Regulars

Every operator can name their regulars, and no two operators use the same test. The word works fine in conversation and stops working the moment it has to appear in a report, because a system cannot act on a feeling about who belongs. Any platform reporting regulars is applying a threshold, and the threshold is a setting rather than an industry standard.

How many visits makes a guest a regular?

There is no standard, and the number that works depends on the room’s cadence. Three visits inside a year is the most common default in guest platforms and a reasonable starting point for a dining room doing normal covers. A neighbourhood bar whose regulars come weekly will find three a year almost meaningless, and a destination tasting room where a devoted guest returns twice a year will find three unreachable.

Visit count on its own also misses the shape of the visits. Three visits across one week is a traveller on a trip, and three spread evenly across a year is a relationship. Adding recency separates them, which is why most useful definitions read as a count inside a rolling window rather than a lifetime total. Loyalist uses three visits in 365 days as the default and lets a restaurant change both numbers, so the definition can match how the room actually fills.

Where should a regulars list show up?

In the reporting a team reads before service, and on the reservations themselves. A regulars list living in a marketing tool influences what gets emailed and changes nothing about a Friday night. The moment that matters is a host looking at the book, so the flag has to reach the book.

The mechanism is worth being precise about: this is scheduled reporting rather than a live alert at the door. A guest crossing the threshold appears in the next pre-shift report, and a tag written back onto their reservation is what the host sees when the booking comes up.

Why is my regulars count lower than it should be?

Because a regulars count is the metric most exposed to identity problems, and the exposure is one-directional. A guest with four visits split across three duplicate profiles never crosses a three-visit threshold, so they sit in the guestbook as three near-strangers. Anonymous covers do the same thing more quietly: a genuine weekly regular who walks in and pays cash accumulates no visits at all.

Both errors subtract. Nothing in the data invents regulars who do not exist, so a restaurant’s regulars list is reliably shorter than its actual population of regulars, and the gap widens with walk-in volume. This is worth stating in a group setting, where the roll-up is often taken as a measure of loyalty performance. It is closer to a measure of how well the venue’s records are joined, and the cheapest way to grow the list is usually to fix the join rather than to run another campaign.