Digital transforms the mechanisms of the mix, not only the media.
Evidence status
What was reproduced, and what it does not prove.
The CSV and Python verify the incremental revenue and the illustrative contribution. They prove neither real incrementality nor causality.
- Technical specification of the digital mix matrix, specification in French (opens in a new tab)
v1, 31 August 2026 · CC BY 4.0 · specification in French · Method document written by marketing-mix.net. The figures are illustrative: they are not market measurements.
Key distinctions
Three conditions before calculating.
Unit economics before platform metrics
Clicks, installs and leads are not economic outcomes. Every journey must be traced back to net price, variable margin, cost to serve, return rate and future value. A positive advertising ROAS can hide a loss-making cohort if discount, logistics or churn are excluded.
Loops rather than functional silos
Search influences acquisition, but also the architecture of the offer; the product influences conversion and retention; CRM changes frequency and marginal cost. Mapping these loops avoids paying two teams for the same sale, or removing a lever that feeds another.
Data as a conditional asset
Data is only worth something if it improves a decision. Consent, identifier, matching quality and time to availability determine that value. Collecting more without a usage rule mainly increases cost, legal risk and false signals.
Method
Designing a decision-oriented digital system
- 01Define the economic behaviours
Start from the behaviours to change — discover, try, buy, upgrade, renew or recommend — and for each write the expected contribution and the horizon; this step prevents confusing digital activity with value created.
- 02Assign a role to each lever
Qualify paid acquisition, content, product, pricing, distribution, onboarding and CRM by primary role, mechanism and dependency; a lever can have several effects, but only one outcome should drive its budget decision.
- 03Calculate the full cost of the loop
Add media, discount, commission, creative, technology, fraud, returns and service; deduct transfers from other channels; compare incremental contribution rather than revenue attributed by the last interaction.
- 04Install a learning cadence
Pair each loop with a descriptive measure, a periodic causal proof and a reallocation rule; daily signals serve execution, while experiments and cohorts govern changes of level.
The economic map of the digital mix
Each lever is linked to the targeted behaviour, the full cost, the data created and the next decision.
A profitable campaign turns loss-making once the full loop is counted
A brand spends €120k on paid social and is attributed €480k of sales, a ROAS of 4. Gross margin is 55%, the average discount 12%, returns 14%, and service costs €9 across 2,400 orders. A holdout test estimates that 62% of the sales are incremental.
Incremental sales: 480 × 62% = €297.6k. After discount and returns, retained revenue is 297.6 × 88% × 86% = €225.2k. Gross margin: €123.9k. After €21.6k of service and €120k of media, contribution is −€17.7k. The attributed ROAS of 4 becomes an incremental contribution ratio of −0.15.
The channel is not cut automatically: targeting, discount, promise and onboarding are tested together. The next tranche is funded only if incremental contribution per order exceeds €50.
Acceptance conditions
What must be true to act.
- 01
Every digital metric is connected to a behaviour and to a contribution.
- 02
Discount, return, service and commission costs sit in the same account.
- 03
Sales transferred from other channels are deducted.
- 04
A causal proof regularly calibrates the platform metrics.
Limits
What this analysis does not prove.
- Incomplete identifiers make customer loops only partly observable.
- Brand and network effects often exceed the attribution window.
- An average contribution can hide both profitable and value-destroying segments.
References
Works cited.
- Lemon, K. N. & Verhoef, P. C. (2016), “Understanding Customer Experience Throughout the Customer Journey”, Journal of Marketing.
- Li, H. & Kannan, P. K. (2014), “Attributing Conversions in a Multichannel Online Marketing Environment”, Journal of Marketing Research.
- Berman, R. (2018), “Beyond the Last Touch: Attribution in Online Advertising”, Marketing Science.