Guests across reservation, POS, ordering and loyalty: how they match
What the answer looks like
Illustrative output on sample data
Illustrative example: how Lexi answers this for a multi-concept restaurant group with 12 venues. It shows how records from four systems were matched to guests, and on which identifiers.
- 1.6 million records across reservations, POS, online ordering and loyalty
- 640,000 unique guests after matching
- Matched on phone number: 41% of links
- Matched on email: 33%
- Matched on loyalty number or card token: 26%
What the data shows
84,000 guests appear under two or more email addresses, most often one personal and one work email.
What it means
Before matching, those guests were counted as separate people, which understated visit frequency and retention.
What to do
Use the matched guest count in all reports, and ask for phone number at reservation, since it links the most records.
Every match shows the identifier used. Want me to list guests matched on a single weak identifier for review?
Why it matters
The same guest often books with a work email, orders with a personal one and pays by card. Without matching, they look like several people, and every guest measure is wrong. Seeing how records were matched, and on what, lets the data team trust and explain the guest count.
Frequently asked questions
How do restaurants match guests across systems?
Match records on identifiers they share, such as email, phone number, loyalty number and payment card tokens. Each link should record which identifier was used, so the data team can check and explain how a guest profile was built.
Why do guests appear under different email addresses?
Guests use a work email for bookings, a personal one for ordering, and sometimes an old address for loyalty. Phone numbers and loyalty numbers often link these records together when emails differ, which is why collecting them at booking and checkout matters.
How do you check that guest matching is accurate?
Review a sample of matched profiles, especially those linked on a single identifier, and check for false matches such as shared family phone numbers. Keep matching rules visible and adjust them when errors appear, recording each change so reports stay explainable.