Configuration recommendation for attribution windows




What is an “attribution window” and why is it important?

An attribution window (sometimes called a conversion window) is the fixed period following an ad click or impression during which any conversion can still be attributed to that touchpoint.

For example, if your attribution view is set to a seven-day post-click window, a purchase made up to seven days after someone clicked on the ad will be attributed—fully or partially—to that click.

Choosing the right window is crucial for an attribution platform because it defines the time interval that the system considers a cause-and-effect relationship.

Too short a window underestimates high-funnel or high-consideration campaigns; too wide a window overestimates obsolete impressions, inflates ROI, and sometimes doubles attribution across channels.
By calibrating windows to realistic buying cycles and impression quality, your attribution view can more accurately measure incremental contribution, direct budgets more intelligently, and ensure fair evaluation across all channels.


1. Fundamental principles

Principle
Why does it matter?
Implication
Align the window with the decision cycle duration
Complex purchases take weeks or months to complete, while impulse purchases take hours or days to complete.
Base it on the average purchase time in the sector, then adjust the window around this reference.
Differentiate between post-click and post-impression
Clicking shows explicit intent, so it deserves a longer analysis period than passive exposure.
The post-click window should be ≥ the post-impression window (often 3 to 5 times longer).
Weight by print quality
“Real views” (≥ 50% of pixels visible for ≥ 1 second for display; IAB/MRC standards) have a greater impact on memory and conversion.
Use longer windows for visible prints and shorter ones for non-visible prints.


2. Recommended attribution windows by industry

Indusy
Conversion cycle
Post-Click
Post-Impression Visible
Post-Impression
Travel / Tourism (B2C) – flights, hotels, OTAs
2–6 weeks
30 days
7 days
1 day
Food / Mass distribution (B2C)
Same day to 48h
1 day
12 hours
1 day
Fashion / Beauty (B2C)
3–7 days
7 days
1 day
6 hours
Specialized retail (B2C) – electronics, sports, home
1–3 weeks
14 days
3 days
1 day
Culture / Entertainment (B2C) – tickets, streaming
A few hours to 1 week
3 days
1 day
6 hours
Daily services (B2C) – energy, banking, insurance
3–6 weeks
30 days
7 days
1 day
Automotive (B2C) – leasing, testing
2–3 months
60 days
14 days
3 days
Telecoms (B2C) – mobile plans, internet
1–3 weeks
14 days
3 days
1 day
Education / Training (B2C) – online courses
1–2 months
30 days
7 days
1 day
Health / Well-being (B2C) – teleconsultation, products
A few days to 1 week
7 days
1 day
Same day
DIY / Garden / Decoration (B2C)
1–3 weeks
14 days
3 days
1 day
Consumer electronics (B2C) – smartphones, household appliances
2–4 weeks
21 days
3 days
1 day
Gaming (B2C) – games, consoles, subscriptions
A few hours to 1 week
7 days
1 day
6 hours
Travel / Tourism (B2B) – professional tools
2–3 months
60 days
14 days
3 days
SaaS / Software (B2B)
3–6 months
90 days
30 days
7 days
HR Services (B2B)
2–4 months
60 days
14 days
3 days
Professional Training (B2B)
2–4 months
60 days
14 days
3 days
Events / Trade Fairs (B2B)
4–8 weeks
30 days
7 days
1 day
Finance / Insurance (B2B)
3–6 months
90 days
30 days
7 days
Logistics / Transport (B2B)
2–4 months
60 days
14 days
3 days


3. Application & Window Optimization

  • Start with recommendations by industry.
  • Monitor conversion time curves : if ≥ 90% of conversions occur in the first half of the window, shorten it; if ≥ 20% occur after, lengthen it.
  • A/B test the length of attribution windows to understand the impact in terms of performance transfer between channels.
  • Document changes for effective learning history.


4. Sector nuances

  • Seasonal peaks (Black Friday for fashion, summer for travel) shorten conversion cycles
  • High-end products in a sector (luxury fashion, premium electronics) require windows close to specialized retail.
  • Subscription (SaaS) models often have immediate trials but delayed conversions.
  • Assisted offline conversion (test drive, bank appointment) involves additional delays.

By anchoring attribution windows to realistic purchase journeys and weighting them by impression quality , you improve reporting accuracy – valuing upper-funnel media while limiting bias from low-value impressions.