When a media buyer is worth it (and when they are not)

A good media buyer multiplies capacity. A mediocre one is an expensive way to outsource your confusion.

Short version: A media buyer is worth it when spend is high enough that small efficiency gains pay for the seat, creative supply exists, stock and offers are stable enough to learn from, and you are the bottleneck on account hygiene. Wait if you cannot explain the offer in one sentence, inventory is chaos, or you expect them to be strategist, copywriter and therapist. Keep offer, margin and stock policy; let them own structure and testing.

Signs you actually need one

• Spend is high enough that small efficiency gains pay for the seat
• You already have creative supply, not a blank Figma
• Stock and offers are stable enough to learn in-platform
• You are the bottleneck on day-to-day account hygiene

Signs you should wait

• You cannot explain the offer in one sentence
• Inventory is chaos and you want ads to “create demand” for dead sizes
• You expect them to also be strategist, copywriter, and therapist
• Monthly spend is still in “learning what works” territory and you have not fixed landing pages

Founders often hire to escape the feed. The feed follows you into the hire if the system underneath is mush.

A healthier split

Keep ownership of offer, margin, and stock policy. Let a buyer own structure, testing cadence, and platform craft. Meet on a fixed review. Kill vanity reporting.

Software will not replace judgement. It can replace the scavenger hunt across tabs so humans argue about decisions instead of data plumbing. That is the honest line between a retainer and an operator stack.

Onboarding a buyer without chaos

Give them offer rules, margin floors, and stock policy on day one. If you cannot write those, you are not ready for the hire: you are ready for a clearer business.

Scorecard for the first 30 days of a hire

• Account hygiene improved (naming, structure, waste cut).
• Tests have a hypothesis and a kill date.
• They escalate stock conflicts instead of ignoring them.
• Reporting leads to decisions, not show.

If none of that shows up by day 30, you hired hope. End it cleanly.

What good looks like in week one

They ask about margin and stock before they ask for admin access. They clean naming. They do not rebuild the account for sport. If week one is only mood and big promises, believe them, just not in the way they hope.

How to hire and onboard a media buyer without chaos

The hire is a workload and skill decision, and it goes wrong in predictable ways. Here is how to check you need one, then run the first month so you can actually tell.

1. Check it is a capacity problem, not an offer problem. A buyer improves execution. They cannot fix a weak offer, thin margin, or a product people do not want. If your conversion rate and repeat rate are poor, hiring someone to buy more traffic makes the problem more expensive. Fix the offer first.
2. Write the job as three tasks, not a title. For example: own daily budget decisions across Meta and Google, brief and ship eight creative concepts a month, and produce a weekly decision log. Specific tasks let you evaluate candidates and, later, tell whether it is working. "Own paid" tells you nothing.
3. Test with a paid task rather than a portfolio. Give shortlisted candidates read access to a real account for two hours and ask what they would change and why. Pay them for it. You will learn more from that than from any case study, because case studies are written after the fact by whoever had the best month.
4. Pressure-test how they handle constraints. Ask directly: what do you do when the best performing product has four days of cover left. A good buyer talks about pulling it from prospecting and protecting retargeting. A weak one talks about scaling the winner. This single question sorts candidates faster than anything else on the call.
5. Give them the constraints in writing before day one. Margin floors, the SKUs that must never be paused, buffer rules, brand no-go areas, and who approves spend increases. Handing someone an account without these and then criticising their judgement is a management failure, not a hiring one.
6. Make week one about access and a written audit. No structural changes in the first week. Access confirmed, tracking verified, and a written account audit including the uncomfortable parts. If someone rebuilds your account on day two without asking why it is like that, that is a warning, not initiative.
7. Score the first 30 days on process, not on ROAS. Thirty days is not enough to judge results, and it is plenty to judge process. Did the decision log get written every week. Did creative actually ship. Were bad results reported without being asked. Those predict month six far better than a fortnight of ROAS.
8. Decide the split with software and stick to it. Buyers are worth their fee on judgement, creative direction and hard calls. They are expensive for monitoring and preparation. Decide which parts of the job are theirs and which belong to tooling before they start, so nobody is paying senior rates to read a stock report.

If the honest answer is that you need judgement occasionally rather than daily, a fractional buyer plus better tooling usually beats a full-time hire. The economics are in what a retainer costs.

If you want stock, ads and the morning list in one place: that is what we built Ralph for: a growth operator for Shopify, with review before anything ships. Not magic. Fewer tabs. See Ralph · Docs