nine red flags in an agency pitch.

agency pitches are optimised. the deck is good because decks are the product of the sales process, not the work. so instead of judging the presentation, judge the structure underneath it — how they get paid, what they measure, and who actually does the work. every shortlist contains at least one shop calling itself the top digital marketing agency in the market; the structure is how you tell whether that claim survives contact with a contract. these nine signals come up repeatedly, and each of them predicts a specific kind of disappointment.

none of them are hard to spot once you know to look, and most surface in the first meeting. the difficulty is that pitches are pleasant, deadlines are pressing, and it feels rude to interrogate someone who has just spent two weeks preparing for you. do it anyway: an hour of awkward questions is cheaper than a year of polite disappointment.

— 01the four that should end the conversation.

a guaranteed ranking or a guaranteed roas. nobody controls the auction or the algorithm. a guarantee means either the target is set low enough to be trivial or the contract has an escape clause you have not read yet.

a fee that is a percentage of ad spend. it pays the agency to grow your budget whether or not it works, and it makes "spend less, earn the same" a recommendation that costs them money to make. flat fees remove the conflict entirely.

reporting that leads with reach or impressions. those are inputs — useful diagnostics inside a proper performance report, useless as a headline. an agency that headlines them either cannot connect spend to enquiries or would rather not.

no named people. if you cannot meet the strategist, writer and editor before signing, you are buying a pool. the pitch team is not the delivery team, and the gap between them is where most disappointment lives.

a fifth borderline case deserves mention: an agency that will not let you speak to a current client directly. references are stage-managed everywhere, but a flat refusal usually means the roster is either thinner than the logo wall suggests or unhappy enough to be risky.

— 02the five that need a hard question.

a twelve-month contract with no exit — ask for the notice period in writing and who keeps the assets. an unwillingness to name a price band before a second meeting; that is a sales process, not a scoping problem. a proposal that lists every service you asked about and nothing you did not; an agency with a point of view disagrees with something.

a portfolio with logos and no numbers — ask for one anonymised monthly report instead, and watch whether the answer is proud or awkward. and a team that has never told you no, which is less a red flag in the pitch than a pattern to watch for in month two.

one more that people miss: an agency that owns your ad account and page admin rather than being added as a user on yours. it is presented as convenience and functions as exit friction. every asset built with your money should sit in your name from day one, and the agencies confident in their work will offer that before you ask.

— 03why incentives beat credentials.

most agency relationships do not fail because the work was bad. they fail because the commercial structure rewarded something other than your outcome — volume over quality, spend over efficiency, activity over decisions. you can see that structure in the contract before you have seen any work at all, which makes it the highest-leverage thing to read carefully.

the useful reframe: assume the agency will behave rationally within the incentives you agree to. then read the fee model and the reporting commitment as a prediction of their behaviour, because that is what they are.

this is also why credentials mislead. a shop with excellent case studies and a percentage-of-spend fee will still, eventually, recommend spending more — not out of cynicism but because that is the direction the incentive points. conversely a modest studio on a flat fee has no reason to inflate anything, which tends to show up in reporting that admits when a month was flat.

— 04what a good pitch looks like instead.

it quotes your own numbers back at you. it disagrees with at least one thing in your brief and says why. it names people, with roles. it states what month one produces as deliverables rather than adjectives, and it says out loud what it will not do — the exclusions list is the most honest slide in any deck.

and it gives you a number without three meetings. a shortlist you can compare on scope beats a shortlist you can only compare on charisma.

if you want one question to carry the whole review, use this: "what would make you tell us to stop spending on a channel?" an agency with a real answer has recommended cutting its own scope before and will do it again. an agency without one has never had that conversation, which tells you how the next flat quarter will go.