Marketing agency RFP template: sections, pricing table and scoring weights that make answers comparable

What a marketing agency RFP template has to do

An agency RFP (request for proposal) gets comparable answers when every agency fills the same sections in the same order, prices the work in a table you supply, and is scored against weights you publish before any response arrives. The skeleton we use:

• Eight fixed sections, each with a required format
• One pricing table, filled by every agency without changes to its rows
• A 4-week response window
• A shortlist of 5 agencies receiving the RFP

The scoring criteria and their weights are printed in the RFP itself, next to section 8, so every agency knows how it will be judged.

RFI, RFP or RFQ

An RFI (request for information) screens the field. An RFP asks a shortlist for approach, team and price. An RFQ (request for quotation) asks for price on a scope you have already fixed. Use an RFP when the approach is still open.

Project or retainer

Why public RFPs are useful evidence

We use public RFPs as evidence because public bodies publish their full terms, including scoring weights and budgets that private briefs keep internal. The decisions they force are the same ones a company with $20M to $2B in revenue faces in an agency search.

What is the difference between an RFI and an RFP

What each stage should ask

An RFI is a screen. It asks for facts an agency can supply without doing strategic work: ownership, size, current clients, conflicts, relevant capabilities and the people who would staff the account. Keep it short enough that a good agency can answer it in a day.

The RFP asks agencies to respond to your actual brief. It covers approach, named team, a draft scope of work and pricing in a common format. The pricing table and scoring weights in this piece belong here, because this is the first stage where answers need to be compared line by line.

How many agencies to carry forward

The guidelines name the stages. The counts below are an example, not a benchmark.

In this example, 15 agencies answer the RFI and 5 move to the RFP. We think the deepest cut should come between those two stages. Every agency you invite to the RFP spends real hours on a proposal, and your team spends real hours scoring it. In our reading, a long RFP list rarely improves the decision and makes the scoring step harder.

Write the cut criteria down before you read a single RFI response. If an agency fails a hard requirement, such as a client conflict or no capability in a channel you need, it stops at the RFI stage. For the wider selection sequence, see our playbook on choosing a marketing agency.

The RFP template, section by section

Copy the table into your document and replace the right-hand column with your own text.

What the source documents add to each row

How to make agency proposals comparable

Comparable proposals come from fixed answer formats. If each agency chooses its own structure, the scoring committee spends the review translating answers before it can judge them. Set the format in the RFP and say that answers outside it will not be scored.

Four formats to fix

The markup row needs a basis. In an example, a 15 percent markup on 100 of media adds 15, and an agency that quotes it on a different base will not line up with one that quotes it on net spend. Carry the agreed rows into your contract checklist.

State the budget

In our reading, a stated budget is what makes the pricing table comparable. Agencies scope to the same number, so the committee compares what each delivers for it rather than whose guess about the money was closer. If you allow bundles, ask for the bundled price and the per-part prices in the same table, so a combined offer can be checked against separate awards.

How should agency proposals be scored

Score proposals against the criteria the agencies saw in the RFP, with the weights printed next to them. The success criteria set out earlier become arithmetic here, and every agency knows the arithmetic before it writes a word.

Step 1: publish criteria and weights

Change the weights to fit your brief. Do not change them after proposals arrive.

Step 2: write a definition for each score

Use a 1 to 5 scale and give every point a written meaning, so one scorer's 4 is not another's 2.

• 1: Does not address the criterion.
• 2: Addresses it in general terms with no evidence specific to us.
• 3: Meets the requirement as written in the RFP.
• 4: Meets it with evidence drawn from comparable clients or our own data.
• 5: Meets it with evidence and a specific plan we could start on day one.

Step 3: score alone, then meet

Each named decision-maker scores every proposal independently and submits the sheet before the group meets. We think independent scoring before discussion is the single best guard against the most senior voice in the room setting everyone else's numbers. The meeting then covers only the criteria where scores diverge, and each scorer explains the evidence behind the number.

Step 4: do the arithmetic the same way for everyone

Convert each 1 to 5 score into points: divide it by 5 and multiply by the criterion's weight, so a criterion's maximum equals its weight and the total is out of 100. In this example, an agency scoring 5 on strategic response, weighted 30, earns 30 points on that line; a 3 earns 18.

Treat the total as the start of the decision, not the end. We think a close finish on points usually means the chemistry meeting and reference calls carry the call.

Step 5: use the sheets for feedback

A worked scoring example with three agencies

This example is ours, and every figure in it is an example input, not a benchmark. Three hypothetical agencies answered the same request for proposal (RFP), scored on the five criteria and weights from Step 1. Four evaluators scored them, and the panel recorded agreed points for each criterion. Each criterion's maximum equals its weight, so the points are already weighted and the five rows add to 100.

Reading the totals line by line

Agency A won price outright, earning 15 points against Agency B's 5, a 10-point edge.

That edge did not hold. On strategic response, Agency B earned 30 against Agency A's 15, a gap of 15 points. On team, Agency B earned 20 against Agency A's 10, another 10 points. Strategy and team together gave Agency B 30 plus 20, or 50, against Agency A's 15 plus 10, or 25.

A 25-point lead on strategy and team outweighed a 10-point price gap. Agency B finished at 85 and Agency A at 70, a difference of 15.

In our reading, this is the point of setting weights before proposals arrive. A board member who asks why the panel chose the more expensive agency gets a line-by-line answer, not a feeling. For how to frame that answer for the CEO and the chief financial officer, see our alignment guide.

Checking the spread between evaluators

The agreed points hide how far apart the evaluators were. Before closing the scoring, list each evaluator's points for every criterion and subtract the lowest from the highest.

Take Agency C's 20 on strategic response. The four evaluators gave 30, 25, 15 and 10, which average 20. That is a spread of 30 minus 10, or 20 points. Compare Agency B's team score, where all four evaluators gave 20 and the spread is zero.

Our working rule is to flag any criterion where the spread exceeds 10 points. A wide spread usually means the evaluators read the criterion differently, not that they disagree about the agency.

When a criterion is flagged, take these steps:

1. Ask each evaluator to state in one sentence what earned or lost points.
2. Rewrite the criterion so it names what counts. For measurement, specify whether the panel expects marketing mix modeling, incrementality testing, or platform reporting, and what a full-marks answer contains.
3. Rescore that criterion for every agency, not only the one that triggered the flag.
4. Record the revised definition in the scorecard so the next search starts from it.

How long agencies should get to respond, and the steps to run the search

Two weeks suits a small scope with a short page limit, as Atascadero had. In our reading, a multi-part scope with a pricing table needs closer to a month. Agencies have to staff the response, build the budget and pull case work. In our reading, a short window favors incumbents and firms with idle teams.

Work back from the start date

A timeline you can copy

The week numbers below are an example, not a benchmark. Adjust them to your own start date.

Before the decision date, agree the scope of work and fee structure you will sign. The winner then starts on terms it has already priced.

Spec work, chemistry meetings and other RFP traps

Spec work and who owns it

Our fix is to decide both questions before the RFP goes out and write the answer into it. If you want creative thinking, ask for a strategic response to a real brief rather than finished executions. If you do ask for spec work, state in the RFP whether you pay for it and who holds the intellectual property (IP) afterward. Carry that language into the agency contract checklist so it survives the award.

Chemistry before scoring

The traps at a glance

A feedback template

Send every agency that reached the proposal stage three lines within the window you published:

1. Your total weighted score and your rank.
2. The criterion where you scored highest.
3. The criterion that cost you the most points, with one sentence of reason.

This takes minutes per agency once the scoring sheet exists, and it closes the last trap on the list.

What this page does not cover

This page does not cover contract negotiation in detail, fee benchmarks for agency work, or how to run the transition from an outgoing agency. It is not legal advice on procurement rules, intellectual property or contract terms, so have counsel review the language before you issue an RFP.

Related on CMO Tribune: Agency contract checklist: the terms a CMO should read twice; How to choose a marketing agency, with a five-stage method and a scoring sheet you fill in yourself; Marketing board deck template: eight slides and the numbers a board expects.