How to Choose a Meta Ads Agency (Without Regret)

This is the most important question and the one most often skipped. Meta's system has automated most of the targeting and bidding work, which leaves creative as the main performance lever. An agency that does not produce creative is managing the variable that matters least while waiting for you to supply the one that matters most.

Ask directly: do you make the ads, or do I? If the answer is that you supply assets, you are effectively taking on a video production workload alongside the retainer, and the account's performance will be capped by how quickly you can produce new angles — which for most businesses means the account starves within two months.

If creative is included, ask how much of it: how many new assets enter the account per month, who writes the hooks, and whether they can film or direct you to film. Vague answers here usually mean a freelancer will be found after you sign.

Three pricing models dominate. Percentage of ad spend aligns the agency with growth but creates an obvious incentive to recommend higher budgets and works poorly at low spend levels. Flat retainers are predictable and easier to compare but need to be matched to a defined scope. Hybrids combine a base fee with a percentage above a threshold.

None of these is inherently better, but each needs a clear scope attached. The question that makes quotes comparable is what is included beyond media buying: creative production, landing page work, tracking setup, reporting frequency, and how many campaigns or products are covered.

Be especially careful with quotes that look cheap. A low management fee with creative billed per asset frequently costs more in total than an inclusive quote, and it puts the agency's incentive in the wrong place — they earn more when you need more assets, rather than when the account performs.

Campaigns should run inside your Business Manager and your ad account, with the agency added as a partner. This is standard practice and any resistance to it is a serious warning sign.

The reason matters beyond principle. Your ad account accumulates pixel history, conversion data, custom audiences and creative — assets that materially affect performance and that take months to rebuild. Running under an agency's account means that when the relationship ends, so does all of it, and your next agency starts from zero.

Ask explicitly what happens at the end of an engagement: who keeps the account, whether you receive the creative files, and whether audiences and tracking configuration transfer. Get the answer before signing, not during a handover.

Meta's reported conversions and what you can independently verify rarely match exactly, for structural reasons. An agency that presents platform numbers as literal fact is either inexperienced or comfortable letting you misread them, and both cost you money over a year.

Good reporting shows both sides: platform-reported results and your own verifiable outcomes — enquiries, booked calls, orders — with an explanation of why they differ. Ask to see a sample report before signing, and see whether it explains decisions or simply displays numbers.

Also ask what they report weekly versus monthly. Weekly reporting should be about upcoming decisions: what is being scaled, what is being cut, what creative is next. Monthly reporting should step back to cost per acquisition trends and the plan for the following month.

This question is worth more than any case study. An agency that has run enough accounts has lost money on some, and how they describe it reveals how they think: whether they diagnosed the cause, whether they told the client early, and whether the lesson changed their process.

Answers that deflect entirely — blaming the client's product, the market, or the platform without any self-assessment — predict how they will behave when your account has a bad quarter. Answers that describe a specific diagnosis and a specific change are a good sign.

A related question: when would you tell me Meta is the wrong channel? Agencies willing to name the conditions under which they would recommend stopping are far more trustworthy than those who believe paid social suits everybody.

Guaranteed ROAS figures are the clearest one. Nobody can guarantee a return before seeing your offer, margins, tracking and market, and a specific promised multiple is a sales device rather than a forecast.

Long lock-in contracts with no performance review, refusal to run in your own account, screenshots without context, and an unwillingness to say who will actually work on the account day to day are all worth taking seriously. So is an agency that never asks about your margins — because without them, nobody can tell you what a good cost per acquisition would even be.

One subtler flag: an agency that talks exclusively about targeting, audiences and bidding strategy without mentioning creative. That vocabulary is several years out of date and usually signals a process built for how Meta worked before automation took over the levers they are describing.

Month one should be unglamorous: tracking audited and fixed, account structure rebuilt, the first creative batch produced and launched, and a baseline cost per result established. Expect information rather than profit.

Month two is where testing produces direction — which angles resonate, which audiences respond, which landing page issues are limiting conversion. Costs usually become more stable here even if they are not yet at target.

Month three is where scaling decisions become sensible, because there is finally enough data to scale on. Any agency promising target performance in week two is either lucky or overstating, and the expectation it sets usually causes the account to be switched off before it works.

A freelance media buyer is the cheapest option and can work well when you already have a strong creative pipeline and someone internally owning strategy. The risk is single-person dependency and the frequent gap where nobody is producing new ads.

Hiring in-house makes sense at higher spend levels where the salary is small relative to the budget being managed, and where there is enough work to keep a specialist busy. Below that, the same money buys more capability from an agency because you are sharing a team rather than employing one.

An agency that combines media buying with creative production tends to suit businesses that need both and do not want to coordinate between a buyer and a separate video team — which is the arrangement where most of the friction and most of the delays live.