choosing between two proposals.
two proposals are never comparable as written. normalise the scope first, then judge what price cannot show.
convert both to per-month deliverables, per-month senior hours and named people. then score on thinking, honesty, measurement and fit. price is a tiebreaker, not a criterion.
agencies write proposals to be persuasive rather than comparable, which means the first task is arithmetic and the second is judgment. done in that order it takes an hour. done in the other order you end up choosing on presentation quality, which correlates with sales investment rather than delivery.
here is the method we recommend to clients even when we are one of the two.
one thing to settle before you start: who decides. a comparison run by three people with different priorities produces a compromise choice, and compromise choices tend to pick the least objectionable proposal rather than the most useful one. name a decision-maker and let everyone else advise.
— 01step one: normalise the scope.
build a single table with four rows and two columns: countable deliverables per month, shoot or production days per month, named senior people and their roles, and channels in scope. fill it from each proposal, and where something is not stated, write "not stated" rather than assuming.
the "not stated" cells are the most informative part of the exercise. they are what you will argue about in month three, and they are usually the difference between two prices that looked far apart.
then convert to a rough unit cost — fee divided by monthly deliverables, and fee divided by production days. this is crude and it still reveals a great deal: a proposal that is 40% cheaper per month but produces half the output is more expensive per unit, and a proposal that is more expensive with two shoot days included is often the better value.
finally, separate media spend from fees in both. an agency bundling them together may look cheaper because it has assumed a smaller budget, which is not a discount — it is a different plan.
watch for one-time fees hidden in the monthly number too. onboarding, setup, tracking implementation and playbook development are legitimate charges, but a proposal that folds them into month one without labelling them looks cheaper than it is over three months.
| compare on | how | weight |
|---|---|---|
| deliverables per month | count them; mark anything "not stated" | high |
| production days per month | shoot days, crew included or not | high |
| named senior people | names and roles, met before signing | high |
| measurement commitment | what is reported, and the enquiry definition | high |
| disagreement in the proposal | did they push back on anything | medium |
| structural fit | similar funnel, not necessarily same sector | medium |
| price | unit cost, not monthly total | tiebreak |
| case studies and logos | — | ignore |
| deck design | — | ignore |
— 02step two: judge what the table cannot show.
four things, and they matter more than the price difference. thinking — did they quote your own numbers back at you, and did they say something you had not considered? a proposal that restates your brief is a template with your logo on it.
honesty — did they disagree with anything, name a risk, or tell you something would not work? a proposal that agrees with every element of your brief has no point of view and will not push back when you are about to waste money.
measurement — what specifically will be reported, and how is a qualified enquiry defined? vagueness here predicts a year of reach screenshots — ask both to describe their standard monthly report and compare the two answers directly.
fit — have they worked in something structurally similar, and did the people you would actually work with attend the meeting? sector experience matters less than most clients think; structural similarity matters more. an agency that has run a long-cycle, high-value funnel will adapt to your category quickly; one that knows your industry but only ever ran ecommerce will not.
— 03step three: the two questions that break a tie.
ask both agencies the same two questions and compare the answers rather than the decks. first: "what would make you tell us to stop spending on a channel?" a specific answer means they have done it. a general one means they never have.
second: "what happens in month one, in deliverables?" the useful answers are unglamorous — audit, tracking, a written thesis, foundation fixes. an answer that begins with posting or campaigns has skipped the part that makes the rest work.
if both answer well, you have two good options and the decision is about people rather than proposals. that is a legitimate basis for a choice and worth admitting openly.
if you are genuinely undecided, buy a small project from your first choice before committing to a retainer. one well-scoped piece of work tells you more about how a team communicates, meets deadlines and handles feedback than any number of further meetings, and it costs a fraction of a wrong twelve-month decision.
— 04what to ignore.
the case study slides, mostly. everyone shows their best work and nobody shows what it cost or how long it took. one anonymised monthly report tells you more about how an agency operates than ten case studies about how it sells.
the logo wall, entirely. large logos often represent a small piece of work for a big client, and it says nothing about who would be on your account.
and the deck design. it correlates with how much the agency invests in new business rather than delivery, and occasionally inversely with how much attention your account will get. our own scope and pricing sit on a public page for exactly that reason — it removes the performance from the comparison.