Dema

Measure what your marketing actually adds.

Run geo tests that change spend by region, compare against what would have happened without it, and read the lift in sales and profit.

  • Test and control regions matched
  • Lift in sales and profit
  • Results feed your attribution

Geo test, Meta, Nordics

Started 24 Aug Ends 18 Oct

Net sales

Lift in net sales

+18%

Test regions vs without ads

Confidence

95%

Interval excludes zero lift

Incremental revenue

€65K

Over 8 weeks

Used to calibrate attribution

×0.60

Meta's reported sales

Used by the teams at

  • Acne Studios
  • Scuffers
  • NOTHS
  • Ridestore
  • Axel Arigato
  • Adlibris
  • Osprey London
  • Villoid

Start with the comparison

A deliberate change.
A clearer answer.

Use the pre-test data to build comparable treatment and control groups. Change spend in the treatment regions while preserving the comparison.

Estimate the outcome that would have occurred without the intervention, then assess the difference and its uncertainty.

A controlled change
Field overshirt campaign creative
Core collectionMeta · Prospecting
Regional test
Treatment

Change spend

Control

Keep the plan

Compare outcomes across comparable regions.

Design around the decision

Different questions.
Different experiments.

Investigate an existing channel, try a new one or test a change in spend. Each design answers a different question.

Explore the examples to see how a holdout and a scaling test create different comparisons.

Choose the question before the test

What does this spend contribute?

Pause ads in treatment regions

Spend changesIllustrative holdout experiment: pause ads in treatment regions. Observed outcomes are compared with an estimated counterfactual after the intervention.
Before the testAfter the change
Observed outcomeEstimated counterfactual

Estimates depend on test design, statistical power and uncertainty.

Go beyond the headline

Read the whole result.

Timing, business outcomes and uncertainty all matter. A single uplift number is only part of the evidence.

Give the effect time.

Plan the treatment and post-treatment windows before reading the result.

Give the effect time to arrive
BeforeCheck comparability

Use the pre-test period to assess the design.

DuringChange the spend

Keep treatment and control conditions clear.

AfterLet outcomes mature

Include the lag before finalizing the result.

Follow the economics.

Read sales alongside contribution and new-customer outcomes.

What you keep
€20,000Contribution after advertising

Keep the uncertainty.

Use the range around the estimate to judge what the test can support.

Read the estimate and its range
+12%Estimated lift
−10%0%+30%
Lower estimate
+3%
Upper estimate
+21%

A point estimate is more useful when you can see its uncertainty.

Questions

A good test starts with good questions.

Bring one commercial job

Find out what the spend really adds.