Как мыслит Head of Media Buying: кого берут, а кого увольняют

Business-focused summary (Head of Media Buying interview)

Who they hire / what they look for in interviews

• Soft skills first (prioritized during screening)

  • How the candidate communicates and structures explanations.
  • Ability to answer clearly without “mumbling.”
    • Otherwise, it triggers extra clarifying questions and signals higher risk.

• Hard skills second (validated through experience)

  • Concrete ability to “pour” traffic
    • Typically demonstrated via months of experience, volumes, and execution details.
  • Interviewers form follow-up questions based on answers—this is treated as evidence of real understanding.

• Red flags / “dismissal triggers” (manager’s view)

  • Primary trigger: the buyer does nothing when results are bad

    • If you have 2 months in a row (or several weeks) of negative performance → “change everything” (e.g., reassemble funnel, creatives, etc.).
    • A special observation: repeating the same method for ~6 months with no results is a strong negative signal.

  • Secondary trigger: laziness / reduced engagement

    • Buyers who “relax” after making money often experience result decay.

  • Instant dismissal triggers (fraud / deception / serious waste)

    • Spending money on left-wing sources/links.
    • Deliberate misreporting (e.g., claiming a launch occurred when it didn’t).
    • Large unexplained losses (example: $100,000 wasted in a referenced case).

---

How they motivate buyers to avoid “laziness”

• Not one-size-fits-all

  • For some buyers, strict pressure / scolding works.
  • For others, pressure demoralizes, so the approach must be individualized.

• Effective alternative motivation method

  • Set the buyer’s personal goal (e.g., “how much money to earn for a car”).
  • Translate that goal into a step-by-step production plan:
    • expected volumes (what needs to be poured),
    • approximate profit,
    • what daily funnel execution looks like.

• Practical anti-procrastination process

  • Break work into smaller steps—“get up from the sofa” + begin with a small physical/action step (analogy: preparing a backpack/training kit).

---

Career ladder / becoming independent (buyer → team lead)

• Core principle

  • You become capable through 24/7 systematic work + practice.

• Testing and iteration are mandatory

  • New buyers/assistants shouldn’t be afraid to spend small amounts—testing is the learning loop.

• Speed-to-independence varies

  • Some become independent in ~3 months.
  • Others may need ~1 year.
  • Talent matters, but execution (focus + desire + time invested) dominates.

---

Team strategy: internal leadership pipeline + knowledge sharing

• They prefer to grow leaders internally, not hire only from outside.

• Team design example

  • “Head / CMO” above team leads.
  • Each team lead manages multiple buyers.
  • Example structure mentioned: team leads had 2 buyers each, with designer/technical functions under them.

• Knowledge sharing vs isolated work

  • They share stats across the team (not fully isolated by CPA/stream).
  • This helps quickly determine whether a problem is:
    • global (offer/funnel issue), or
    • local (one buyer’s execution).
  • Rationale: shared analytics prevents “everyone fails except me” blindness and reduces wasted spend.

---

Infrastructure playbook (what “good ops” look like)

They frame infrastructure as a way to accelerate performance through speed and diagnosis.

Key components

• Fast technical support for media buying issues

  • Quick answers for tracker/PVA/app/domain/setup problems.
  • Preferably includes rapid turnaround and sometimes weekend responsiveness.

• Correctly working tracker

• Transparent analytics

  • Purpose: identify bad flows, stop them, then scale good ones.
  • Must be implemented so buyers can use it easily (e.g., fast loads, clear breakdowns).

• Flexible testing environment

  • Ability to vary:
    • PVA
    • OS / iOS / PVAs
    • traffic sources (multiple setups/streams)

Operational expectations

• Fast design and creative production

  • Buyers need to test hypotheses immediately.

• Design department requirements

  • Strong understanding of what converts (not just aesthetics).
  • Designers must follow clear technical specs so testing remains valid.
  • Designers may also produce creatives based on trends/observations; regular output can improve conversion quality.

---

Scaling and stop rules (anti-hype, anti-bad-budget expansion)

When to scale a campaign / EO (example decision rule)

• Not based on “first day wins.”
• Require stability:
  • At least 2–4 days consistently with acceptable metrics and reasonable prices.
  • Example criteria:
    • CPC / installs / registrations at “normal” price levels,
    • conversions with ratios that don’t look broken (avoid random spikes like 10 registrations vs 4 depots inconsistently).
• Budget scaling timing
  • They wait for early volatility to settle (noting that audience updates can be weekly/daily).
  • They minimize the risk of “tomorrow it flies off” by observing multiple days.

How they judge traffic/geo/offer stability (not random luck)

• Emphasis on funnel stability and consistent metric ratios—not just one KPI like “deposits only.”
• They address a common mistake:
  • “Cheap deposits” can look good but later fail.
  • They want evidence of consistent economics.

---

Metrics + KPIs mentioned (and their role)

No formal table/targets are provided, but decision logic repeatedly references these KPIs:

• CPC (cost per click)

  • If CPC is too high vs their math, everything downstream breaks.

• Click → Install (Click-to-Inst)

• Install → Registration (Inst-to-Reg)

• Registration → Deposit (Reg-to-Dep)

• Price of each stage

• Deposit counts / deposit price
• Payback (whether traffic economics works for the advertiser)
• Stability of metrics over time
  • Variance and consistency across 2–4 days (sometimes “a few days”)

Their scaling math idea

• They compute whether CPC is realistic based on:
  • expected conversion rates,
  • and the target deposit economics.
• From regs/deposits economics, they infer allowable CPC.

---

Funnel troubleshooting (root-cause analysis)

They use stage-based diagnosis.

If the issue is early in the funnel

• Bad CPC → likely Facebook targeting/audience (or creatives/design/PVA setup).
• Bad install/register → likely creatives / design / PVA / domain / app launch flow.
• Bad reg/deposit → likely deeper funnel elements (offer, offer presentation, post-install user journey).

Integrated approach

• Multiple variables may be present simultaneously:
  • Facebook audience quality,
  • creatives,
  • PVA/app configuration,
  • domains and tracker signals.
• Method:
  • Find the “cut line” where the funnel starts failing,
  • then isolate variables around that stage.

Avoid blaming the offer too early

• Common claim: “offer doesn’t convert.”
• Response: check earlier stages first:
  • deposit attempts / passability %

---

Testing approach and hypothesis validation

• They accept new hypotheses if they meet “adequacy” criteria:
  • a new creative/funnel idea must be reasonable, explainable, and not “complete trash.”

• Validation often includes adjusting funnel components and running the test.

• “Proof” isn’t required if:

  • the idea falls within known safe boundaries (adequate creative/funnel logic), or
  • it resembles previously successful patterns.

• They frequently refine and reassemble funnel elements based on feedback.

---

Budgeting/execution tactics (FB specifics at high level)

• Budgets: CBO or ABO
• Structured testing is emphasized; they mention avoiding “crooked testing” where analytics per creative doesn’t work reliably.
• Use cases for duplicate strategies
  • Duplicate/take money snapshots during favorable performance.
  • Note: Facebook behavior changes over time; duplicates that work in one period can fail later.

---

Concrete examples / anecdotes (operational learnings)

• Stopping “too good metrics”

  • Cheap clicks/deposits may come from wrong audience:
    • clicks are cheap, but deposit economics fails downstream.

• Scaling based on experienced intuition

  • A top buyer scaled from ~$500 to ~$800 after observing metric timing/behavior.
  • It worked because the “intuition” was grounded in observed patterns.

• Expensive operational mistakes

  • Scaling too aggressively without validating funnel assumptions.
  • Forgetting to stop a campaign, causing unwanted spend (example: PVA/account unbanned midstream and budget flowed incorrectly).

---

Frameworks / playbooks implicitly referenced

• Funnel stage diagnostics

  • Identify the breaking point (CPC vs install vs reg vs deposit) → isolate the variable.

• Test-to-scale gating

  • Require 2–4 days stability before scaling budgets.

• Zero tolerance for stagnation

  • If no improvement after ~weeks/months, “change everything” (funnels/creatives/buying setup).

• Team-wide RCA using shared statistics

  • Determine whether an issue is offer/funnel-wide or buyer-specific.

---

Presenters / sources

• Admatrix — podcast/company host/representative (interviewer)
• Lesha — Head of Media Buying (guest; primary speaker)