Social media marketing proposal: the structure that wins

A social media marketing proposal is written for the person who already likes you and read by the person who signs the cheque. Those are rarely the same person. That gap is where good agencies lose work to worse ones.

I have written a lot of these and lost plenty. The ones that got signed were consistently shorter than the ones that did not. Below is the social media marketing proposal structure I use now, and the reasoning behind each section, because copying the shape without the reasoning gets you a nice looking document that still loses.

What goes in a social media marketing proposal?

Seven sections, in this order: the problem in the client’s own words, the objective as a number, the channel plan, the creative plan, the measurement plan, the fee, and the terms. The order matters, because it moves the reader from their problem to your fee rather than from your credentials to their budget.

1. The problem, in their own words. Not a summary. The actual sentence they said on the call. “Our cost per purchase went from 40 riyals to 95 in March and nobody can tell us why.” Quoted back, it proves you listened, and it sets the standard the rest of the document has to meet.
2. The objective, as a number. One number. What a customer is allowed to cost, or how many leads a month at what price. Not “grow engagement”.
3. The channel plan. Which platforms, and one line on what each is for. Crucially, name the platform you are not going to run, and say why.
4. The creative plan. How many assets a month, who makes them, and what happens when the person who makes them goes on leave.
5. The measurement plan. What tracking exists today, what is broken, and who fixes it.
6. The fee. In money, not in percentages alone.
7. The terms. Notice period, account ownership, what leaves with you.

Why does the objective have to be a number?

The objective section of a social media marketing proposal is where most of them quietly give up. “Increase brand awareness and drive qualified traffic” commits to nothing, so it can never be missed, which also means it can never be met.

A real objective starts with the client’s margin. If they make $60 gross on an order, a customer cannot cost $80. That single sentence is the whole plan, and half the time the client cannot tell you the number. When that happens, say so in the proposal and make finding it deliverable one. It is more useful than anything you were going to write about content pillars.

Then state what you will not chase. Reach, impressions, follower growth and video views are diagnostic numbers. They explain results. They are not results.

What should the fee section say?

Say the price. A social media marketing proposal without a number asks the reader to book another meeting just to find out whether you are affordable, and most of them will not bother. Name the model, name the monthly figure in real money, and show what the percentage works out to on their spend.

Fee model	Typical cost	Where it breaks
Flat retainer	$2,000 to $10,000 a month	The agency has no reason to grow the account, and quietly caps effort
Percentage of ad spend	10 to 20 percent	The agency earns more for spending more, whatever it returns
Flat floor, then stepped percentage	Flat under a threshold, rate falls as spend grows	Harder to explain, but honest at both ends
Performance only	A share of tracked revenue	Only works when attribution is clean, which on social it almost never is

Whichever model you pick, show the arithmetic. Ten percent sounds small until it is $4,000 a month on $40,000 of spend, and the client works that out later, on their own, in a worse mood.

We publish our pricing in full before anyone gets on a call, including what each percentage costs in real money. It loses us some meetings. The meetings it loses were never going to close.

How long should a social media marketing proposal be?

About eight pages. The proposal is not the sale. The call is the sale. The document exists so that in month four, when someone asks why TikTok is not in the mix, both sides can point at the same paragraph.

Thirty page decks are usually doing one of two jobs. Justifying a fee that the work will not justify on its own, or hiding the fact that nobody has decided anything yet. Buyers who have been through a few agency relationships read them exactly that way.

The one thing worth adding length for is a worked example. One month of the plan, with real numbers, showing what gets spent, what it should return, and what you would change if it does not. That page gets read twice.

Ad account ownership and the notice period

Two clauses decide what happens when the relationship ends, and most proposals skip both. The brand must own the ad accounts, with the agency holding partner access. And the notice period has to cover media commitments as well as the fee, because media commitments bind on a different clock.

Who owns the ad account. The brand does. The agency takes partner access to the brand’s own Meta Business Manager and Google Ads account. Google’s documentation on manager accounts covers how the linking works, and it is worth reading before you sign anything. If an agency insists on running your budget inside their own account, you lose the entire conversion history the day you leave. That history is what the bidding models are trained on, so you are not just losing a spreadsheet, you are resetting the learning.

What the notice period actually covers. Thirty days on the fee is standard. Media commitments are a separate thing, and they bind harder. The 4A’s and IAB standard terms and conditions, version 3.0, set out the norms the whole industry drafts against: seven days notice on non guaranteed buys priced by click, lead or action, fourteen days on guaranteed CPM, and thirty days on flat fee fixed placements. If your proposal promises a client they can stop next week, and you have signed a fixed placement, you own the gap.

What kills a proposal after the meeting?

Three things kill a social media marketing proposal after the room empties. Silence, because you sent the document with no next step and no date. A guarantee, because nobody controls the auction or the client’s landing page. And a fee buried inside the ad spend, so the client cannot see what they paid you.

Silence. You sent the document and waited. Send it with a specific next step and a date attached, and make the date close.

A guarantee. Nobody controls the auction, the client’s landing page, or a competitor deciding to discount. Promise process instead: concepts per month, review cadence, the number that triggers a stop, the report and the day it arrives. Those you can actually keep.

And the fee buried inside spend. If the client cannot separate what went to the platforms from what went to you, they will assume the worst when results dip. They always do.

What we put in ours

Our social media marketing proposal runs seven pages, and the price is on the website before we send it. The objective section names the cost per customer we aim at, and the number that would make us tell you to stop. The measurement section says that Meta’s reported purchases and your back office will not match.

It says roughly by how much, too. Putting that in month one is information. Putting it in month three is an excuse.

If you want the wider planning context, we have written up how to write a digital marketing plan, which is the document a proposal should sit inside, and media buying, which is the work the proposal is actually selling.

Below roughly $5,000 a month in ad spend, no social media marketing proposal is worth signing, including ours. There is not enough data for anyone to optimise against and the fee makes the arithmetic worse. Above that, fifteen minutes is usually enough to tell whether a proposal is worth writing at all.