How to choose a digital marketing agency as a DTC brand

Choose the agency that can tell you, before you sign, what your ad budget should be and where your growth is actually stuck. If you have hired an agency before, the usual checklist of case studies and references won’t separate anyone, because every agency passes it in a pitch. The eight questions below are built so a weak answer is hard to fake, and you can ask all of them on a first call.

I run an agency, so weigh this accordingly. These are the questions I’d want a founder to ask us, and they work on us too.

Why the usual checklist doesn’t help if you’ve hired before

Most guides on how to choose a digital marketing agency tell you to set goals, check case studies, call references, ask about reporting and read the exit terms. That’s sound advice for a first hire.

If you’ve already been through two or three agencies, you know it doesn’t separate anyone. Every agency arrives with a case study deck and a good pitch. The one that let you down last time passed the same checklist.

What actually differs is whether the agency will do the senior thinking: work out what the business can afford to spend, find the real reason growth has stalled, and take responsibility for fixing it. The other kind runs the ad account, sends a report and leaves the thinking to you. If you’ve had that kind, you know where it ends. You’re back in the marketing seat, briefing creative and reading dashboards at night, doing the work you paid someone else to do.

Each question below is easy to answer well if the agency does this work, and hard to bluff if it doesn’t.

1. “What should our ad budget be?”

Ask it exactly like that. If the answer is “what’s your budget?”, you have your answer.

An agency that asks what your budget is doesn’t know what your budget should be. The budget falls out of your numbers: what you keep from an average order after product cost, shipping, payment fees and returns, how often customers buy again, and how much cash and stock you have to grow into. A good agency tells you which of those numbers it needs and what it will do with them.

Here is the shape of it, with made-up numbers. Say your average order is $80 and you keep $32 of it after product cost, shipping, payment fees and returns. That's a 40% contribution margin before ads, so a first order breaks even at a ROAS of 1 divided by 0.40, which is 2.5x. Ads that return 2.0x on first orders lose money on the first purchase and only pay if repeat purchases make it back.

A good answer then goes one step further. Each extra dollar of ad spend tends to bring back less than the one before, so the right budget is the level where the next dollar stops paying for itself. You find it by raising spend in steps and watching store revenue, not the platform’s dashboard. An agency that can explain that on a first call has done it before.

You can run this test before any call. Email your current agency and ask what your ad budget should be, worked out from your revenue and margin. The reply tells you a lot.

2. “What would you need to see before telling us where the problem is?”

A price before a diagnosis is a red flag at the doctor’s, and it’s a red flag here. If the pitch deck already has your channel plan and a package in it, the agency has prescribed before it has looked.

A good answer asks for things outside the ad account: your margins, or the P&L if you’ll share it, store analytics, the path from ad to checkout, your offer, your retention numbers and how tracking is set up. Then it gives you hypotheses and says what data would confirm or kill each one.

This matters because the reason a brand can’t scale often sits outside the ad account. On accounts we’ve worked on, it has been a conversion leak on the site, a basket too small to pay for cold traffic, and tracking that counted the wrong thing. An agency that only runs Meta can’t find a problem that isn’t on Meta. It can only prescribe more Meta.

Our own framework starts in the same place: find the biggest bottleneck in the business first, then decide what channel work it needs.

3. “Which number will you judge yourselves on?”

Meta and Google each credit sales to their own ads by their own rules. Google lets you pick an attribution model, which decides how much credit each ad interaction gets, and Meta applies its own attribution setting. The same order can be claimed by both, and neither tells you what would have happened without the ad.

The gap can be large. On Loose Leaf Tea Market, Meta was reporting 2.23 times the revenue the store’s own records credited it with. It runs the other way too. We’ve seen a brand whose ads were carrying far more of its revenue than Meta claimed, which only showed up when the ads were switched off.

So ask which number decides. The answer you want is store revenue and margin, with platform ROAS as one input. A better agency will also tell you how it would test what the spend actually adds, for example by measuring what happens to store revenue when budgets move, or by pausing spend in a controlled way. If the answer is “ROAS in Ads Manager”, the agency is marking its own homework.

We hold ourselves to the same standard: before we start, we agree targets for revenue, blended MER and the cost of a new customer, and every monthly report is measured against them.

4. “What will you check before you spend more?”

Google’s Smart Bidding sets a bid in every auction to get more of the conversions you’ve told it to count, and only actions marked primary are used for bidding. If the wrong thing is marked primary, the bidding buys more of the wrong thing, and it does that very efficiently.

We’ve seen this on a lead-gen account. Its best-looking campaign was reporting cheap conversions, and every one turned out to be an accidental tap on a phone link inside apps and on low-quality sites. Because that action was primary, the bidding kept moving budget toward it.

A good agency checks tracking before it touches budgets: which conversion actions are primary, whether each purchase is counted once, and whether the platforms’ order counts roughly match the store’s. If tracking is broken, it fixes that first and changes nothing else until it’s fixed. If the plan for week one is new campaigns, ask what they checked first.

If you’ve asked your current agency these four and the answers worried you, put all eight to us on a strategy call, which is for brands doing at least $50,000 a month in revenue and spending at least $5,000 a month on ads. We’ll tell you straight if we’re not the right fit.

5. “Who will make the daily decisions on our account?”

Senior strategy means little when junior hands make every daily decision. The person who pitches you is often not the person who changes your budgets on a Tuesday.

Ask who that person is, how many other accounts they carry, and whether you can meet them before you sign. Then ask them to walk you through a decision they made last week on an account like yours. You’ll learn more from that conversation than from the deck.

At Growth Mate one senior lead owns each account day to day, with specialists in paid media, creative and email behind them. One owner, not two.

6. “When did you last switch off something you built?”

Every agency can tell you about a win. Ask for the last time they turned off their own work because it wasn’t paying, and what they did with the money.

Here’s ours. On Loose Leaf Tea Market we ran 155 days of spend through the store’s own order records to measure what each platform’s next dollar actually bought. Two campaigns we had built ourselves came back at zero, so we stopped one and raised the target on the other.

An agency with no example has either not measured its own work or not told you about it. My rule for Growth Mate is simple: if I don’t believe something will make you money, I won’t sell it to you. The same goes for keeping it running.

7. “What does your report ask me to decide?”

ROAS is the easiest number to make look good, because it tends to rise when spend falls. Cut the budget back to the most efficient campaigns and ROAS goes up while revenue and new customers go down. The report looks better and the business gets smaller.

I’ve seen a weekly report call it a great week while daily spend had collapsed and ROAS held. Technically right, commercially a disaster.

A useful report leads with store revenue, new customers and what you made after ad spend, then says what changes next week and why. Ours go out monthly too, so hold us to the same test: every report should end in a decision.

8. “What would make you turn us down?”

An agency that would take any brand at any budget is optimizing for volume. A doctor who accepts every patient at any price isn’t doing much diagnosing. Ask what the agency would need to see to say no.

Good answers are specific: margins too thin to carry paid spend, no access to the store’s data, or a target the unit economics can’t reach. “We work with brands of every size” is not an answer.

Ours starts with size, and it is written down: our strategy call is for brands doing at least $50,000 a month in revenue and spending at least $5,000 a month on ads. Ask any agency where its line is and why it sits there.

How to score the answers

Question	An answer that should worry you	The answer you want
What should our ad budget be?	“What’s your budget?”	The numbers they need, and how they’ll turn them into a break-even and a target
What would you need to see first?	A channel plan and a package in the pitch	Margins, store data, the site, the offer and tracking, then hypotheses
Which number will you judge yourselves on?	Platform ROAS	Store revenue and margin, plus a way to test what the spend adds
What will you check before spending more?	New campaigns in week one	Conversion actions, double counting, platform orders against store orders
Who makes the daily decisions?	“A dedicated team”	A named person you can meet, who can walk you through last week
When did you last switch off your own work?	No example	A specific campaign, why it went and where the money moved
What does your report ask me to decide?	A PDF of platform metrics	Store numbers first, then what changes next week
What would make you turn us down?	“We work with everyone”	Specific deal-breakers on margin, data access or targets

One worrying answer is a conversation. Several is a pattern.

When these questions are overkill

If you’re still finding product-market fit, or ads are a small test rather than a channel the business depends on, you don’t need a partner to do the senior thinking yet. A good freelancer who runs the account well costs less and is enough for now.

They’re also the wrong questions if you want someone to run ads exactly the way you specify. That’s a legitimate thing to want. Hire for clean execution, keep the thinking yourself, and judge the agency on how well it follows your plan.

If an agency gets uncomfortable at question one, that tells you most of what you need to know.

How this was made: drafted with AI help from Growth Mate’s working playbooks and case records. Every figure links to the page it comes from.