Setup
Client, reference, one time zone for every date, submission instructions and a pricing schedule.
Reqify Tenders · for consultancies advising buyers
Run your client’s
vendor selection.
Turn the project’s BRD requirements into an RFP, issue it with a response workbook, and let Reqify check every vendor’s claims against the vendor’s own documents — with the quotes. The AI suggests; your evaluation panel decides, and every step is on the record.
There’s no sample tender yet. The sample bid shows the same project, requirement quoting and source checks that Tenders builds on.
RFP from the BRD, wording keptClaims checked against vendor documentsBlind scoring, people decide
Built for implementation partners and advisers running a client’s RFP, RFQ or RFI.
Vendor responses · illustrative data
T-014 · Functional requirements
The solution shall provide IVR menus and prompts in English and French.
MandatoryVendor B’s workbook: Comply
AI check · vendor’s own documents
Contradicted
“IVR prompts are currently offered in English and Spanish.”Vendor B proposal · p. 14 · quote checked word for word
Review flag · needs review
The flag doesn’t fail the vendor on its own. An editor confirms or dismisses it, with a reason that goes in the report.
Confirm failure Dismiss
How it works
From the BRD
to an award you can defend.
A tender runs inside the client’s delivery project, on its Procurement tab. Until the tender is issued, a step guide shows what is done and what comes next.
Criteria
Rated and pass / fail criteria with weights, a price method, evaluators and blind scoring.
Requirements
Imported from the project’s BRD and FRD, numbered T-001 onwards, wording kept.
Draft the RFP
Background and scope written from the project’s documents; the rest comes from the tender itself.
Issue
Checks that weights total 100%, the deadline is set and no SME ask is open — then the files are fixed.
Vendors respond
Their completed workbooks are read back by T-number: the claim, the response and the price.
Assess
Each claim checked against the vendor’s own documents, with verified quotes.
Score
Evaluators score blind; editors agree a consensus beside the AI’s suggestions.
Award
Choose a ranked vendor. Anyone but the top-ranked one needs a written rationale.
Report
An evaluation report composed from the recorded evaluation, ready for the client.
Write the RFP
The RFP starts from requirements you already have.
Import the business, functional and non-functional requirements from the project’s BRD and FRD. Each keeps its own wording and its project reference under its T-number, and priorities map to Mandatory or Rated.
- Reqify drafts the RFP document: only the background and scope are AI-written, from the project’s BRD, FRD, charter and SOW.
- Dates, instructions, criteria, requirements and pricing come straight from the tender — never re-worded.
- Terms are quoted word for word from approved records in your library. Reqify never writes terms, a budget or a figure your documents don’t state.
- The Word RFP, Excel response workbook and addenda are issued in the buyer’s name, not your firm’s.
Requirements · illustrative data
Project BRDFR-012
The system shall route calls by caller language and agent skill.
Tender requirementT-012 · Mandatory
Same wording, never rewritten. Must priority becomes Mandatory.
Response workbookCompliance ▾
Comply, Partial or Does not comply, with the vendor’s response and page reference.
Evaluate
The AI reads the fine print. The panel decides.
Assess with AI checks each vendor’s response against the vendor’s own proposal, pricing and other files. The completed workbook is the vendor’s claim, never its evidence.
- Verdicts per requirement: Supported, Partly supported, Unsupported claim, Contradicted or Not addressed — each with quotes checked word for word.
- A suggested score per criterion with strengths, weaknesses and questions to put to the vendor, labelled as a suggestion.
- Blind scoring on by default; consensus shows each evaluator’s score, the spread and the AI’s suggestion side by side.
- Mandatory failures can be waived only with a written reason, which appears in the report.
Consensus · Vendor B · illustrative data
| Criterion · 0–5 | Weight | Evaluators | Spread | AI suggestion | Consensus |
|---|---|---|---|---|---|
| Functional fit | 40% | 4 · 3 · 4 | — | 3 | 3.5 |
| Technical and security | 20% | 3 · 3 · 4 | — | 3 | 3 |
| Implementation approach | 20% | 2 · 4 · 4 | Flagged | 3 | 3.5 |
| Experience and references | 20% | 4 · 4 · 3 | — | 4 | 4 |
Built for a defensible procurement
Fair to vendors.
Clear to the client.
Your client will be asked how the vendor was chosen. Tenders keeps the answer in one place.
Blind scoring
Nobody sees another evaluator’s draft. Evaluators see the others’ submitted scores, the AI’s suggestions and the results only after submitting their own; revealing early is logged.
Vendor files kept apart
Vendors’ documents never enter the project’s knowledge base, and their responses are shown only to the tender’s editors and evaluators.
An append-only record
Submissions, reopened scores, consensus changes, waivers, the shortlist and the award are written to the security audit log, with who did it and when.
One project, both sides
Win the work.
Then run the tender.
Tenders run in delivery projects. If the project started as a bid in Reqify, the tender is pre-filled with the bid’s client, and the RFP is written from the BRD the project already produced.
Bid wonStart delivery in the same project
The RFP pack, your proposal and the clarifications stay in the project’s knowledge.
RequirementsBRD and FRD with their sources
Generated from the project, with the buyer’s requirements carried in as R-### numbers when the project came from a bid.
Procurement tabA tender from those requirements
Imported as T-001 onwards, issued, assessed, scored and awarded — then the evaluation report.
What Reqify won’t do
The AI suggests.
People decide.
A procurement has to stand up to questions from vendors and the client’s own governance. So some things are left to people by design.
Score or rank vendors
Results are computed only from the scores your panel agrees. The AI’s suggested scores are shown beside them and never used as a score.
Fail a vendor on its own
An AI flag on a mandatory requirement waits for an editor to confirm or dismiss it, with a reason. The award waits for every flag.
Contact vendors
There is no vendor portal and no email to vendors. You send the RFP, workbook and addenda yourself.
Write terms or a budget
Terms are quoted only from your approved library records; figures the documents don’t state become questions for your team.
Mix vendor files into the project
Ask Project and document generation never read a vendor’s files, and the RFP is never drafted from them.
Reopen an award
An awarded tender locks its scores, consensus and waivers and stays on record. It can’t be reopened or cancelled.
Plans and costs
Free to set up.
A document per AI run.
Tenders uses your plan’s document allowance. Everything people do — setup, scoring, consensus, results — uses none.
| Professional | Team | Enterprise | |
|---|---|---|---|
| Active tenders at a time | 1 | 5 per workspace | Unlimited |
| Draft or regenerate the RFP | 1 document | 1 document | 1 document |
| Assess one vendor with AI | 1 document | 1 document | 1 document |
| Write the evaluation report | 1 document | 1 document | 1 document |
Free. Setup, importing requirements, the workbooks, vendors’ questions, addenda, scoring, consensus and results.
What counts as active. Draft, issued and evaluating tenders. Awarded and cancelled ones don’t count.
Not on Starter or the free trial. A failed AI run is refunded automatically. Up to 30 AI runs per tender a day.
The practical details
Questions teams ask first
Who is Tenders for?
Consultancies that help a client choose a product or supplier — the platform you will implement, say. Tenders runs from the client’s delivery project in Reqify, so the RFP is written from requirements the project already has.
Does the AI score the vendors?
No. For each requirement it checks the vendor’s claim against the vendor’s own documents and gives a verdict with the quotes behind it, and it suggests a score per criterion with its reasons. Evaluators score every criterion themselves, the panel agrees a consensus, and results are computed from the agreed scores only. The AI’s suggestions sit beside them for reference.
Can the AI fail a vendor?
Not by itself. When it marks a mandatory requirement Contradicted, Not addressed or Unsupported claim, that becomes a review flag. An editor confirms or dismisses each one with a reason, the decision is recorded, and the award waits until every flag is decided.
Does Reqify send the RFP to vendors?
No. Reqify doesn’t email vendors or host a vendor portal. You download the RFP (Word), the response workbook (Excel) and each addendum — all issued in the buyer’s name — and send them yourself. Vendors return the workbook, and Reqify reads their answers back row by row.
Will the RFP give away internal information?
It is built not to. Only the background and scope are AI-written, from the project’s BRD, FRD, charter and Statement of Work. A drafted paragraph that mentions money, a budget, internal risks, what stakeholders think or anyone by name is held back as a question for your team. The vendor-facing text carries no citations or internal document names, and terms are quoted only from your own approved library records.
Do vendors’ files end up in the project’s knowledge?
No. A vendor’s files never go into the project’s knowledge base — Ask Project and document generation never read them — and a vendor’s responses and documents are shown only to the tender’s editors and evaluators.
What does a tender cost?
Setting up, importing requirements, the workbooks, questions, addenda, scoring, consensus and results use no documents. Each AI run uses 1 document: drafting the RFP, assessing one vendor, or writing the evaluation report. A failed run is refunded automatically. Professional includes 1 active tender, Team 5 per workspace and Enterprise unlimited; Tenders isn’t on Starter or the free trial.
Can the decision be defended later?
That is the point of the record. The evaluation starting, each evaluator’s submission, reopened scores, revealed blind scores, consensus changes, waivers, the shortlist and the award — including whether it went to the top-ranked vendor — are written to Reqify’s append-only security audit log. The evaluation report is composed from that recorded evaluation.
Start from the evidence
Run the next selection
on the record.
See how Reqify quotes and checks sources in the sample bid, then talk to us about running your client’s next vendor selection as a pilot.