Media buyers manage capital whether they realize it or not

A budget is capital.

Every dollar assigned to Google, Meta, TikTok or another channel is a decision to exchange cash today for an expected economic outcome later.

Yet many media teams are trained primarily on efficiency ratios.

They learn ROAS.

They learn CPA.

They learn CTR.

Those metrics describe campaign performance.

They do not by themselves answer:

How much should the business spend?

Where should the next dollar go?

When does profitable scaling stop?

How long can the company wait to get the cash back?

This learning path is designed to close that gap.

Common Thread Collective’s 2026 research and operating material is a useful commercial benchmark because it explicitly connects channel allocation, incrementality and contribution margin. Its Q1 2026 benchmark covers 299 DTC brands and $231 million in spend and illustrates the gap between platform-reported and incrementality-adjusted returns. Haus contributes an experimental perspective on incrementality and diminishing returns.

The Radar curriculum is vendor-neutral.

The goal is to teach the reasoning.

Prerequisite — basic performance metrics

The learner should already understand:

  • revenue;
  • spend;
  • ROAS;
  • CPA;
  • conversion rate;
  • AOV;
  • gross margin.

The path begins when those metrics stop being enough.

Stage 1 — Unit economics

Competency

Build a per-order economic model.

For ecommerce:

Revenue
– COGS
– fulfillment
– payment fees
– discounts
– variable service cost
= pre-ad contribution

Then subtract acquisition cost.

Exercise

Compare two products.

Product A

  • revenue: $100
  • gross margin: 70%
  • shipping and fees: $15

Product B

  • revenue: $100
  • gross margin: 35%
  • shipping and fees: $10

Ask the learner to calculate the maximum CAC if the business requires $15 contribution after ads.

The lesson is immediate:

Equal revenue does not mean equal bidding value.

Checkpoint

The learner can derive break-even ROAS rather than search Google for “good ecommerce ROAS.”

Stage 2 — CAC is not one number

Competency

Distinguish:

  • blended CAC;
  • paid CAC;
  • new-customer CAC;
  • marginal CAC;
  • channel CAC;
  • cohort CAC.

Scenario

CAC rose from $40 to $55 while contribution dollars increased.

Is performance worse?

Not necessarily.

If the business had profitable headroom, higher CAC may be the cost of scaling.

The learner must stop treating lower CAC as universally better.

Stage 3 — LTV and payback

Competency

Understand the difference between eventual value and cash timing.

A customer can be worth $500 over two years and still be dangerous to acquire for $300 today if the company lacks working capital.

The learner should calculate:

  • first-order contribution;
  • 30/60/90-day value;
  • twelve-month value;
  • payback;
  • retention curve.

Exercise

Compare two acquisition channels.

Channel A: CAC $80, 30-day contribution $60, twelve-month contribution $180.

Channel B: CAC $110, 30-day contribution $115, twelve-month contribution $160.

Which channel can scale faster under cash constraints?

Which has higher long-term value?

There may be different answers.

Stage 4 — Contribution margin

Common Thread Collective’s current materials emphasize contribution margin because ROAS ignores costs outside advertising.

That framing is useful.

Competency

Connect media performance to:

  • product cost;
  • fulfillment;
  • discounts;
  • returns;
  • payment costs;
  • ad spend.

Exercise

Give the learner a 4x ROAS campaign.

Then reveal:

  • gross margin 25%;
  • shipping 8%;
  • payment fee 3%;
  • returns 10% of revenue.

Ask whether the campaign is profitable.

The operator should never again treat 4x as self-explanatory.

Stage 5 — Incrementality

Attributed return is not necessarily created return.

CTC’s Q1 2026 benchmark is commercially produced, but it demonstrates the problem clearly: in its dataset, Google branded search had much higher reported ROAS than its incrementality-adjusted return, while some other channels moved in the opposite direction.

Do not memorize the factors.

Learn the principle.

Competency

Understand:

  • platform ROAS;
  • incremental ROAS;
  • incrementality factor;
  • cannibalization;
  • halo effect;
  • geo testing.

Exercise

Channel A reports 6x ROAS but tests at a 0.4 incrementality factor.

Channel B reports 2.5x but tests at 1.2.

Compare incremental return.

Then decide how that should affect budget allocation.

Stage 6 — Marginal return

This is the stage that changes scaling behavior.

Competency

Distinguish average from marginal performance.

A channel can have 4x average ROAS and 1.8x marginal ROAS on the next spend band.

The next dollar should be judged on the second number.

Exercise

Build a simple response curve.

Spend:

  • $25k
  • $50k
  • $75k
  • $100k
  • $125k

Estimate incremental revenue added at each step.

Calculate marginal ROAS.

Then mark the point where marginal contribution becomes unacceptable.

Checkpoint

The learner understands why “we can spend more because ROAS is above target” can be wrong.

Stage 7 — Cross-channel allocation

Competency

Treat budget allocation as ranking marginal opportunities.

Imagine:

  • Google next-dollar iROAS: 3.0x;
  • Meta: 2.6x;
  • TikTok: 2.1x;
  • YouTube: uncertain, expected 2.8x.

The learner should decide:

  • where to deploy capital;
  • how uncertainty affects the decision;
  • how much to reserve for learning;
  • when to rebalance.

A portfolio needs an exploration budget as well as an exploitation budget.

If every dollar goes to the historically best-known channel, new opportunities never earn evidence.

Stage 8 — Forecasting

Media economics becomes operational when it enters a forecast.

Competency

Build:

  • revenue target;
  • spend target;
  • contribution target;
  • new-customer target;
  • channel allocation;
  • expected efficiency.

Exercise

Create three scenarios:

Base Stretch Downside

For each, specify:

  • spend;
  • CAC;
  • contribution;
  • cash requirement;
  • channel mix.

The learner should understand that forecasts are decision models, not promises.

Stage 9 — Daily and weekly operating rules

A finance model that never changes media decisions is not useful.

Competency

Translate economics into operating triggers.

Examples:

  • increase spend when contribution is ahead and marginal efficiency remains acceptable;
  • protect cash when payback deteriorates;
  • move budget when channel incrementality changes;
  • reduce spend when inventory limits value;
  • invest more when creative unlocks new marginal capacity.

CTC’s current “media plan” material is useful because it argues against rigid upfront allocations and instead connects spend to changing business conditions.

That is the right direction.

Failure mode

Hitting the budget because the budget exists.

Budget is a ceiling or plan.

It is not evidence that the money should be spent.

Stage 10 — Finance/media reconciliation

Competency

Run a weekly meeting where finance and media use the same definitions.

The learner should be able to present:

  • revenue;
  • spend;
  • contribution;
  • CAC;
  • marginal return;
  • incrementality evidence;
  • payback;
  • forecast variance.

Scenario

Media says performance is excellent.

Finance says cash is tightening.

Both can be correct.

The operator should identify the bridge: payback timing.

Capstone — allocate a real portfolio

The learner receives:

  • P&L;
  • product margins;
  • inventory;
  • twelve months of channel spend;
  • platform ROAS;
  • incrementality estimates;
  • LTV cohorts;
  • cash balance;
  • creative capacity.

They must produce:

  1. unit economics;
  2. allowable CAC;
  3. channel incrementality adjustment;
  4. marginal-return curves;
  5. total budget recommendation;
  6. channel allocation;
  7. cash-payback model;
  8. downside case;
  9. test budget;
  10. weekly decision rules.

Then the instructor changes one assumption:

  • gross margin falls five points;
  • Meta creative capacity doubles;
  • Google branded incrementality is revised down;
  • cash-payback target tightens.

The learner must reallocate.

A model that cannot react to changed assumptions is not a decision system.

Suggested 10-week sequence

Week 1: Unit economics.

Week 2: CAC and LTV.

Week 3: Contribution margin.

Week 4: Incrementality.

Weeks 5–6: Marginal return.

Week 7: Cross-channel allocation.

Week 8: Forecasting and cash.

Week 9: Operating cadence.

Week 10: Capstone.

Graduation standard

The learner should be able to answer one question without relying on a platform dashboard:

Where should the next dollar go, how much should we be willing to pay for the result, and what evidence would make us change that decision?

That is media economics.

More from Radar