RFP Evaluation Criteria: How to Score Bids (With a Worked Example)

RFP evaluation criteria are the factors a buyer uses to score and compare bids before awarding a contract, typically price, quality, service level and ease of doing business, each given a weight. Most content on RFP scoring is written for vendors trying to win. This guide is for the buyer: the person who sets the criteria, scores the responses and has to defend the award afterwards.

It covers how to choose and weight your criteria, how to build an RFP scoring matrix, a worked example scoring three bids, and the calls a spreadsheet can’t make for you: tied scores, suspiciously low bids and partial awards.

A note on timing: this guide covers evaluating bids before you award. Rating a vendor’s performance after the job is finished is a separate process, and it produces the past-performance evidence you’ll use here.

In this guide:

Matching your evaluation effort to the risk of the purchase
The core RFP evaluation criteria
How to weight them (the weighted scoring model)
An RFP scoring matrix template
A worked example: three bids, scored end to end
The RFP evaluation process, step by step
Lowest bids, tied scores, red flags and partial awards
FAQ

Start with the stakes, not the template

Before you pick criteria, decide how much evaluation the purchase needs.

I’ve spent my career buying for construction and property operations, and no two RFPs call for the same level of evaluation. A strategic, high-value scope, where poor performance would mean major delays and millions of dollars in cost, deserves a full weighted evaluation. A quick pricing RFQ for a low-risk need may not need scoring at all. You just make a quick decision based on price.

A simple rule of thumb:

Low-risk, interchangeable goods: a price comparison is enough.
Services, or anything expensive to get wrong: use weighted RFP evaluation criteria.
Strategic scopes: weighted criteria, several reviewers and a stakeholder decision meeting.

The core RFP evaluation criteria

On my most recent evaluation, my team scored four suppliers on four criteria. Together they cover most service and construction procurement:

Price. The total cost to deliver the full scope, not just the headline number.
Quality of service. How good the work will be, based on the approach in the proposal and your past experience with the vendor.
Service level. Responsiveness, availability, staffing and what’s included.
Ease of doing business. How flexible and cooperative the vendor will be once the job starts.

For larger or riskier scopes, add these:

Capacity and financial stability. Can they take on a job this size and keep paying their crew through it?
Scope completeness. Did they price the whole scope, or leave items out to look cheaper?

Score every bid on two kinds of evidence: what the proposal says, and what you already know about the vendor from past work. The second kind is why keeping a record of vendor performance matters so much.

How to define “ease of doing business” so it isn’t a gut score

Ease of doing business is the criterion most likely to become a gut feeling. I define it as how flexible the vendor is in supporting your business to reach its goals. To make that scoreable, break it into specific behaviours:

Schedule flexibility. If an earlier trade runs late, can this vendor move their start by a few days or a month?
Handling surprises in scope. If a section has to be redone after an incident on site, do they say “I’m already here, let me patch it and repaint,” or insist on extra cost and a separate visit?
Emergency response. When something happens at short notice, can they show up? For a waste hauler: when a bin fills ahead of schedule, can they send a truck right away?

Then write down what each score means, for example 5 = “moved schedule at no cost on two previous jobs” and 1 = “every change needed a new contract.” With written definitions, every reviewer scores against the same standard, and the score holds up if someone questions the award.

Weighting your criteria: the weighted scoring model

RFP evaluation criteria scoring matrix weighting price, quality, service level and ease of doing business

A weighted scoring model gives each criterion a percentage weight. The weights add up to 100%, and each vendor’s score on a criterion is multiplied by that criterion’s weight:

Weighted score = score on criterion × weight of criterion

Add up the weighted scores for each vendor to get a total you can compare directly.

There’s no universal set of weights. They should follow the risk of the purchase:

The more a failure would cost you, the less weight price should carry.
The more the work depends on the vendor adapting on site, the more weight ease of doing business should carry.
For commodity goods, price can carry most of the weight, or all of it.

Set the weights before you open the bids. Weights chosen afterwards look like they were picked to justify a vendor you already liked, and they’re harder to defend.

RFP scoring matrix template

An RFP scoring matrix (also called a bid evaluation matrix or RFP scorecard) is a grid with criteria in the rows and bidders in the columns. For each criterion, record:

Criterion: for example, ease of doing business.
Weight: for example, 20%.
What a 1 looks like: the minimum acceptable, or evidence of problems.
What a 3 looks like: meets requirements.
What a 5 looks like: clearly exceeds them, with evidence.
Evidence source: the proposal, references or past performance records.
Score per bidder, then weighted score per bidder.
Notes and flags: anything that affects confidence but doesn’t fit a number.

Keep the matrix to the criteria that actually separate bidders. Four to six is enough for most service RFPs.

Worked example: scoring three bids

This is an illustrative example, not a real tender. It shows how the numbers work. Imagine a two-year janitorial contract for a small office portfolio, with three bids.

Weights: price 30%, quality of service 25%, service level 25%, ease of doing business 20%. Scores run from 1 to 5.

Bid A: new vendor, lowest price

Price: 5 × 0.30 = 1.50
Quality: 3 × 0.25 = 0.75 (reasonable proposal, no track record)
Service level: 3 × 0.25 = 0.75
Ease of doing business: 2 × 0.20 = 0.40 (rigid change-order terms)
Total: 3.40 out of 5 (68 out of 100)

Bid B: known vendor, about 8% above A

Price: 4 × 0.30 = 1.20
Quality: 4 × 0.25 = 1.00 (consistent work on past jobs)
Service level: 4 × 0.25 = 1.00
Ease of doing business: 5 × 0.20 = 1.00 (has absorbed schedule changes at no cost before)
Total: 4.20 out of 5 (84 out of 100)

Bid C: premium vendor, about 30% above A

Price: 2 × 0.30 = 0.60
Quality: 5 × 0.25 = 1.25
Service level: 4 × 0.25 = 1.00
Ease of doing business: 4 × 0.20 = 0.80
Total: 3.65 out of 5 (73 out of 100)

Result: Bid B wins, even though it isn’t the cheapest. The award is easy to defend because the reasoning is recorded: B’s premium over A is small, and it’s backed by performance evidence that A can’t offer yet.

Two checks before you finalize:

Ask about the 30% gap. When one bid is far above the rest, find out why. Often the higher bidder included critical scope that the lowest bidder left out just to keep the price down. Make sure you’re comparing the same scope.
Question the lowest bid, too. Can Bid A actually deliver at that price? See the next sections.

How to evaluate RFP responses: the process, step by step

Decide the level of evaluation: price-only, weighted or full stakeholder review.
Choose your criteria and write down what a 1, 3 and 5 mean for each.
Set the weights before any bids arrive.
Check scope completeness so every bid covers the same work.
Have each reviewer score independently, citing their evidence.
Combine the scores. Average them across reviewers, and flag any criterion where reviewers are far apart.
Discuss the flags and the close calls with stakeholders.
Decide and record why, including any tie-breaking reason.
Award, either in full or as a partial award (see below).
Review performance at the midpoint and end of the scope, so your next evaluation has better evidence.

Is the lowest bid ever the right award?

Yes. In my experience, the cheapest bid is the right choice in two situations:

Low-risk, identical goods. If it’s the same office chair, buy the cheaper one.
A vendor you already trust happens to be lowest. If they’ve worked with you many times and know all the surprises in a project, you’re choosing them for confidence, and the price is a bonus.

The danger is an unfamiliar vendor with a very low bid. Ask whether the proposal has traps, whether they’ve left out conditions they’ll charge for later, and whether they’ll even make money on the job. I’ve seen a drywall subcontractor with a $4 million scope go under partway through a project. It looked healthy but had priced very aggressively and couldn’t keep up with payroll, so we had to step in and manage the work.

The test for paying more: if a vendor you know is slightly above the lowest bid, that’s defendable, provided your confidence rests on recorded experience. If they’re far out of line, either talk to them or accept they aren’t the right fit for this job.

When reviewers disagree, or scores tie

Averaging scores across several reviewers is a sound starting point. But an average can hide a real disagreement, and two bids can end up almost tied.

My approach is to make the final decision a group one. It shouldn’t be a one-person decision. Bring in the stakeholders, especially the superintendent or project manager who will manage the vendor day to day, and let them put their views on the table.

When totals are neck and neck, relationship and confidence break the tie. In one of my evaluations, two security services bids were nearly identical. The building’s developer had worked with one of them before and had confidence in them, and that decided it. Write the tie-breaking reason in the evaluation notes so the decision can be explained later.

What to do with a bad feeling about a bid

Sometimes a vendor scores well on paper and something still seems off. My go-to example is an aggressive proposal for a large scope from a bidder using a Gmail address and no website.

That instinct belongs in the process, but as a documented risk, not a hidden deduction:

Record it in the internal notes on the proposal and on the vendor’s record, so other reviewers take a second look.
Raise it at the evaluation meeting. Is the bidder legitimate? Are they too small for the project? Do they have the organization to deliver?
Check it against history. Some old-school firms do excellent work and simply don’t market themselves well. You only know that if you have records from past jobs.
Limit the exposure. For a first job with an unproven vendor, cap the award (I’d suggest something like $50,000 to $100,000) and raise the limit as confidence grows.

Partial awards: when to split one RFP across vendors

RFP partial award: a building portfolio split cleanly across three vendors, each responsible for its own group of buildings

A partial award splits one RFP across more than one vendor. It’s rarely covered in RFP guides, and it works well when the split is clean.

Split when responsibility divides cleanly. My team once tendered window cleaning for a portfolio of 60 buildings in Vancouver. No single vendor had the crew or equipment to do that volume in the summer season, so they split it across three companies, about 20 buildings each. It worked very well. Each vendor owned its buildings completely. If one fell behind, another could take on an extra building because their crew was already mobilized. And the team ended up with performance data on three vendors instead of one.

Don’t split a single scope. Two cleaners alternating days in one building, or one contractor doing half a job and another finishing it, will blame each other when something goes wrong. Clean scope means clean responsibility.

Don’t split without enough volume. Two buildings shared between two vendors means neither gets enough business to care about keeping it.

Midpoint reviews: carrying the evaluation past the award

Evaluation doesn’t end when the contract is signed. For longer scopes, I collect field feedback at the midpoint. For a multi-year service contract, that means every six months. At the end of the scope there’s a final review, and for work under warranty, another one later.

Vendors change during a job. Some slip near the end. Others start slowly and finish strong. Those reviews become the past-performance evidence in your next evaluation, so each RFP is easier to score than the last.

RFP evaluation criteria FAQ

What are the most common RFP evaluation criteria?

Price, quality of service, service level and ease of doing business. For larger scopes, add capacity, financial stability and scope completeness.

How do you weight RFP evaluation criteria?

Weight them by the risk of the purchase, and set the weights before opening bids. The more costly a failure would be, the less weight price should carry.

What is an RFP scoring matrix?

A grid that lists your evaluation criteria, their weights and what each score means, with each bidder’s scores and weighted totals side by side. It’s also called a bid evaluation matrix or RFP scorecard.

Should the lowest bid always win?

No. The lowest bid is right for low-risk, identical goods, or when a vendor you already trust happens to be lowest. An unfamiliar vendor far below the others needs scrutiny first.

What’s the difference between bid evaluation and vendor performance rating?

Bid evaluation happens before the award and compares proposals. Performance rating happens after the work and records how the vendor actually did. The second feeds the first.

When should you split an RFP award?

When the work divides into separate pieces with clear ownership, such as groups of buildings, and there’s enough volume for each vendor. Avoid splitting a single scope.

Score your next RFP in one place

PowerRFP puts every bid side by side, lets several reviewers score against the same criteria, and supports partial awards across multiple vendors on one project. The proposal scorecard is part of PowerRFP Pro. Start a Pro trial to run your next RFP evaluation from one place.