To compare media buyers side by side on the same accounts, give them the same business goal, measurement rules, decision window and comparable account scope. Judge each buyer against agreed targets and controllable decisions, then use the comparison to share working methods, explain differences and improve the whole buying team.
That is the foundation. If one buyer gets the strongest markets, another inherits weak tracking and a third is measured against a different attribution window, the table may look precise while answering the wrong question.
Set the comparison rules before looking at results
A fair comparison starts with a written operating agreement. It does not need to be long. It needs to make clear what counts, when it counts and what each buyer controls.
Use this checklist before the evaluation period begins:
| Comparison rule | What to agree on | What can distort the result |
|---|---|---|
| Business goal | Leads, purchases, revenue or another defined outcome | Buyers optimizing toward different outcomes |
| Primary KPI | CPA, ROAS, link CPC or another stated measure | Comparing one buyer on CPA and another on platform ROAS |
| Target | The account target plus any campaign exceptions | Applying one blended target to campaigns with different economics |
| Attribution | Source of record and attribution settings | Mixing platform-reported and first-party results |
| Evaluation period | Start date, end date and planned review points | Giving one buyer more time or a better seasonal window |
| Account scope | Campaigns, markets, platforms and budgets included | Comparing a mature market with a new launch |
| Decision rights | What the buyer may change and what requires approval | Scoring someone on decisions outside their control |
| Constraints | Offers, brand rules, exclusions and budget limits | Treating a required constraint as buyer error |
The target hierarchy matters. An account can have a fallback target while specific campaigns use their own CPA or ROAS targets. Groups of campaigns can also share a rule when they serve the same job.
Example: Say an account has a $50 target CPA. A high-value campaign may justify a different CPA target, while a group of TikTok campaigns may be judged on link cost per click because they are supporting a different stage of the buying process. The comparison should follow those stated rules instead of forcing every campaign into the $50 account target.
Put the exceptions in writing before reviewing performance. Otherwise, every exception starts to look like an excuse after the fact.
Build a scorecard around the work buyers control
To compare media buyer performance, separate business results from operating quality. Both matter, but they answer different questions.
Business results show what happened. Operating measures show how the buyer managed the account, responded to evidence and protected the plan.
Start with the primary result
Choose the metric closest to the agreed goal:
- Use CPA when the buyer is accountable for a defined conversion with roughly comparable value.
- Use ROAS when reliable revenue is available and revenue is the operating target.
- Use MER when leadership needs a combined view of total revenue against total media spend.
- Use a campaign-group KPI, such as link CPC, when that group has a distinct job and target.
Do not reward a lower CPA automatically if the leads have different value. Do not reward higher platform ROAS automatically if the buyers are working from different attribution settings. Define the source of truth first.
Add a cost and efficiency grid
This grid keeps the calculation visible and prevents a blended headline from hiding what changed.
| Measure | Calculation or source | Why it belongs in the review |
|---|---|---|
| Spend | Ad account delivery | Shows the budget actually deployed |
| Planned spend | Approved media plan | Gives pacing a clear reference |
| Conversions | Agreed source of record | Keeps the denominator consistent |
| CPA | Spend divided by conversions | Measures cost against the defined action |
| Revenue | Agreed revenue source | Prevents different revenue definitions |
| ROAS | Attributed revenue divided by spend | Compares campaign return where revenue is reliable |
| MER | Total measured revenue divided by total media spend | Gives leadership a cross-platform view |
| Tracking exceptions | Recorded gaps or changes | Shows where the numbers need qualification |
The point is not to load the scorecard with every available column. The point is to make each number interpretable.
Review decision quality
Add a short decision record beside the metrics:
| Review item | Practical check |
|---|---|
| Pacing | Did spend land close to the approved plan? |
| Target discipline | Were campaign targets and account fallbacks applied correctly? |
| Change discipline | Were material changes recorded with a reason? |
| Test quality | Did the test isolate a useful question? |
| Evidence | Was the decision supported by current account data? |
| Restraint | Did the buyer leave a working campaign alone when the evidence did not support a change? |
| Follow-through | Was the effect of an approved change checked later? |
This makes the review useful even when two buyers finish with similar topline results. One may have found a repeatable method. The other may have benefited from favorable conditions that will not repeat.
Design a comparison that does not create account conflict
“On the same accounts” should not mean several people changing the same campaign at once. That destroys the test because no one can tell which decision caused the result.
Use one of these structures instead.
Assign clear campaign ownership
Give each buyer a defined set of campaigns within the same account or account group. Match the sets as closely as practical by objective, budget, market maturity, offer and conversion path.
For a franchise, that might mean each buyer manages a group of locations with similar business conditions. Record the differences that remain. Comparable does not mean identical.
Use controlled tests for specific decisions
When the question is narrow, test the decision rather than the person.
Example: One buyer proposes new copy for an existing Google ad group. The clean comparison is a test that runs the new version beside the original, with the result reviewed at preset checkpoints. If the volume is too low to support a call, mark it inconclusive and state why.
An inconclusive result is not a buyer failure. It is a statement about the available evidence.
Rotate ownership with a formal handoff
Rotation can work when campaign splitting is impractical. Capture the starting condition, active tests, recent changes, approved constraints and open tracking issues before ownership moves.
Do not compare the incoming buyer’s first period with the outgoing buyer’s settled period without noting the handoff. The incoming buyer may be dealing with learning resets, expired offers, unfinished tests or budget changes that were already in motion.
Keep one change log
The change log should answer five questions:
- What changed?
- Why was it changed?
- Who approved it?
- When was the result checked?
- What happened afterward?
A buyer should be able to explain a decision in plain language without reconstructing it from platform history. The log also protects the buyer when an approved business decision produces a weak result.
Teams evaluating PPC management software for agencies should treat this record as a core operating requirement, not a reporting extra.
Read the scorecard as an operating tool
A side-by-side view should help leadership ask better questions. It should not become a leaderboard with the context removed.
Start with the largest meaningful differences and work backward:
- Did the buyers have the same target?
- Were the accounts and offers comparable?
- Did one group have a tracking interruption?
- Did spend pace differently against plan?
- Was one buyer working within tighter constraints?
- Did a completed test explain the performance gap?
- Was the apparent winner driven by one campaign or a repeatable pattern?
Then separate three types of finding.
A method worth sharing
If one buyer consistently frames clearer tests, catches pacing issues sooner or documents stronger reasons for changes, turn that method into a shared operating rule.
The goal is not to make every buyer use identical tactics. It is to spread the parts of the process that produce clearer decisions.
A condition that needs context
A market, offer, landing page or sales process may explain a performance difference. Record it as context instead of assigning it to buyer skill.
This is especially important for franchises. Local economics can differ even when the brand, platform and campaign structure look similar.
A decision that needs follow-up
Some decisions cannot be graded at the first review. Mark them open and set the next check. Do not force a verdict because the meeting is on the calendar.
This is also how agencies can scale account oversight without turning every review into manual reconstruction. The related guide on scaling a PPC agency without hiring a media buyer for every new client covers the broader operating model.
When you compare media buyer performance this way, the output is not just a ranking. It is a clearer account plan, a record of what the team has learned and a fair basis for the next decision.
Frequently asked questions
How do you compare media buyers fairly?
Give each buyer the same business goal, KPI definitions, attribution rules, evaluation period and level of control. Compare matched accounts or campaigns, then document differences in budget, offer, market and tracking before drawing conclusions.
What metrics show a good media buyer?
Use the metric tied to the business goal, such as CPA or ROAS, together with pacing, tracking quality, change discipline and the results of completed tests. No single metric can explain performance without the target and operating context.
Should different buyers run the same accounts?
They can work within the same account structure, but they should not make competing changes to the same campaign at the same time. Assign clear campaign ownership, use controlled tests or rotate responsibility with a recorded handoff.
How do franchises manage several agencies?
Set common KPI definitions, naming rules, tracking requirements and reporting periods across every agency. Keep local targets where economics differ, but roll the results into one portfolio view so the brand can see both local and combined performance.
How long should you test a media buyer?
Run the comparison long enough to cover the normal buying cycle and collect enough conversion data to evaluate the agreed target. Set review dates before the comparison starts, and mark the result inconclusive when the data cannot support a fair call.
What Ads Arsenal does for this problem today
Ads Arsenal lets a business give each media buyer a property and compare those properties side by side in a portfolio. One property can cover a group of lookalike accounts. Buyers can use an account-level target as the fallback, set campaign-level CPA or ROAS targets where needed, and apply one KPI rule to a group of campaigns.
Ads Arsenal is in a free design-partner beta for agencies and in-house teams running real spend on Google, Meta and TikTok. Become a design partner.