How to tell whether a marketing proposal will actually work

By Jimmy Dodgson, Client Services Director, Maitland

Published 2026-08-20

Six checks to run on any marketing proposal before you sign it, including the arithmetic most agencies never show you and the tell that should worry you most.

Executive summary

Judge a marketing proposal on arithmetic before ideas: divide the revenue target by customer value to get the customers required, then divide the budget by that to see the cost per customer the plan assumes. Compare it with what a customer costs you today. Then run six checks: the maths, the timescale against the channels, whether it answers the brief or the problem, whether there is a baseline and a named number, what is excluded, and who does the work. The tell that should worry you most is the proposal that agrees with everything.

You have three marketing proposals on your desk and they all sound convincing. Which of them will actually produce the number you need, in the time you need it?

I have written this because we lost a tender this month by refusing to answer the brief as written, and I would rather show you the checks we ran than tell you we are honest.

Here is what this covers: the arithmetic that decides whether a plan is possible before anything else matters, the six checks to run on any proposal, the tell that should worry you more than any red flag, what to do when the answer is that the plan cannot work, and why we still think we did the right thing.

Start with the arithmetic, not the ideas

Most proposals are judged on their ideas. That is the wrong order, because ideas are cheap and arithmetic is not negotiable.

Before you read a single tactic, do this sum. Take the revenue target in the brief. Divide it by your average order value or contract value, which gives you the number of new customers the plan must produce. Now divide the proposed budget by that number of customers. What you are left with is the cost per customer the plan is quietly assuming.

Then ask one question. Is that figure anywhere near what a customer actually costs you to acquire today?

If the plan assumes £80 and your current reality is £300, the proposal is not ambitious. It is arithmetically impossible, and no amount of creative thinking closes a gap of that size. Everything else in the document is decoration on top of a sum that does not work.

This is the check we ran on the tender we lost, and it took about 10 minutes.

The six checks

Run these on every proposal you receive, in this order.

One: does the maths stack up? The sum above. Cost per customer implied by the plan, against cost per customer in your business today. A stretch of 20 or 30 per cent is ambition. A gap of three or four times is fantasy.

Two: is the timescale honest for the channels proposed? Paid search and paid social produce signal in 30 to 60 days. SEO and content compound over six to 12 months. A proposal promising a search-led return inside a quarter has either misunderstood the channel or is telling you what you wanted to hear.

Three: does it answer the brief, or the problem? These are frequently different. A brief is written by somebody inside the business, working from assumptions that may themselves be the reason growth has stalled. A proposal that follows every instruction faithfully may simply be an expensive way to repeat an existing mistake.

Four: is there a baseline and a named number? The plan should state where you are starting from and what it expects to produce, by when. Without a baseline, nobody can be held to anything, which is a topic we have covered in how to check an agency's results .

Five: what is excluded? Ad spend, tools, production, third-party costs. A proposal that presents a total without naming exclusions is understating the true figure, and you will discover the difference in month two.

Six: who does the work, and what happens when they are away? Named people, not a capability slide. The team on the pitch is not always the team on the account, and it is a fair question to ask directly.

A seventh check is worth adding if the proposal runs to any length. Read the first page and the last page together, and see whether they describe the same job. Long documents drift. A proposal that opens on your revenue target and closes on a schedule of deliverables has quietly changed the subject somewhere in the middle, and the deliverables are what will end up in the contract.

If you only have 20 minutes, spend them on checks one, two and five. Those three catch most of the proposals that were never going to work, and they can all be answered from the document in front of you without a meeting.

The tell that should worry you most

Here is the one that catches people, because it does not feel like a warning at all.

The proposal that agrees with everything.

If you wrote a brief and every agency came back saying it was an excellent brief and here is exactly how they would deliver it, be careful. You have either written a genuinely flawless brief, which is rare, or you have found several suppliers who would rather win the work than tell you something awkward.

The most useful proposal you receive is usually the one that disagrees with part of your brief and explains why, with numbers. That agency has done the arithmetic, has reached an uncomfortable conclusion, and has decided you are worth telling. It costs them the easy path to the sale. It is also the only signal you will get, before signing, about how they will behave when something goes wrong later.

What we did when the numbers did not work

We were invited to tender recently. The brief was thorough, the business was one we wanted, and the internal appetite to win it was high.

Then we ran the sum. The budget and the plan as specified would not produce a return on ad spend that justified the investment, and certainly not inside the timescale the brief set out. We went back over it looking for the reading that made it work, because we wanted the work. There was not one.

So we did not answer the brief. We proposed something materially different: a different sequence, a different allocation, and a different first phase aimed at fixing what we thought was actually holding the business back.

It got us through the door and into the final rounds. We did not win it.

I am not going to dress that up as a moral victory, because we would rather have won. But the team are proud of how we behaved, and I would make the same call again tomorrow. Telling a prospect their plan will not work is the same behaviour they would want from us in month eight of a partnership, when the awkward news is about our own performance rather than their brief. An agency that will not be honest before the contract will not suddenly find the nerve afterwards.

That is also our answer to a question we get asked constantly, which is what we will not do and why .

What to do when a proposal cannot work

If the arithmetic fails, you have three options and only one of them is bad.

Raise the budget to match the target, which is the honest response when the target is genuinely fixed. Our guide to what to spend on marketing sets out the UK bandings and a method for working the number backwards from your goal.

Or lower the target to match the budget, and set a realistic timeline for reaching the original ambition over two years rather than one.

Or change the plan, which is what we proposed. Often the constraint is not the budget at all. It is that the money is aimed at the wrong stage of the problem, and no amount of spending fixes a conversion issue at the bottom of the funnel by pouring more traffic in at the top.

The bad option is to proceed with a plan you know does not add up, and hope. That decision is usually made because a target has already been promised upwards, and it costs a great deal more than the conversation it was designed to avoid.

If you are the person who has to have that conversation internally, the arithmetic helps you rather than exposing you. A sum on one page showing the cost per customer the plan assumes, next to the cost per customer you actually achieve, is far easier to defend than an opinion about whether an agency is any good. It moves the discussion off judgement and onto a number, which is where it belongs.

How we handle this at Maitland

We build our proposals backwards from the number, which is why the arithmetic surfaces early rather than in month six.

Every GROW partnership opens with a diagnostic month before any delivery: we audit what exists, set the tracking, record a baseline and come to see you in person. That month exists so that the plan is built on your actual numbers rather than on assumptions, and so that if the sums do not work we find out together at the start.

What that produces when the plan and the budget do fit each other: Complete Weed Control saw leads rise 452 per cent with a 16 times return on ad spend. The Auckland Project saw a 14 times return on investment. Across all our clients in the last 12 months, growth averaged 52 per cent. Those numbers came from plans that were possible before they were ambitious.

Frequently Asked Questions

How do I know if a marketing proposal is realistic?

Run the arithmetic first. Divide the revenue target by your average customer value to get the customers required, then divide the budget by that number to get the cost per customer the plan assumes. Compare it with what a customer costs you today. A gap of three or four times means the plan cannot work, whatever the tactics say.

What are the red flags in a marketing agency proposal?

No baseline or named target. A timescale that ignores how the proposed channels actually work. Costs presented without exclusions. Capability slides instead of named people. And the one most people miss: a proposal that agrees with every part of your brief without challenging anything.

Should an agency ever refuse to answer the brief?

If the brief cannot produce the result it asks for, yes. A proposal that follows flawed instructions faithfully is an expensive way to repeat an existing mistake. The useful response explains the disagreement with numbers and offers an alternative, rather than quietly submitting something that will not work.

What should I do if my budget cannot hit my target?

Three workable options: raise the budget to match the target, lower the target and set a longer timeline, or change the plan so the money is aimed at the real constraint. Proceeding with a plan you know does not add up is the only genuinely bad choice, and it is the most common one.

How do I compare marketing proposals fairly?

Send every agency the same brief, including your current revenue, your target, your budget and your actual cost per customer. Then score them on the six checks rather than on presentation quality. Comparable inputs are what make the outputs comparable, and most tender processes never achieve it.