Media Buyer Interview Questions and Preparation Guide

A media buyer interview usually tests whether you can reason safely with incomplete information. Strong answers clarify the business goal, check measurement, make assumptions explicit, define a controlled action, and explain what evidence would change the decision. Memorizing platform trivia is less useful than showing a reliable diagnostic process.

Prepare examples from real or clearly labeled practice work. Do not reveal confidential data and do not invent a winning campaign. A thoughtful account of a failed test can demonstrate more judgment than an unexplained success.

What the interviewer may be assessing

Questions often cover six areas:

1. campaign and funnel fundamentals;
2. metrics and business economics;
3. diagnosis and optimization;
4. creative or audience testing;
5. operations, access, and communication;
6. compliance and professional integrity.

The expected depth depends on the level. A junior should recognize limits and escalate correctly. A senior should compare trade-offs and improve the operating system. Ask what scope the role owns before assuming the same answer fits both.

A useful answer framework

For a scenario question, use C-C-H-G-D:

• Context: product, market, objective, and time horizon;
• Checks: tracking, data maturity, recent changes, and source of truth;
• Hypotheses: two or three plausible explanations;
• Guardrail: budget, quality, or policy condition that limits downside;
• Decision: next action and the evidence required for another change.

This framework prevents premature optimization. You do not need to recite the letters; make the logic visible.

Fundamental questions

“How do you structure a new campaign?”

A useful answer begins before the ad platform: define the valuable conversion, audience and message hypothesis, measurement, constraints, and decision rule. Then explain how you avoid unnecessary fragmentation while keeping the test interpretable. Mention pre-launch QA and who approves budget and creative.

“How do you know tracking is ready?”

Describe a reproducible check: expected events, test actions, parameters, deduplication where relevant, time zone, consent behavior, and comparison with the source system. State that you will not increase spend while a critical conversion is unverified.

“What metrics do you monitor?”

Connect metric layers: delivery, engagement, conversion, approved quality, and business value. The correct metric depends on the funnel. Explain why a high CTR can coexist with weak customer value and why platform ROAS may differ from financial reporting.

Diagnosis and optimization questions

“CPA increased by 30%. What do you do?”

Do not immediately propose changing bids. Ask: relative to which baseline, over what period, with what conversion definition and volume? Check tracking, spend mix, auction cost, click behavior, conversion rate, approval quality, offer or page changes, and data maturity. Then prioritize the cheapest reversible check and protect the budget with a guardrail.

“When do you stop a test?”

There is no universal number. A responsible answer considers pre-agreed loss limits, minimum operational signal, conversion delay, business risk, and whether the test can still answer its question. Stop immediately for policy, customer-harm, security, or critical tracking issues. Do not continue merely because money has already been spent.

“How do you scale a winning campaign?”

First verify the result: data source, downstream quality, stability, capacity, and creative freshness. Describe staged increases, observation windows, and rollback conditions. Scaling changes the system—audience mix, auction dynamics, operations—so a result at one spend level is not a guarantee at another.

“A platform and internal dashboard disagree. Which is correct?”

Neither should be accepted blindly. Compare attribution rules, event time, time zones, identity, deduplication, cancellations, and reporting latency. Choose a documented commercial source of truth for business decisions while using platform data for its appropriate operational purpose.

Creative questions

“How do you develop test ideas?”

Start with audience research, customer objections, offer evidence, product use, and prior concept performance. Turn an observation into a concept hypothesis. Distinguish concept, execution, hook, and format so one asset does not represent an entire idea.

“How do you identify creative fatigue?”

Avoid treating frequency alone as proof. Look for a pattern across delivery, auction cost, response rate, conversion, audience saturation, and cohort quality. Compare with other concepts and recent changes. Plan refreshes before the portfolio depends on a single asset.

“Tell me about a failed test.”

Use a case where your process is clear. Explain the hypothesis, controls, outcome, limitation, stop decision, and next learning. Do not blame the algorithm, designer, or audience without evidence. Acknowledge what you could have designed better.

Behavioral and operational questions

“Tell me about an expensive mistake.”

Choose a real example you are permitted to share. Describe detection, immediate containment, communication, root cause, and a control added afterward. Avoid minimizing the impact. If you have not managed meaningful spend, use a smaller operational error rather than inventing one.

“How do you prioritize when several campaigns need attention?”

Rank issues by customer or compliance harm, financial exposure, time sensitivity, data certainty, and reversibility. Communicate what is intentionally delayed. Constant context switching is not proof of seniority.

“How do you work with creative teams?”

Describe an evidence-based brief, agreed definitions, a predictable review cadence, and respect for craft. A buyer should provide the problem and learning—not request arbitrary copies of the latest “winner.”

Compliance questions

You may be asked what you would do if a high-performing creative makes an unsupported claim or a manager suggests an evasion tactic. The correct direction is to pause or reject the risky action, preserve evidence, follow the escalation path, and propose a compliant alternative. Performance pressure does not transfer legal or platform responsibility away from the team.

Be cautious if the interviewer treats account bans as routine, asks you to hide the advertiser, or expects personal payment methods and identities to operate business accounts.

Practical assignments

A reasonable take-home task should be time-bounded, use anonymized or synthetic data, and assess job-relevant thinking. Clarify:

• expected time and deliverable;
• whether data are fictional or confidential;
• what assumptions are allowed;
• who owns the work;
• whether the task resembles unpaid production work;
• how your submission will be stored or deleted.

For a case analysis, open with definitions and data checks. Then prioritize observations, hypotheses, and next tests. Do not manufacture certainty from a small dataset.

Questions to ask the employer

• What business metric ultimately defines acquisition quality?
• Which system is the source of truth for performance and bonuses?
• Who approves budgets, creatives, and market expansion?
• What happens when tracking is unreliable?
• How are failed tests reviewed?
• How are platform and legal changes communicated?
• What is the expected creative supply and who owns production?
• How is the first 90 days assessed?
• Which actions require immediate escalation?
• Why is the role open, and what made the previous operating model difficult?

Concrete answers reveal more than promises of “unlimited budgets.”

Seven-day preparation plan

Day 1: map the company’s business model using public, legitimate information; write down what remains unknown.

Day 2: prepare three cases—successful learning, failed hypothesis, and operational problem.

Day 3: review measurement concepts and the current official policies for the stated channel.

Day 4: practice two diagnostic scenarios using C-C-H-G-D.

Day 5: edit your résumé and portfolio so dates, scope, and terminology match.

Day 6: prepare employer questions and your compensation boundaries.

Day 7: run a mock interview, shorten long answers, test links, and rest.

Final checklist

• I can explain every metric and claim in my résumé.
• My cases identify personal and team contributions.
• I can say what I do not know without becoming vague.
• I check tracking and business context before optimizing.
• I include budget, quality, and compliance guardrails.
• I have questions about source of truth, review, and accountability.
• I will not disclose confidential information in a live exercise.
• I can recognize an unsafe or deceptive operating request.

The goal is not to produce a perfect answer instantly. It is to show that the company can trust how you arrive at one.