RFP Evaluation Criteria: How to Choose, Weight and Define Them

When a monitor asks why the second-ranked proposal lost

A grants manager sits with a monitor who has the award file open. The monitor points to a proposal that ranked second and asks a plain question: why did it lose to the top-ranked offeror? The manager has all the scoresheets, but the RFP lists only "technical approach," "experience" and "cost." Each evaluator, it turns out, read those words a little differently. One treated "experience" as years in business, another as similar projects completed, and a third as the quality of references. The totals add up, yet nobody can show what a 4 meant on the day it was given.

This is the gap that well-built RFP evaluation criteria close. Criteria are not a formality that sits in the solicitation until scoring begins. They are the measuring stick the whole file depends on, and they have to be public, weighted and defined before any proposal is opened.

After reading this article, you will be able to choose factors that fit your solicitation, weight them to total 100 and write a definition for every rating point.

What 2 CFR 200.320(b)(2) asks you to publish

When a recipient or subrecipient spends federal award money through a competitive proposal, 2 CFR 200.320(b)(2)(i) requires public notice. The notice must identify all evaluation factors and their relative importance, and proposals must be solicited from multiple qualified entities. The 2024 revision of 2 CFR Part 200 took effect October 1, 2024 (89 FR 30046, published April 22, 2024). Check your award terms to see which version applies to your award, and cite that section in your procedures rather than an older edition.

Two neighboring requirements shape how the criteria get used. Under 2 CFR 200.320(b)(2)(ii), an entity using the proposals method must have written procedures for conducting technical evaluations and making selections. The basis of the award is set out in 2 CFR 200.320(b)(2)(iii):

"Contracts must be awarded to the responsible offeror whose proposal is most advantageous to the recipient or subrecipient considering price and other factors."

Read together, these give the criteria three jobs. The notice tells offerors what will be judged and how much each factor counts. The written procedures tell your evaluators how to judge it. The award basis tells everyone what the resulting ranking is for.

The file has to show all of this afterward. Under 2 CFR 200.318(i), records must be detailed enough to show the history of each procurement. They include the rationale for the procurement method, the contract type selection, the contractor selection or rejection, and the basis for the contract price. Published criteria with stated weights are the clearest evidence of why one contractor was selected and another rejected.

Confirm the current text of 2 CFR 200.320(b)(2) at eCFR and record the effective date in your file. eCFR describes itself as authoritative but unofficial. The official edition is the annual CFR on govinfo.gov. State and local codes may be stricter than the federal rules, and where both apply your process may need to satisfy both. Ask your awarding agency or pass-through entity how a rule applies to your specific award.

How to choose RFP evaluation criteria that fit the solicitation

If you are working out how to write RFP evaluation criteria, begin with the scope of work rather than a template. A criterion earns its place when it describes something that would make one proposal better than another for this particular purchase. A generic list copied from the last solicitation tends to measure what the last project needed.

Start from what the work requires

Read the scope of work and ask what separates a strong response from an adequate one. For a services contract, the answer may involve the proposed approach, the people assigned and relevant past performance. For a software or equipment purchase, it may involve how well the offer meets the stated requirements and what support follows delivery. The answers become candidate factors.

Write each candidate as a short phrase that an offeror could respond to directly. "Staffing plan and key personnel" gives a proposer something to address. "Quality" does not. Our companion piece on RFP evaluation criteria examples shows how factors read across several kinds of purchases.

Keep each factor distinct

Overlap is the most common structural flaw. If "experience" and "past performance" both reward the same references, those references are counted twice, and the weights no longer mean what the notice says. Test each pair of factors by asking whether a single piece of evidence could earn points under both. If it could, merge the factors or narrow their definitions.

Keep the number manageable too. Every added factor spreads the available points thinner and gives evaluators more to read and more room to disagree. A small agency scoring a few proposals a year is usually better served by a short list of factors that are each clearly defined.

Leave out what the rules do not allow

Some choices are limited by the rules rather than by preference. Four points deserve a check before the criteria are final:

• Price. For competitive proposals, the award basis is the proposal most advantageous "considering price and other factors," so price is ordinarily part of the evaluation. Under 2 CFR 200.320(b)(2)(iv), the qualifications-based method in which price is not a selection factor "can only be used to procure architectural/engineering (A/E) professional services." Under that method, the entity evaluates qualifications, selects the most qualified offeror and then negotiates fair and reasonable compensation.
• Geographic preference. 2 CFR 200.319(a) requires full and open competition. The text in force before October 1, 2024 expressly barred state, local and tribal geographic preferences in evaluating bids or proposals; the 2024 revision removed that paragraph. State or local law, your award terms and the full-and-open-competition standard can still limit location-based criteria, so check them before using one.
• Conflicted evaluators. Under 2 CFR 200.318(c), no employee, officer, agent or board member with a real or apparent conflict of interest may take part in the selection, award or administration of a contract supported by the federal award. Criteria do not fix this, but your evaluator roster has to.
• School district and E-Rate purchases. Under 47 CFR 54.503, bids are carefully considered with price as the primary factor. A district applying for E-Rate support should confirm how that weighting interacts with its own state and local rules.

Qualifications-based selection, state QBS statutes and local procurement codes vary. Check your state's law and your own written procedures before you finalize the factor list.

RFP evaluation criteria weighting: setting relative importance

Weighting turns a list of factors into a decision rule. The notice must state each factor's relative importance under 2 CFR 200.320(b)(2)(i), and weights are the plainest way to do that. Many evaluation files express them as points that total 100, so each factor's weight is simply its share of the whole.

This article does not recommend particular weights, and no weight is correct for every purchase. A weight records a judgment about what matters in this procurement, made by people who understand the work. The mechanism is what you can teach, and the judgment stays with your organization.

What the mechanism requires

Three checks make weights hold up when someone reads the file later:

1. The weights total 100. A set that sums to 98 or 103 usually means someone edited one factor and forgot the others. In a spreadsheet, nothing forces the total, so make it a visible check cell that you confirm before release.
2. The weights match what the RFP published. The weights in the scoring tool must be the same ones offerors saw. If the RFP is amended, record the amendment and the date, and update the scoring sheet to match.
3. The weights are fixed before proposals are opened. Adjusting a weight after seeing who submitted what invites the accusation that the weights were tuned to favor someone. Even without any intent, the file can no longer show otherwise.

How price enters the ranking is part of weighting. If price is a scored factor, the method for converting price into points belongs in the solicitation too, since offerors are entitled to know how it will be treated. Our guide to weighted scoring for RFPs walks through the arithmetic, including the standard lowest-price formula.

Weighting in practice

A useful discipline is to explain each weight in a sentence while drafting. If you cannot say why a factor carries its weight, the committee probably has not agreed on it yet. Write the explanation into the planning record. It does not need to be published, but it helps show later that the weights followed from the purpose of the procurement.

Weights are also the place where a committee's disagreements surface early, which is when they are cheapest to settle. A debate over whether technical approach should count for more than staffing is far easier to have before proposals arrive than after.

Building an RFP scoring rubric: define every rating point

A weight says how much a factor counts. A rating definition says what each score on that factor means. Without definitions, two evaluators can give the same proposal a 3 and a 5, and both can be sincere, because the scale lives only in their heads.

An RFP scoring rubric solves this by anchoring each rating point. An anchor is a short written description of what a proposal must show to earn that point. Every point on the scale gets one, written before scoring opens and used by every evaluator.

A sample anchor set

The table below is an illustration for one hypothetical factor, "Staffing plan and key personnel," on a 0–4 scale. It is not a recommended scale.

Rating	Definition
0	No staffing plan is provided, or the plan does not address the scope of work.
1	Roles are named, but qualifications are missing for most key positions.
2	Qualifications are stated for key positions, but availability or time commitment is unclear.
3	Qualified staff are identified for all key positions, with stated time commitments.
4	All of the above, plus a described plan for coverage if a key person becomes unavailable.

Each definition describes evidence that is present in the proposal, not an impression. An evaluator can point to the page that supports a 3 or explain what is missing for a 4. Definitions that rely on words like "good" or "excellent" push the judgment back into private interpretation, which is what the rubric is meant to prevent. Our piece on RFP scoring scales and rating definitions covers how to write them for different scale lengths.

A worked example

Here is how a rating becomes a weighted score. Assume the factor carries a weight of 20 points, chosen only to keep the arithmetic simple, and an evaluator gives it a rating of 3 on the 0–4 scale.

Weighted score = rating ÷ maximum rating × weight = 3 ÷ 4 × 20 = 15.

A rating of 4 would earn the full 20, and a rating of 2 would earn 10. The same calculation runs for every factor, and the weighted scores sum to a total out of 100. Record the formula in the file so that someone rebuilding the totals later gets the same result.

Why the definitions matter in a dispute

Evaluation records are where challenges tend to land. The GAO Bid Protest Annual Report for fiscal year 2025 (GAO-26-900695, December 12, 2025) lists the top three grounds on which protests were sustained: unreasonable technical evaluation, unreasonable cost or price evaluation, and unreasonable rejection of a proposal. These are federal agency procurements, not grant-recipient or local ones. They still show where evaluation reasoning gets examined.

Under 2 CFR 200.318(k), the recipient or subrecipient settles source evaluation, protests, disputes and claims arising from its own procurements. Your own record is therefore the main evidence you have. Anchored ratings let you show what a score meant, not just what it was.

Independence and consensus

Anchors work best alongside independent scoring. Each evaluator should rate every factor on their own before anyone discusses results, so that one voice does not set the others' scores. Consensus comes afterward, and the rationale for each final score should be written down. Our RFP evaluation process guide covers that sequence from criteria through award.

What the file should record before proposals open

The criteria are only as useful as the evidence that they existed in final form before scoring began. A short pre-scoring record does that work. At minimum, the file should show the following:

• The final factor list, worded exactly as published in the RFP, with the date of release and any amendments.
• The weights, totaling 100, and how price is converted to points if it is scored.
• The rating scale and an anchor for every point, written before the first proposal was opened.
• The tie rule, chosen in advance, so that a tie is resolved by a rule and not by a judgment made after the totals are known.
• The evaluator roster and roles, along with each person's conflict-of-interest attestation under 2 CFR 200.318(c), completed before scoring starts.
• The rule version you followed and its effective date. Thresholds and procedures change. The Uniform Guidance was substantially revised effective October 1, 2024, and a file that does not record the rule and its date cannot show which rule it followed at the time.

That last point is not hypothetical. On May 29, 2026, OMB published a proposed rule, "Regulation for Federal Financial Assistance" (91 FR 32198), that would revise 2 CFR Part 200 again. As of October 2026, the Federal Register shows no final rule. A proposed rule is not a final one, so check federalregister.gov and confirm the current status with your awarding agency or pass-through entity before changing anything.

Retention matters as well. Under 2 CFR 200.334, records are generally kept for three years from the final financial report, unless an exception applies, such as a written notice extending the period. Local public-records rules may require longer, so check yours.

Tools can help with this record, and so can a well-built spreadsheet with a locked criteria tab. RFPManagement is designed to hold a weighted criteria builder in which weights must total 100 and every rating point is anchored before a round opens. All scoring in it is done by people, with no AI scoring or score suggestions, which keeps the record a human one. Whichever tool you use, the point is that nothing about the criteria should be loose when the first proposal comes in.

Putting the criteria in place

Before you release your next RFP, take the factor list through a short sequence. Check that each factor comes from the scope of work and does not overlap with another. Confirm that the weights total 100 and match the published notice. Write an anchor for every rating point, set a tie rule, and file the whole set with a date before proposals are opened.

The RFP Evaluation Criteria Builder is a planned workbook designed to record and check exactly this: the factor list, weights that must total 100, and an anchor for each rating point, along with a starter library of criteria to adapt. See its store page for what it is designed to do. To hear when the evaluation workspace and the workbooks open, join the waitlist.

RFPManagement records and computes. It does not give legal or procurement advice. The award decision stays with your organization.