Data platform
Customer data integration: one record per customer across systems that disagree
How to bring the customer records in your CRM, billing, support and app systems into one consistent view, and keep it current.
Guide Published 5 min read
- Built and supported by Interkey An Interkey product, delivered and supported by Interkey
- One platform for migration and continuous sync The move and the replication that follows it
- Written quote after one meeting The price is discussed in a meeting about your systems
Customer data integration (CDI) brings the records a company holds about each customer, in its CRM, billing, support and app systems, into one consistent view. It means matching the same customer across systems, deciding which system is right for each field, and keeping the result up to date. It must also follow your own data rules on what each record may be used for.
Batch or real time: three patterns
Most projects use more than one.
| Pattern | How it works | When it fits | What it needs |
|---|---|---|---|
| Nightly load | Once a day, records are extracted, matched and loaded into the unified record | Reports and analysis, where yesterday's data is enough | A scheduled job, run when the sources are quiet |
| Continuous sync | Each change in a source is read as it happens and applied to the unified record | Service desks, apps and branches, where the record is seen during service | Change capture on the source databases, and a reference system for each field |
| Request on demand through an API | An application asks each system for the customer when needed, then assembles the answer | Few systems and few requests, each answered quickly | An API on every source, and a plan for when one is down |
The nightly load is a form of ETL. Continuous sync rests on change data capture.
Why the same customer looks different in every system
Each system records the customer for its own job, typed by different people. Sales use the CRM, finance uses billing, the contact centre uses the support tool, and the customer types into the app. Three differences are common in Saudi records.
The same mobile number is stored as 05XXXXXXXX in one system and +9665XXXXXXXX in another. One Arabic name is written in Latin letters as Mohammed in one system and Muhammad in another. And one customer is a person identified by a national ID in the app, but a company identified by its commercial registration in billing. None of these is an error inside its own system; they become errors side by side.
From four systems to one record
-
Four systems, four versions of the customer
CRM Billing Support App -
Matching
The same customer found in each system, and the right value chosen for each field
-
One record per customer
Kept current as the systems change
Matching rules are designed in the project.
The four steps of customer data integration
Each step depends on the one before it.
-
Collect.
List every system that holds customer data, the fields it keeps and who owns it, and bring the records together.
-
Match.
Find the same customer across systems, on a stable identifier where one exists, otherwise on matching rules.
-
Choose the reference for each field.
Decide which system is right for each field: billing for the billing address, the app for contact choices, the CRM for the account owner.
-
Keep it in sync.
Apply new changes to the unified record, and back to the other systems where agreed, so the result does not age.
Your data rules come first
Combining customer records is a data protection question before it is a technical one. Settle these points with the people responsible for your data rules before design starts; this page gives no legal reading.
- Purpose: each record was collected for a purpose; using it for another is your data owners' decision.
- Only what is needed: the unified record carries the fields its users need, not every field every system holds.
- Access: who may see which fields is set by role, and every read and change is logged.
- Correction and deletion: a customer's request to correct or remove a record reaches every connected system.
- Data residency: where the record and its copies are stored is agreed with your security team first.
Where the unified record lives
The unified record needs a home, and there are three usual choices. The first is an existing system, often the CRM, made the reference. It is the lightest choice, but it ties the record to that system's limits. The second is a dedicated customer database that serves the record to apps and service channels quickly; our customer 360 use case describes this pattern. The third is a data warehouse, which suits reporting but is rarely fast enough for a customer at the counter.
Relatedcustomer 360 use case
Where iblync fits
iblync is Interkey's platform for continuous sync between different systems. In customer data integration it carries the fourth step: changes in the source databases reach the unified record as they happen, in one direction or in both directions. iblync does not match identities or clean data: that is designed in the project, and iblync then keeps the result current.
Relatediblyncin both directions
Questions about customer data integration
What is customer data integration?
Customer data integration is the work of bringing what a company knows about each customer, spread across its CRM, billing, support and app systems, into one consistent record. It covers matching the same customer across systems, choosing the right system for each field, and keeping the result current as the systems change.
What is the difference between CDI and a CDP?
CDI is work that gets done; a customer data platform (CDP) is a product you buy to do part of it, usually for marketing. CDI also covers billing, support and operational systems, and the rules that decide which system is right for each field.
What is the difference between CDI and MDM?
Master data management (MDM) is the wider discipline: it covers products, suppliers, locations and other shared data, as well as customers. CDI is the part of that work that concerns the customer record. A company can start with CDI and extend the same rules to other data later.
How do you match customers across systems?
On a stable identifier where one exists in every system, such as a customer number. Where none exists, matching rules compare mobile number, email and name after bringing each to one format, and a person reviews the doubtful cases. The rules are tested on a sample first.
Who decides which customer data may be combined?
Your data owners decide, with your data protection officer or legal adviser, before the design starts. Whether two records may be combined depends on the purpose each was collected for and on your own data rules. The integration project then builds only what has been agreed; it does not take that decision itself.
Should customer data be synced in real time?
Yes, where a customer or an employee sees the record while being served: at a counter, in the app or on a support call. For reports and analysis, a regular update, often nightly, is enough. Many projects sync the service channels continuously and feed reporting on a schedule.
Next
Read next
Next step
Ask an expert one question
Tell us which systems hold your customer records and who needs the unified view. Ask one question about where to start.
What happens next
- We read your request
- An engineer contacts you
- You get an estimate or a demo
Or directly
+966-11-2180999 [email protected]Tawuniya Towers, North Tower, 7th Floor, King Fahad Highway, Olaya, P.O. Box 56835, Riyadh 11564, Saudi Arabia