How to Review a Marketing Agency Proposal: What It Commits Them To, and What It Quietly Leaves Out

Read a marketing agency proposal for commitment, not content. Most of the document describes activity the agency will perform; what you need to find is the four things it binds them to — who is named on the work, what outcome (if any) is promised, what sits outside the price, and what you walk away with at the end. Everything else is packaging.

By Avi Vatsa, CEO, Exchange Four Agency. Avi leads the team that installs and runs AI-leveraged revenue systems for owner-led companies. His background spans law, technology, and marketing; he also co-founded Dialora, an AI voice-agent platform for automated lead capture and booking. (Marketer of the Day #1411)

Why does a proposal feel convincing and still commit to nothing?

Because a proposal is a sales document that has to survive a legal review. The two jobs pull in opposite directions. The sales half wants to sound ambitious; the legal half wants to avoid promising anything measurable. The resolution, almost universally, is a long list of activities — audits, content calendars, campaign management, monthly reporting — written in the future tense and tied to no result.

That is not a scandal. It's the format. But it means the owner reading it is drawing conclusions from the wrong part of the page. You read "full-funnel strategy and ongoing optimisation" and hear they will get my number up. The document says they will do work. Those are different sentences.

The gap shows up later as frustration about proof. An industry survey distributed via Businesswire found 71% of brands cite frustration demonstrating marketing ROI effectiveness — and a large share of that frustration was baked in at signature, in a document where nobody ever wrote down what success would be.

So read the proposal with a highlighter and four questions. Print it. Four passes, each one looking for one thing.

---

Pass one: Who is actually named on the work?

What a good answer looks like: named individuals with roles and rough time allocation. "Strategy lead: [name], ~6 hrs/month. Paid media: [name], ~20 hrs/month." Or at minimum, a named senior owner of the account who appears in the proposal and will appear on your calls.

What a vague one looks like: "our team of specialists," "a dedicated account manager will be assigned," "senior strategists oversee all accounts." Note the passive voice in will be assigned. Something will happen to you later.

Here's the specific thing to check, because it's the one that catches people: compare the names in the pitch meeting to the names in the document. If the person who impressed you is in the room but not on the page, the proposal is not binding them to your account. That's how an engagement sold by a principal gets delivered by whoever had capacity that month.

Turnover is the real downstream cost. The G2 review pattern for agency work is blunt about it — clients describe "constantly changing project managers and the issues with communication made it difficult to work with them." Each handover restarts the context-building. You pay for the ramp-up twice.

The question to send back: "Can we add the named strategy lead and their monthly hours to the scope, with a clause that a change of lead requires notice to us?" The answer tells you more than the proposal did. An agency that genuinely staffs seniors on accounts will say yes in a sentence. If the staffing is fluid, you'll get a paragraph about flexibility and resource allocation.

If you're already living this, the deeper read is what account-manager turnover actually signals.

---

Pass two: Is any outcome committed, or only activity?

This is the pass that matters most, and it's quick. Go through the deliverables section and sort every line into two buckets.

Most proposals you receive will have zero lines in the outcome bucket. That is normal, and on its own it isn't disqualifying — a serious agency may decline to guarantee a revenue number it can't control, and declining is more honest than guaranteeing. Only promise what you can deliver; that's a defensible position.

What is not defensible is a proposal with no outcome language and no definition of what the parties will look at to decide whether it's working. If nobody writes down the measure, month seven becomes an argument about whether the brand is building.

The question to send back: "What number will we both look at in month 6 to decide if this is working, and what's the plan if it isn't moving?"

The second half of that question is the real one. Any agency can nominate a metric. The plan-if-it-isn't-moving is where you find out whether they've thought past the sale. A good answer is specific about what gets changed first — the offer, the targeting, the follow-up — and in what order. If the answer is "we'd reassess strategy," you have learned something.

Watch out for the metric that looks like an outcome and isn't. Impressions, reach, "engagement," keyword positions on terms nobody buys from, MQLs with a definition loose enough to include anyone who downloaded a PDF. Our own view on which numbers belong in a monthly report is set out in the numbers an owner should demand every month and in how to calculate marketing ROI when the dashboard never ties to a dollar.

On timeframe: don't demand a committed outcome by month two. Revenue systems take time to compound, and an agency that promises otherwise is selling you a gamble with confident framing. What to expect month by month is a separate read, but the short version: judge leading indicators early, lagging ones from month four.

---

Pass three: What sits outside the price?

Every proposal has a boundary. Few of them draw it clearly, and the hidden cost of the unclear boundary is the change order you didn't budget for.

Hunt for these specific omissions. They are the ones that recur:

• Ad spend. Is the fee inclusive or on top? If on top, is there a management percentage, and does it scale with spend?
• Landing pages and creative. "Campaign management" frequently excludes building the page the campaign points at.
• Tooling and licences. CRM seats, scheduling tools, call tracking, email platform, SEO software. Who pays, and whose account is it in?
• Revisions. How many rounds are included before extra work is billable?
• Integration and setup. Connecting your CRM to the forms, the calendar, the phone system. Often the largest unpriced item.
• Anything described as "scoped separately" or "subject to a later quote." That phrase is a blank cheque with a polite name.

What a good answer looks like: an explicit Not included section. Genuinely — the best proposals have one, and its presence is a quality signal. It means somebody thought about the boundary before you did.

The question to send back: "What's the first thing you'd expect to come back to me with as an out-of-scope quote in the first six months?" An agency that has run this engagement before answers immediately and specifically, because they know. "Usually the landing pages — your current site builder makes it slow." That's experience talking. A pause and a reassurance is a different signal.

The multi-vendor version of this problem is worse, and quantified: industry vendor-management research puts coordination overhead at 8–15% of annual vendor spend in hidden labour that never appears on an invoice, with businesses reporting roughly 30% more total spend versus one integrated partner. If this proposal is for vendor number three, that arithmetic is already running against you — the hidden coordination bill breaks it down.

---

Pass four: What happens when it ends?

Read the last page first next time. Four items, and a proposal that handles all four well is unusual enough to be worth noticing.

1. Term. Is there a minimum? Three months is reasonable ramp-up. Twelve with no out is a financing arrangement, not a partnership.
2. Notice period. Thirty days is standard. Ninety is a quarter of your year you can't redirect.
3. Account ownership. Are the ad accounts, the domain, the CRM, the tracking properties in your business's name with you as owner, and the agency as a user? This is the single most expensive thing to get wrong and the cheapest to fix at proposal stage — one sentence.
4. Exit deliverable. What do you physically receive on the last day? Raw data, source files, documented processes, credentials, a handover call?

What a vague one looks like: silence. Most proposals simply don't address the ending, because the proposal is written in the optimism of the beginning.

The question to send back: "On the last day of our engagement, whenever that is, what's the list of things I receive?"

Ask it in week one, cheerfully. Nobody is offended. The answer is a direct read on whether the agency builds things that keep working without them or builds a dependency. We've covered who owns your marketing assets when you leave in full, and the contract clauses worth reading closely separately — the proposal is where you catch them early, before anybody's lawyer is involved.

---

What if the proposal fails all four passes?

Send the four questions. Don't walk yet.

A weak proposal and a weak agency aren't the same thing. Plenty of good operators write mediocre proposals, because the document was assembled from a template at 11pm. The four questions are the test — not the PDF.

What you're watching for in the replies is whether the answers come back specific and fast, or general and slow. Specific and fast means they've run this engagement before and know where it gets hard. General and slow means the proposal was the thinking, and there isn't more behind it.

The pattern to avoid resembles what reviewers describe as "cookie-cutter" — where "processes are confusing" and the team lacks the capacity to change direction. If every answer you get back could have been sent to any other prospect that week, you're reading a template. How to spot a templated agency goes deeper on the tells.

And if the whole category is starting to feel like a bet, that's worth examining too. The owner's objection — "every agency promises a lot and ends up being one more vendor I have to manage and chase" — is usually earned. The structural answer isn't a better proposal. It's a different arrangement, where someone owns the result instead of the scope. What changes when an agency owns the outcome sets out that difference, and the alternatives owner-led companies actually weigh prices each option honestly.

The one-page version

Before you sign anything, confirm the document answers these four. If it doesn't, the questions above are a 10-minute email.

"They know exactly how to connect marketing execution to real business outcomes," says Riggs Eckleberry, Chairman of OriginClear, describing the Exchange Four engagement. That connection — execution to outcome — is the thing a proposal should make visible before you sign, not something you find out about in month nine, and it's the same connection you should be pressure-testing every quarter once the work is underway — here's the agenda that keeps an agency accountable.

A proposal that survives all four passes is rarer than it should be. It is also, usually, the one worth signing.