RFI, RFP, RFQ — three acronyms that get mixed up in purchasing offices every day. They look interchangeable, and they are not. Choosing the wrong one wastes weeks, confuses good suppliers into walking away, and leaves you with a pile of offers you cannot compare to each other.
The good news is that picking the right instrument comes down to a single question, and once you ask it the choice is almost automatic. This guide explains what each tool is for, the mistakes that quietly sabotage tenders, how to sequence the three when a category is complex, and how to build an evaluation that produces a decision you can defend.
The one question that decides everything
Every tender starts from what you already know — and what you do not. Two dimensions settle the choice:
- Do you understand the supply market? Do you know who the credible suppliers are and what is technically possible?
- Do you know the solution? Have you specified exactly what you need, down to the drawing, part number, or service level?
Map your answers and the instrument selects itself:
| Specification known | Specification unknown | |
|---|---|---|
| Market known | RFQ | RFP |
| Market unknown | RFI, then RFQ | RFI, then RFP |
Everything below is the detail behind this matrix.
RFI — Request for Information
An RFI is a scouting tool. You use it when you do not yet understand the market well enough to run a real competition: who the players are, what technologies exist, what is feasible, what is standard practice.
What it contains: open questions about capabilities, references, certifications, capacity, and approach. No detailed pricing request.
What it produces: a shortlist and a map of the market — not a winner. The output of an RFI is the input to an RFP or RFQ.
The most common RFI mistake: asking for prices. Without a defined scope, suppliers either refuse to quote or give numbers so padded they are meaningless. You then anchor on a fictional figure. Keep price out of the RFI; that is not its job.
RFIs are also legitimate market intelligence even when you are not about to buy — a way to keep a category map current before the contract comes up for renewal.
RFP — Request for Proposal
An RFP is for problems where you know the outcome you need but not the best way to achieve it. You describe the requirements, the constraints, and the context; suppliers propose their solution, approach, and commercial terms.
RFPs fit services, software, logistics contracts, consulting, marketing — anything where supplier creativity and method genuinely add value and two good suppliers might solve the same problem differently.
What it contains: business requirements, mandatory and desirable criteria, constraints, the evaluation method, and the response format you require (so offers arrive comparable).
The critical discipline: the price in an RFP is one weighted factor, not a line-by-line comparison. Because each supplier proposes a different solution, you cannot simply rank by cost. You need a weighted evaluation matrix, agreed before offers arrive — never built afterwards, when it is too easy (consciously or not) to shape the weights around the answer you already prefer.
A worked evaluation matrix
Suppose you weight five criteria and score three suppliers from 1 to 10:
| Criterion | Weight | Supplier A | Supplier B | Supplier C |
|---|---|---|---|---|
| Solution / technical fit | 35% | 9 | 7 | 8 |
| Price / total cost | 30% | 6 | 9 | 7 |
| Implementation & timeline | 15% | 8 | 7 | 6 |
| Support & SLA | 10% | 7 | 6 | 9 |
| Sustainability / ESG | 10% | 8 | 6 | 7 |
| Weighted score | 7.65 | 7.40 | 7.40 |
Supplier B is the cheapest (price score 9) — and does not win. Supplier A takes it on weighted value (7.65), because superior technical fit, weighted at 35%, outweighs B’s price advantage. This is the entire point of an RFP: the best value wins, not the lowest price, and the weights — fixed in advance — make the result objective and defensible.
RFQ — Request for Quotation
An RFQ is for fully specified purchases. You know exactly what you need — the drawing, the part number, the service level, the quantity — and the main open question is price.
Because every supplier quotes the same defined thing, the offers are directly comparable, the evaluation is largely arithmetic, and the process can be fast — or run as an electronic reverse auction when the market is competitive enough.
The most common RFQ mistake: running it on a poorly specified scope. If the specification is loose, you will receive cheap offers for the wrong thing, then spend the savings — and more — on rework, change orders, and disputes. The rule is simple: no specification, no RFQ. Tighten the spec first, or step back to an RFP.
When the spend is significant and the market competitive, the RFQ is also where e-sourcing earns its keep — structured online quoting and reverse auctions compress the process and surface the real market price. (This is the focus of e-Sourcing and e-Auctions: Tools and Tactics.)
The three tools side by side
| Dimension | RFI | RFP | RFQ |
|---|---|---|---|
| You know the market? | No | Yes | Yes |
| You know the solution? | No | Goal yes, method no | Yes, fully specified |
| Main question | Who and what exists? | What is the best solution? | What is the price? |
| Pricing requested? | No | Yes, within the proposal | Yes — the focus |
| Supplier effort | Low | High | Low to medium |
| Evaluation basis | Qualitative shortlist | Weighted matrix | Largely arithmetic |
| Typical output | Shortlist + market map | Selected solution & partner | Best price, award |
| Speed | Fast | Slow | Fast (can be an auction) |
| Best for | New or unknown categories | Services, software, complex scope | Defined goods, commodities |
Sequencing them: the funnel
The three are not mutually exclusive — for a complex, unfamiliar category the strongest approach is a funnel:
- RFI to understand the market and build a shortlist.
- RFP to select the best solution and partner from that shortlist.
- RFQ (or a final negotiation) to sharpen the price once the solution is fixed.
Not every category needs all three. A commodity you buy regularly goes straight to RFQ. A first-of-its-kind service might need the full funnel. What matters is that you tell suppliers where they are in the process — an RFI is not a commitment to buy, and pretending otherwise burns goodwill you will want later.
Building an evaluation that holds up
Whichever instrument you choose, the evaluation makes or breaks it.
- Agree criteria and weights before offers open. This is non-negotiable. Weights set afterwards invite bias and, in regulated or public tenders, legal challenge.
- Separate technical and commercial scoring. Evaluate the solution before you see the price, so a low number does not create a halo effect on the technical judgement.
- Keep the criteria few and meaningful. A matrix with fifteen criteria at 2% each dilutes the decision into noise. Five to seven weighted criteria force genuine priorities.
- Score against a defined scale. Define what a “7” means versus a “9”, so different evaluators score consistently.
- Document everything. The criteria, the scores, the rationale, the Q&A. In regulated and public procurement this is mandatory; everywhere else it protects you and informs the next tender.
Running the process fairly
A clean process protects the result and your reputation with the supply base:
- One point of contact for all supplier questions.
- Shared Q&A: an answer given to one bidder is published to all. A level playing field is both ethical and practical — it keeps your best suppliers engaged.
- Firm, communicated deadlines, with the same window for everyone.
- Professional debriefs for unsuccessful bidders. The supplier you reject today is one you may need next year; how you decline matters.
Quick decision rule
- Do you need to learn the market? → RFI
- Do you know the goal but not the solution? → RFP
- Do you know exactly what you are buying? → RFQ
Common mistakes to avoid
- Asking for prices in an RFI. It is not a pricing exercise; you will get padded or no numbers.
- Running an RFQ on a vague spec. Cheap offers for the wrong thing, paid back in rework.
- Building the evaluation matrix after offers arrive. It invites bias and challenge.
- Mixing technical and commercial scoring. Price contaminates the technical judgement.
- Too many criteria. Fifteen trivial weights dilute the decision; choose five to seven that matter.
- Hiding the sequence from suppliers. They should know whether this is an RFI, a shortlisting RFP, or a binding RFQ.
- Uneven Q&A. Sharing an answer with one bidder and not the others compromises the whole tender.
Key takeaways
- The choice comes down to two questions: do you know the market, and do you know the solution?
- RFI scouts the market; RFP selects the best solution; RFQ sharpens the price on a defined scope.
- In an RFP, the cheapest supplier should not automatically win — a weighted matrix, fixed in advance, picks the best value.
- For complex categories, sequence RFI → RFP → RFQ, and tell suppliers where they stand.
- A fair, documented process is not bureaucracy — it produces better offers and protects the result.
