RFP Scoring Matrix and Evaluation Criteria: How to Score Proposals

RFP evaluation criteria are the factors you score proposals against, such as functional fit, security and cost. A scoring matrix, sometimes called a vendor evaluation matrix, lists them with a weight for each, so every vendor is judged the same way. This guide shows how to build and use one for software and IT services, filling in the evaluation steps of the RFP process guide.

Common evaluation criteria for software and IT services RFPs

Tie each criterion to a section of your RFP, so every question feeds one.

You may not need a separate RFP evaluation template: each template in RFPhub’s RFP template library has an evaluation section to adapt, and updated ones, such as the ERP RFP template, include a weighted scoring rubric. Section 7 of How to Write an RFP shows how to state criteria and weights in the RFP.

Pass/fail compliance checks vs weighted criteria

Pass/fail checks decide whether a proposal is scored at all. Check that each proposal:

• Arrived on time and complete, with the pricing sheet and any required forms.
• Meets every must-have requirement.
• Accepts your key terms, or lists exceptions the way the RFP asked.

Set aside any that fail, under the rules you published. Decide in advance whether a vendor may fix a minor slip, such as a missing signature; public-sector buyers should check their rules.

Everything else is weighted and scored: the must-haves get a proposal through the gate, and points come from how well it meets the rest. Keep the must-have list short, so a strong vendor isn’t knocked out over a minor gap.

How to set weights before proposals arrive

Agree the weights before the RFP goes out, and write them into it. Weights chosen after reading proposals tend to drift toward the team’s favorite.

Put the most weight on what would sink the project if you got it wrong. As an example only, the RFP process guide splits 100 points as functional fit 35, technical and security 20, implementation and support 15, vendor strength and references 10, and cost 20. A purchase with difficult integrations might move points from functional fit to technical.

• Fold criteria worth a point or two into larger ones as subcriteria; on their own they barely move the total.
• Test the weights on a made-up case: if one vendor is clearly stronger on fit and another is cheaper, which should win? If the matrix disagrees, fix the weights before issuing the RFP.

A defined 1-to-5 scoring scale

Define every score, so a 4 means the same to every evaluator.

Agree how response codes map to scores, for example that a roadmap answer can’t score above 3.

Example RFP scoring matrix

Using the example weights, each evaluator scores their assigned criteria from 1 to 5, then works out:

Weighted score = score × weight / 100

A vendor’s total is the sum of its weighted scores. A score of 4 on functional fit, for instance, adds 4 × 35 / 100 = 1.40. Because the weights total 100, a perfect total is 5.00.

Copied into a spreadsheet, the table works as an RFP scoring template; lock the weights and note the reason beside every score. Teams outgrowing spreadsheets can compare tools in the RFP management software guide.

How to score price

Score total cost over the contract term, from the pricing sheet. Check first that every vendor priced the same scope; if one left out an item, such as data migration, ask it to price the item in a clarification. Ask what any unusually low price assumes.

Then convert price to a cost score, by a method chosen in advance. Common options include:

• Ratio to the lowest price. Cost score = 5 × lowest price / this vendor’s price. A price twice the lowest gets half marks, and no vendor scores zero on cost.
• Straight-line deduction. The lowest price gets 5, and scores fall in proportion to the gap above it, reaching 1 at a level you set, such as [x]% above the lowest.
• Price bands. Set price ranges in advance, relative to the lowest price or your budget, and give each a score from 1 to 5. Bands are easy to explain, but prices either side of a band edge can score a full point apart.

Some buyers don’t turn price into points: they score everything else, then judge whether a higher score justifies the extra cost, and record why. If price alone will decide between bids that meet a fixed specification, use a request for quotation or an invitation to bid instead.

Independent scoring and the consensus meeting

Each evaluator scores only the criteria they’re qualified to judge, in their own copy, without discussing proposals. Beside each score they note the reason and where they found it, such as a requirement ID.

The project lead combines the sheets and flags criteria where scores differ widely, for example by two points or more. At the consensus meeting:

1. Take each flagged criterion, with each evaluator explaining their score.
2. Change a score only for a recorded reason, such as an answer someone missed.
3. Agree a final score per criterion, or average once the gaps are explained, as decided in advance.
4. Agree clarification questions and the two or three finalists.

How demos and references feed the final scores

Proposal scores rest on what vendors claim; scripted demos, a proof of concept and reference calls test those claims. Decide before issuing the RFP whether they revise existing scores (demos update functional fit and technical; references update implementation and vendor strength) or demos get their own weighted criterion.

• Score every demo against the same script and scale, recording why any score changed.
• If a demo shows a must-have isn’t met, such as a “standard” feature that needs custom code, apply your pass/fail rules.
• Judge a proof of concept against success criteria agreed before it started.
• Ask every reference the same questions. Vendors usually choose their references, so use the calls to probe specific risks.

Finish with final scores and a short decision memo; you’ll need both for debriefs.

Common scoring mistakes

• Changing criteria or weights after proposals arrive. It looks like favoritism, even when it isn’t.
• Scoring the writing. A polished proposal with generic answers can outscore a plain one with specific evidence.
• Letting price color the functional scores. Have a separate evaluator score cost.
• Averaging away disagreements. Talk large gaps through; one evaluator may have missed something.

Public-sector evaluation rules

Government agencies, and organizations spending public or grant money, usually have to follow evaluation rules. These vary by country, state, agency and sometimes purchase value, so check with your procurement office or legal counsel before relying on any point here. You may have to publish the criteria and their relative importance, then evaluate only against them. For US federal negotiated acquisitions:

• FAR 15.304(d) says “All factors and significant subfactors that will affect contract award and their relative importance shall be stated clearly in the solicitation.” It adds that the rating method need not be disclosed.
• FAR 15.305(a) requires agencies to evaluate competitive proposals and assess their relative qualities “solely on the factors and subfactors specified in the solicitation,” and allows “any rating method or combination of methods, including color or adjectival ratings, numerical weights, and ordinal rankings.”

FAR references are to the text on acquisition.gov as of October 2026. The FAR is being rewritten under the Revolutionary FAR Overhaul, and many agencies, including DoD, GSA and HHS, have adopted a deviation based on the overhauled Part 15. Under it, the requirement to state all factors and significant subfactors and their relative importance is at FAR 15.104(a), and evaluating proposals solely on those factors is at 15.202(a). The overhauled text doesn’t repeat the older wording on rating methods quoted above. Check which version a solicitation applies before relying on a section number. This isn’t legal advice.

Frequently asked questions

What are the evaluation criteria in an RFP?

They’re the factors a buyer scores proposals against, each weighted by how much it counts. For software and IT services, common ones are functional fit, technical and security, implementation and support, vendor strength and references, and cost. Must-haves are usually checked pass/fail first.

How do you create an RFP scoring matrix?

List the criteria as rows with weights that add up to 100, and add a column for each vendor. Define each score on a 1-to-5 scale, and have evaluators score independently before a consensus meeting. Set it all up before proposals arrive.

How do you calculate a weighted score?

Multiply each score by its criterion’s weight, divide by 100, and add up the results for each vendor. Because the weights total 100, the result stays on the 1-to-5 scale.

How do you score price in an RFP?

Score total cost over the contract term, on the same scope for every vendor. Common methods include a ratio to the lowest price, a straight-line deduction above the lowest, and price bands mapped to scores. Choose one before proposals arrive.

Is a vendor evaluation matrix the same as a vendor scorecard?

No. A vendor evaluation matrix compares bidders once, before a contract, on what each proposes. A vendor scorecard rates one supplier’s performance after the contract is signed, which is what RFPhub’s vendor scorecard template is for.