteria. The three debrief letters are drafted and go out only when Lena sends them.
“I have read the levelling, the moderation record and the declarations of interest.”
The same RFP, seen by five of the people who carry it — what the evaluation looked like, and what it looks like now.
Eight agents read, level, check and challenge every bid; the panel scores and the director signs.
Reads each bid — cover letter, price workbook, appendices — and rebuilds it into the RFP's 11 cost lines, tagging every value with its page.
Prices exclusions from each bid's own rate card at the RFP baselines, converts currency at the bid-close rate, brings payment terms to 60 days and carries indexation over three years, then re-adds every line to the bid's own total.
Checks the 18 mandatory requirements and the contract template's terms in every bid, against certificates, forms and clauses.
Drafts a clarification question for every gap, decides who must receive it, reads the replies and re-levels the bid.
Collects cited passages for each quality criterion from each bid, for the panel.
Compares individual scores, flags spreads of 2 or more and rationales that cite nothing, and prepares the moderation agenda.
Watches declarations of interest, access to sealed bids, late bids and contact outside the portal, and runs the abnormally-low test on normalised cost.
Ranks the bids, tests the weights, and drafts the recommendation memo and the debrief letters, all cited.
Approve the shortlist, sign the award. The agents propose; a named person decides.
Ask about any bid, cost line, requirement or score, or say what to send. Every answer cites the page it came from, and anything that goes to a bidder waits for you.
Why did Orion move from first to third on cost?
Its headline leaves out €370.0k. Orion's fee of €1,840.0k excludes weekend and out-of-hours call-outs, priced from its own rates at the RFP baseline of 1,480: +€214.6k. Mobilisation is quoted separately: +€43.2k. It indexes at CPI + 3.5% against the cap of CPI: +€66.2k. And 60-day payment adds 2.5%: +€46.0k. At €2,210.0k it is third, €171.3k above Meridian.
Which bids miss a mandatory requirement?
Three bids, six gaps. Taskline fails two: its ISO 45001 certificate expired on 30 Jun 2026 and its bid is valid 60 days, not 120. Orion deviates on two: liability capped at €1M against €5M, and indexation above the cap. Brightwater has two to clarify: 45-day terms and no staff transfer. Meridian meets all 18. Each gap has a drafted question.
Send the clarification questions that are ready
Sent through the portal. Q4 went to all four bidders without naming who raised it, because it is about the RFP's call-out baseline. The others went only to the bid they concern. Replies are due Mon 19 Oct, 17:00; I will read them and re-level any bid that changes.
Would the award change if price were weighted 30%?
No. Meridian stays first. Meridian and Orion have the same quality points, so Meridian leads at every price weight from 20% to 60%. This is a test only; the published weights do not change.
13 screens from the working solution, on its sample data. Pick one to see it large.
The rules every bid is levelled and scored against are settings. Changes are saved as a version the director approves, and a live event keeps the version it started with.
| Setting | Default | Choose from |
|---|---|---|
| Criteria and weights | FM services v3: cost 40, service 30, mobilisation 10, sustainability 10, risk 10 | Templates such as Goods v2 or Professional services v1 · weights must total 100% |
| Contract life for comparison | 3 years (the base term) | Set per template or event |
| Currency | Euro; other currencies at the rate on the closing date | Base currency and the conversion rule |
| Payment standard | 60 days; a shorter term priced at the bid's own surcharge, else 8.4% a year | Standard days and the cost of capital |
| Indexation | Cap CPI, max 3%; CPI assumed 2.4%; excess priced over the life | Index, cap and assumption |
| Exclusions | Priced from the bid's own rates × the RFP baselines; no rate → a question | Baselines come from the RFP's price schedule |
| Abnormally-low test | More than 15% below the average normalised cost → explanation requested | Threshold against the average normalised cost |
| Moderation | Flag a spread of 2; a rationale must cite at least one passage | Spread threshold on the 0–5 scale |
Every figure is labelled: a target the solution is built to, an estimate, a typical published result, or a proven one.
“demo” = seen in the working solution, on its sample sourcing event · “target” = the design goal, measured in the live solution · “estimated” = our estimate · “typical” = published figures (Info-Tech Research Group — 2–3 hours per 112-page RFP response; GAO Bid Protest Annual Report FY2024 — 16% of protests decided on the merits sustained). People, companies, suppliers and events named on this page are fictional — characters and sample data in the working solution.
Bid analysis software for sourcing teams. Eight agents read every supplier bid, rebuild it into the RFP's own cost lines over the contract life, price the exclusions, currency, payment terms and indexation, check the mandatory requirements and draft the clarification questions. The panel scores quality itself from cited evidence with prices sealed, and the procurement director signs the award.
The bid reader rebuilds each bid — a fixed fee, unit rates, a lump sum in another currency, a rate per square metre — into the RFP's cost lines, tagging every value with its page. The normaliser prices each exclusion from the bid's own rates at the RFP's baseline volumes, converts currency at the bid-close rate, brings payment terms to the standard and carries indexation over the life. Every rebuilt bid must re-add to its own total.
No. The panel scores quality. The evidence gatherer brings cited passages for each criterion and never proposes a score; the moderation checker flags spreads and rationales that cite nothing, and never averages or changes a score. Price scores follow the published formula once moderation is locked.
The challenge agent drafts a question for every gap and decides who must receive it. A question about the RFP itself goes to every bidder without naming who raised it. Nothing is sent until the sourcing manager presses Send, and the agent may not invite a change of price. Replies are read and the bid re-levelled.
Declarations of interest are collected before any bid is opened, and a panel member who declared a link to a bidder cannot open or score that bid. Prices stay sealed to the panel until moderation is locked, every view of a sealed bid is logged, and late bids are held sealed for the director to decide.
Yes. Agents never send to a supplier, disqualify a bid, accept a late bid or sign. The sourcing manager decides each deviation with a reason code and approves the shortlist, the chair records each consensus with a reason, and the procurement director signs with a reason code. Debrief letters go out only when a person sends them.
Yes. Criteria templates and weights, contract life, currency, payment standard, indexation cap, exclusion pricing, reconciliation tolerance, the abnormally-low test, moderation thresholds, roles and service levels are settings. Changes are saved as a version the director approves, and a live event keeps the version it started with.
The Agentic Solution Engine builds and deploys it from your requirements and documents — your RFP templates, sourcing policy, contract template and the bids from a past event — and it goes live once every quality gate has passed. We will walk you through it on one of your own past events first.
Every dollar classified, and every saving pinned to the invoice lines behind it.
1-min demoRFP CreationHalf the bidder questions, because every gap past bidders asked about is closed first.
1-min demoSupplier Negotiation PreparationPrice increases held to what costs support, every number in the brief cited.
1-min demoWe'll run Bid Analysis & RFP Evaluation on the bids from one of your own past sourcing events.