Data Cleanup Before Matching
Before any two records are compared, Treez standardizes both sides the same way, so that differences in how a value was typed don't make one person look like two people.
Driver's license / state ID number: converted to uppercase, with spaces, dashes, and other symbols removed. For example,
dl-999-888andDL999888are the same.Phone number: reduced to digits only. A leading "1" country code is removed, so
+1 (555) 123-4567and555-123-4567match. Numbers with fewer than 7 digits are ignored.Email address: lowercased and trimmed of extra spaces. Values without an
@are ignored.Name: extra spaces removed and converted to a consistent case. A middle name typed into the first-name field makes it a different name.
Placeholder ID Numbers Are Ignored
Stores often enter a placeholder when a real ID couldn't be captured. Treez treats these values as "no ID on file" so that unrelated customers aren't linked just because they share the same placeholder. Examples include:
Blank values,
N/A,NA,NONE,NULL,UNKNOWN,PLACEHOLDER,UPDATE,XXXXAll zeros, such as
0000000A single repeated digit, such as
1111111Simple sequences, such as
123456or123456789
When a record has no usable ID number, matching relies on the other details on the profile.
Match Types and Confidence
Treez checks the following combinations, from strongest to weakest:
Match | What must agree | Confidence | Result |
Driver's License | ID number | 100% | Links automatically |
Driver's License, different state | ID number matches but the ID states differ | 75% | Staff review |
Name + Phone + Email + DOB | First name, last name, phone, email, and date of birth | 95% | Links automatically if exactly one profile matches; staff review if more than one does |
Name + Date of Birth | First name, last name, date of birth | 84% | Staff review |
Phone + Name | Phone, first name, last name | 79% | Staff review |
Phone + Last Name | Phone, last name | 75% | Staff review |
Email + Date of Birth | Email, date of birth | 65% | Staff review |
Email only | 59% | Staff review |
Exact matches link automatically with no staff action. An exact match is either a validated ID number match, or a Name + Phone + Email + DOB match to exactly one org profile.
Everything else is a possible match. Treez never links a possible match on its own. Staff see the options in a POTENTIAL MATCHES FOUND window and choose. See Central Customer | Linking and Unlinking Customers.
Additional safeguards
Name alone and phone alone are never enough to suggest a match.
Shared values are skipped. If a phone number or email is on file for more than 25 different people (for example, a store's default email), that value isn't used for matching.
Banned customers are never linked automatically. Even a 100% ID match is shown to staff first, with a red BANNED tag.
Backfill: Bringing In Existing Customers
Backfill runs once, automatically, when Central Customer is turned on for your organization.
It reviews existing customer records across all of your stores and consolidates records that are exact matches into a single org profile.
It does not make best-guess merges. Records that aren't exact matches stay as separate org profiles.
Purchase history, visit counts, notes, documents, and groups from each consolidated record carry forward to the org profile.
Store records are linked to their org profile the first time the customer's profile is opened in Customer Management or the customer checks in.
How Data Is Combined
When store records are linked to the same org profile, their data is combined using these rules:
Data | Rule |
Name, phone, email, address, ID details | The most recently saved value wins. A blank value never erases existing data. |
Customer type | Medical always wins and is never changed back to Adult-Use. |
Customer status | Active always wins. Deactivating a customer at one store doesn't deactivate the org profile. |
Opt-in | If the customer opted in at any store, the org profile stays opted in. An opt-out at one store does not carry over. |
Customer groups | Combined across all stores. |
Visits and sales | Added up across all stores. |
Last visit | The most recent date at any store. |
Documents and notes | All are kept and shared across stores. |
Bans | If any store bans the customer, the ban applies everywhere until it is lifted. |
Merging Duplicate Customers
When your store merges two duplicate customer records (with Merge Customers in the POS or Approve match under PATIENT MERGE in Customer Management), and those records were linked to different org profiles, Treez merges the org profiles too.
Store links, notes, documents, and customer groups move to the surviving org profile, and visit and sales totals are combined.
Empty fields on the surviving org profile are filled in from the merged one.
Bans always carry over. If either org profile was banned, the surviving profile is banned. A merge never lifts a ban.
The merge is recorded in the Org-wide Event History.
Your store's merge is never blocked. If the org profile can't be updated right away, Treez retries automatically within a few minutes.
Important Notes
A mistyped ID number creates a separate profile. If the same person's ID was entered differently at two stores, they may end up with two org profiles. Correct the ID number at the store with the mistake.
Records with and without an ID number don't merge during backfill. A customer with an ID on file at one store and no ID at another may have two org profiles. These can be linked later through staff review.
An ID number can belong to only one org profile. If a store saves an ID number that already belongs to a different org profile, the org profile keeps its current ID number and staff see a message explaining why.
Splitting an org profile that was linked incorrectly requires Treez Support.