Optimise retargeting windows for purchases: Use GA4 to track `view_item` to `purchase` events with user IDs; Calculate elapsed days between qualifying visit and purchase for first-time buyers; Find cumulative purchase share by day to set the ideal retargeting window
Image: Paid Social Guide

Audience Targeting

Part of Retargeting through social advertising

Choosing a retargeting window from the purchase cycle

A retargeting window is the time a past action remains eligible for a reminder. Choose it from the buyer's decision period, not from the longest setting a platform …

A retargeting window is the time a past action remains eligible for a reminder. Choose it from the buyer's decision period, not from the longest setting a platform offers. A week may be too late for an urgent service and too early to judge a costly considered purchase.

Look at the observed delay between a qualifying visit and a completed purchase for the relevant product category. Separate first-time and repeat buyers, then chart the cumulative share of purchases completed by each day after the visit.

Google Analytics 4 (GA4) uses an event-based model: view_item, add_to_cart and purchase are separate ecommerce events, and actions can be tied to the same user ID. Pair the event that defines your qualifying visit with purchase for the relevant product category, where your implementation records the ID.

For each matched purchase, calculate the elapsed days from the qualifying event. Group purchases by elapsed day, then divide the cumulative purchases up to each day by the cohort's total purchases to see the cumulative share.

Choose a coverage target that reflects the trade-off between reaching more of the decision period and extending the reminder period; set the window at the earliest day that reaches that target. Check whether the curve starts to flatten, which can indicate that later days add relatively few purchases. The resulting day count comes from your observed curve, not a universal setting.

For high-ticket electronics and tech, 45–90 days is a useful comparison range. Treat it as a sense-check, not a recommendation: your window should be the earliest day your own cumulative purchase curve reaches its chosen coverage target.

Test shorter and longer boundaries when the audience size allows a comparison, changing one boundary at a time. Define the events precisely: entry on a product view or cart action, exit on a recorded purchase, expiry at the end of the window.

If a buyer purchases through another channel and the platform does not receive that event, the exclusion may miss them. State that limitation in internal reporting.

Meta audience windows can range from 1 to 180 days, depending on audience type. The official Postman collection is named Facebook Marketing API Targeting. Check the current platform settings for eligible signals and retention controls, which may change.

Before deploying a tracking pixel, consider what information it collects, how it will be used and what people are told. The Office of the Australian Information Commissioner (OAIC) has guidance on tracking pixels and privacy obligations.

Purchase Cycle Timeline: From Qualifying Visit to Purchase

  • Day 0Qualifying visit (e.g., view_item or add_to_cart)
  • Day 7Cumulative purchase share begins to rise for low-cost items
  • Day 30Mid-point for many considered purchases; e.g., electronics, services
  • Day 45–90High-ticket items often complete purchase within this window
  • Day 180Maximum retargeting window available in Meta platforms

Key Retargeting Window Insights (Australia)

GA4 Event-Based Tracking
Uses `view_item`, `add_to_cart`, `purchase` with user ID linking
Privacy Compliance
OAIC guidance required for tracking pixels

More from Audience Targeting