Attribution Window

Attribution windows come from digital advertising, where a click and a purchase join directly. Restaurants have no such link, since nobody clicks their way to a table. The hospitality version is a timing rule applied across two systems: the message log on one side, the reservation book and the check on the other. Only a platform holding both can apply it, so the setting belongs to a restaurant CRM rather than to an email tool.

How long should an attribution window be?

Thirty days is the usual default and a reasonable starting point. Loyalist’s default is 30 days. The right length depends on the decision cycle the message is trying to influence, and two cases pull in opposite directions.

A tonight-only offer to fill a quiet Tuesday should be measured in days. Crediting it with a visit three weeks later attributes a booking that had nothing to do with it. A save-the-date for a holiday menu works the other way, because the whole point is that the guest books later. A 7-day window on that campaign will report near zero and the campaign may have filled December.

The important discipline is consistency. Comparing a campaign measured over 7 days against one measured over 30 is not a comparison, and changing the window mid-year quietly rewrites the trend. Pick a default for routine reporting, then vary it deliberately for individual campaigns and label it when you do.

What does a restaurant campaign get credited for?

A campaign gets credited with bookings, visits and revenue, and the credit runs through the guest rather than through a link. The message went to an identified guest, that guest appeared on the book or on a check, and the appearance fell inside the window.

There is a step in front of that worth stating plainly, because it changes how the number should be read. Credit typically begins from an engagement signal, an open or a reply, so a guest who never interacted with the message is generally not counted even if they came in the next day. Attributed revenue is therefore a subset of two things: guests who engaged, and guests whose visit could be matched to a profile at all. A venue that matches 60% of its checks is measuring campaign results on 60% of the evidence.

What does an attribution window not tell you?

An attribution window does not establish cause. It establishes coincidence inside a chosen period. That is a useful proxy and not the same claim. The regular who eats at the restaurant every Thursday will land inside any window applied to any message sent to them.

The window length therefore has an ethical edge as well as a technical one. Widening it always increases attributed revenue and never requires better marketing. Separating what a campaign coincided with from what it caused takes withholding the message from part of the audience and comparing the two groups.