eCommerce Agency vs In-House: How to Actually Decide

Agency or in-house comes down to two ownership questions. Who owns the supply of new creative every month, and who owns the scoreboard that says whether the spend is making money. Headcount follows from those answers. Hire in-house when the work can keep senior specialists busy full time and you want the testing system as a company asset. Use an agency when you need creative volume and buying discipline without three senior hires first. Plenty of brands run both.

We're an agency. Read this knowing that. We've kept the cases where hiring beats hiring us, and put our own costs on the table alongside yours.

Who owns the creative supply and the scoreboard

Growing an eCommerce brand on Meta runs on three jobs: creative production, media buying, and retention in email and SMS. Doing each once is easy. Doing all three every week, at volume, with a testing system behind the creative and a profit number over the top, is what breaks teams.

Creative is a supply problem before it's a quality problem. The auction rewards new creative more than new targeting, which is why the house line is "Creative is the new targeting." So the first question is mechanical. How many new ads can this option put into the account in a month, and what share of them win? That share is the hit rate: winners divided by ads tested. The long version is our creative testing framework.

The scoreboard is the second question. Whoever owns the number reports in lifetime gross profit to CAC. CAC is what you pay to acquire one new customer. ROAS is revenue divided by ad spend in a dashboard, a diagnostic rather than a verdict. AOV is average order value. Lifetime gross profit is the margin one customer returns across their whole life with you, and it's the only figure that says whether an acquisition cost was worth paying. Unit economics covers how to build it.

Answer both and the org chart writes itself.

The real cost of in-house

Payroll is the visible number and the smallest part of the decision. Here's the model. Lift it into a spreadsheet.

• Fully loaded role cost = base salary
• + payroll taxes and benefits
• + tools for that role
• + recruiting cost spread over expected tenure
• + your management hours x your hourly value
• Ramp cost = fully loaded monthly cost x months before full output
• Churn cost = (recruiting cost + ramp cost) / expected tenure in years
• True annual in-house cost = sum of role costs + ramp cost + churn cost
• Cost per tested ad = (creative payroll + creator fees + editing) / ads shipped

Most founders have never calculated that last line, and it decides whether an in-house team can outrun fatigue. Run it on your own production, then compare it to what an outside team quotes for the same ad count.

• Senior media buyer. What it covers: Account structure, budget allocation, scaling calls. The hidden cost: Ramp. They inherit history they didn't build, and read it wrong before they read it right.
• Creative strategist. What it covers: Angles, briefs, hooks, iteration trees. The hidden cost: Management time. Approval queues are where creative volume dies.
• Editor or motion designer. What it covers: Turning footage and stills into shippable variants. The hidden cost: Tools. Software, stock, storage, and an asset library to maintain.
• Creator sourcing. What it covers: Finding, briefing, and paying creators. The hidden cost: Churn. Those relationships leave with the person who built them.
• Retention owner. What it covers: Flows, campaigns, segmentation, list growth. The hidden cost: Tools plus coverage. One person is one opinion and no backup.
• Analyst, or you at night. What it covers: Cohort reporting, margin, the scoreboard. The hidden cost: Your time. Founders absorb this role silently.

Salary figures vary by market and seniority, so we won't invent a range. For a public reference, the US Bureau of Labor Statistics lists 2025 median pay for advertising, promotions, and marketing managers at $165,780 per year, with marketing managers at $166,790 as of May 2025 (BLS Occupational Outlook Handbook). Specialist buyer, editor, and retention pay isn't broken out there. Price those against live listings in your market.

The structural risk of in-house is concentration, not cost. One generalist covering buying, creative, and retention will quietly let creative starve, because creative is the most labor hungry job and the easiest to postpone.

The real cost of an agency

Agencies have hidden line items too, and most articles on this topic skip them because agencies write them.

• Fees. A retainer is fixed cost that never becomes an internal asset. Stop paying and the capability leaves.
• Onboarding. Some charge a setup fee on top of month one, before a single ad ships.
• Lock-in. Long minimum terms protect the agency's forecast. A twelve month minimum is eleven months of being wrong if the fit is bad.
• Junior staffing. The senior operator who sold you is often not the person in your account on Tuesday. Ask who is, by name.
• Shared attention. You're one of several accounts. That's also where the advantage comes from, since a good team sees what's working across many accounts this month.
• Asset ownership. Ad accounts, pixels, creative files, and flows. Get it in writing that they stay yours.

Our published terms are: "No onboarding fees. No lock-in contracts. No junior marketers." (email.hayesmedia.co). Hold every agency you talk to, including us, against all three.

The full evaluation checklist is in how to choose a Meta ads agency, and we ranked the field in best Meta ads agencies for eCommerce.

The alternatives to hiring in-house

Agency and in-house aren't the only two doors. Here's the real menu, and where each runs out.

• Freelancer. What it covers: One job: editing, a flow build, campaign management. Where it runs out: No system behind them. You're the strategist and the QA.
• Fractional media buyer. What it covers: Buying decisions and account structure, part time. Where it runs out: No creative. A buyer with no supply optimizes a shrinking pool.
• Creative-only agency. What it covers: Volume of new ads, angles, iteration off winners. Where it runs out: Someone still has to buy, and the loop breaks when they aren't in the room.
• Full-service growth agency. What it covers: Creative, buying, and retention on one scoreboard. Where it runs out: Fixed fees, shared attention, and quality rides on who is staffed on you.
• Retention-only agency. What it covers: Email and SMS flows, campaigns, list growth. Where it runs out: Acquisition keeps setting the ceiling. See eCommerce email marketing agency.
• Done-for-you operating partner. What it covers: The whole machine, reported in profit. Where it runs out: You give up day to day control. Our take: done-for-you eCommerce marketing.

The common mistake is buying a buyer when the shortage is creative. If three ads carry the spend, restructuring the account won't fix it.

Hybrid: the model most brands land on

In-house owns the number. One internal person, usually the founder or a head of growth, owns lifetime gross profit to CAC and has final say on brand, product claims, offers, and approvals. That role never gets outsourced. Nobody outside your building can carry your margin.

The agency owns volume. Creative production at a rate an internal team can't sustain, media buying with pattern exposure across accounts, and retention when your operator isn't stronger than theirs.

Three conditions make it work. One scoreboard judges both sides. Approvals move fast enough that creative volume doesn't queue. And the internal owner reads the account well enough to challenge the agency.

Choose in-house, an agency, or hybrid

• Spend and product range keep a senior buyer busy full time. Choose an agency if: You need creative, buying, and retention capability now. Choose hybrid if: You have a strong internal owner but no creative bench.
• You have a hiring pipeline and can survive a bad hire. Choose an agency if: Creative supply is the bottleneck. Choose hybrid if: Brand judgment stays inside, volume comes from outside.
• Product complexity means outsiders stay a step behind. Choose an agency if: You want pattern exposure from many accounts. Choose hybrid if: You want the machine proven before you hire against it.
• You want the testing system as an asset you keep. Choose an agency if: You want the option to stop without severance. Choose hybrid if: You tried one side alone and know which half failed.

How to judge either one

Same checklist, both sides of the table.

1. Do they report in lifetime gross profit to CAC? If the first slide is platform ROAS, they're grading their own homework.
2. What's the hit rate? Winners divided by ads tested. If nobody can state it, nobody is measuring creative as a system.
3. How many new ads per month, and who makes them? Names and roles, not a promise of volume.
4. What happens to a winner? A real answer describes iteration trees off the winning angle.
5. Who is in the account on Tuesday? The person, their seniority, their other accounts.
6. How is retention connected to acquisition? If the ads team and the email team never talk, you pay twice for one view of the customer.
7. What do you keep if this ends? Ad accounts, pixels, creative files, flows, documentation.
8. Can I talk to a brand at my size who chose this path? Then call them.

Across our accounts we also benchmark creative against the engagement graph, watching hook rate above 40% and retention at fixed timestamps. Those are internal heuristics, not a platform standard. Any team you hire should be able to say what they watch and why.

A worked example, using the mechanism

Take a brand where acquisition is unprofitable on a first-order basis and retention is an afterthought. The in-house instinct is to hire a media buyer, because the ads look broken. The mechanism says otherwise. If first-order margin doesn't cover CAC, the fix is more creative at bat until the hit rate produces cheaper winners, or more lifetime gross profit per customer so the same CAC clears.

That's what the Remi engagement looked like. The work covered both sides: "producing high-quality conversion-optimized ad creative, revamping product pages with A/B testing, building landing pages to increase conversion rate, and simplifying the media buying strategy on Meta," plus "new email flows to increase new customer conversion rates" and "direct mail campaigns to convert one-time customers into subscription customers."

The published outcome: "The efforts on both customer acquisition and customer retention turned what was an unprofitable CAC to a profitable CAC for new customer acquisition, and led to tremendous revenue and LTV growth." The headline results on the same page: "12,400% Revenue Growth" and "150% ROAS Increase" (hayesmedia.co/case-studies/remi).

Note what moved: creative volume, conversion rate, and retention, on the same scoreboard. A single buyer hire fixes one of those. Your own figures will differ with margin, AOV, and repeat rate in your category.

When this does not apply

• Pre product-market fit. If you don't know which offer sells, hire nobody and retain nobody. Make ads yourself, cheaply, until something works.
• Thin margin with no repeat purchase. If gross margin can't fund a CAC and customers buy once, staffing doesn't solve it. Pricing or the offer does.
• Spend too low to justify a senior specialist. A full-time buyer on a small budget is an expensive way to press buttons.
• The product is the constraint. Stockouts, shipping times, and weak product pages aren't a staffing question.
• You already have a strong internal creative team. Buy buying or retention, not the whole package.