Time to purchase: your retargeting window is probably wrong
Most stores pick a retargeting window out of habit. Your own data has the answer, and it is rarely the number you guessed.
Ask a store how long its customers take to buy and you will usually get a confident answer that nobody has checked. The distribution is easy to measure and almost always surprises people.
What the shape usually looks like
For most stores it is heavily bimodal. A large group buys almost immediately - same session, or within the hour. Then a long tail stretches out over days or weeks. The average sits in the middle and describes almost nobody.
- The immediate group was already decided; the ad found them at the moment of intent.
- The tail group is where consideration actually happens, and where remarketing either works or wastes money.
- The percentile that matters is not the average but the point where nine in ten purchases have happened.
Setting the window from the number
If nine in ten of your purchases happen within fifteen days, a seven-day retargeting window is switching off before a tenth of your revenue arrives, and a sixty-day window is paying to reach people who already bought. The right answer is close to the ninetieth percentile, and it is a number rather than an opinion.
This is also the number that tells you whether a seven-day cookie cap is hurting you. If a meaningful share of your purchases land after day seven, browser-side tracking is losing exactly those customers.
What to do next
Measure the time from first touch to order for last quarter and find the ninetieth percentile. Then compare it to the retargeting windows currently set in your ad accounts. If they disagree, the data wins.
Read next
Touches to purchase: how many times you have to be seen before it works