Careers
Part of The honest guide to marketing attribution
Common marketing attribution questions, answered
Common marketing attribution questions answered: models, platform settings, privacy, and when to use incrementality or marketing mix modeling.
What to take away
- Attribution is a controlled crediting system, not automatic causal evidence.
- Model, container, lookback, and outcome definitions must travel with the number.
- Different reports can be internally correct while answering different questions.
Common marketing attribution questions arise when teams see different channel totals for the same business period. The quickest resolution is to compare definitions and settings before debating performance.
The essential answers
| Question | Working answer |
|---|---|
| What gets credit? | Eligible recorded touchpoints within a stated path |
| What is credited? | A defined conversion or value at a stated maturity |
| Can credit be fractional? | Yes, when a model distributes one outcome across touchpoints |
| Does credit prove impact? | No; incrementality needs a counterfactual |
| Who owns the model? | A named business owner with data and method stewards |
| When is it reviewed? | When decisions, data, products, cycles, or methods change |
Common attribution models answer the credit-allocation question in different ways:
- Last-clickassigns credit to the last eligible touchpoint; first-click assigns it to the first.
- Lineardistributes credit across eligible touchpoints; time decay gives more credit to touchpoints nearer the conversion.
- Position-basedgives selected positions, such as the first and last touchpoints, more credit; data-driven uses a product's method and observed data to allocate credit.
For example, if someone sees a social ad, later clicks a search ad, and then converts after returning directly, the models may credit different touchpoints. None of those allocations alone proves which touchpoint caused the conversion. Choose a consistent model for reporting, and use an incrementality method when the decision depends on causal lift.
Adobe's attribution components guide separates model, container, and lookback window and illustrates how credit can be distributed within its product. Those product definitions do not establish causal contribution.
Why platforms differ
Campaign Manager 360's attribution overview describes product-specific conversion paths, models, windows, and reporting behavior. It helps explain the platform's crediting logic, not the full customer journey outside its coverage. Coverage gaps of that kind are what web analytics for marketers works around when a visit continues outside the tracked property.
Common tools include Google Analytics 4, Google Ads, Meta Ads Manager, and Adobe Analytics. They are not interchangeable: analytics products report configured site events and paths, while ad managers attribute conversions under their ad-interaction eligibility and account settings. Check the model, conversion definition, lookback window, and view-through eligibility in the specific report; there is no shared set of product defaults.
Reconciliation Checklist
- Compare conversion action and counting rule
- Align event time and timezone
- Check eligible networks, devices, offline uploads
- Inspect identity and deduplication coverage
- Match model, container, and lookback window
- Wait for equivalent processing and maturity
If totals still differ, preserve both with clear names. Reconcile each to the source it claims to represent. A finance order total, analytics event total, ad-platform attributed total, and experiment estimate should not be forced into artificial agreement.
Ask a second reviewer to explain one channel's number without opening the original analyst's notes. If that reviewer cannot identify the outcome, model, window, population, and limitation from the report itself, the view is not ready for a material decision. The same readability test decides whether marketing dashboards work without their author in the room.
Resolve the question with a decision record
A useful answer identifies the decision owner, eligible population, evidence period, method, and action. It also states what the result cannot establish. Save the definition and date with the answer. If the same question returns, compare the records before changing the metric. That step exposes a changed population or source instead of treating every difference as new behavior.
For a budget decision that requires evidence of incremental lift, consider a holdout test or a geo experiment. A holdout test compares outcomes for an eligible group exposed to marketing with a group withheld from it; a geo experiment compares outcomes across selected markets. The design and its limitations matter, so do not treat a platform's attributed conversions as the experiment result.
Marketing mix modeling (MMM) is another complementary method: it analyzes spend and outcome history for budget planning rather than assigning person-level credit along a tracked path. Google Meridian and Meta Robyn are named MMM tools.
The Interactive Advertising Bureau (IAB) and the Media Rating Council (MRC) publish standards and guidance relevant to digital advertising measurement. These can support consistent definitions and measurement practices, but they do not make platform reports equivalent or establish that attributed credit is causal.
Privacy rules also affect which data can be collected and used for attribution. The General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA), as amended by the California Privacy Rights Act (CPRA), may apply depending on the organization and data involved. Review applicable notice, consent, opt-out, vendor, and retention obligations; these laws do not prescribe an attribution model.
Keep the evidence record beside the decision so a reviewer can reproduce the reasoning. Set the next review date and name the change that would trigger an earlier check. Where credit rules cannot settle a budget question, marketing mix modeling answers it from spend and outcome history.
Common questions
Can direct traffic receive attribution credit?
The answer depends on the product and report. Google Analytics 4 can record direct/(none) when usable source information is unavailable, while its attribution reports may assign credit based on the selected model and eligible paths. Google Ads and Meta Ads Manager attribute conversions to eligible ad interactions under account settings; a direct visit by itself is not an ad interaction. In Adobe Analytics, direct handling depends on the report's attribution model, marketing-channel rules, and lookback window. Record each product's rules before comparing totals.
Are view-through and click-through conversions equivalent?
No. A view-through or engaged-view conversion can be credited after an eligible ad exposure without a click, under the product's rules. Google Ads and Meta Ads Manager use product-specific eligibility and windows; GA4 reports do not automatically reproduce those ad-platform totals, and Adobe Analytics depends on the impression data and attribution rules configured in the implementation. Record each product's displayed view-through or engaged-view window separately from its click-through window, and report the two types separately.
What should appear beside every attribution number?
Show outcome, population, source, period, model, container, lookback, time basis, freshness, and material limits.
