Knowledgebase

Modelling Customer Data Print

  • dataengineering, data, billing, guide, howto, solution, zillionkinghost, hosting
  • 0

A single view of a customer.

WHAT IT MEANS

One record per customer, combining information from every system.

WHY IT IS DIFFICULT

Systems identify customers differently Values conflict Definitions of customer differ

WHAT TO ESTABLISH FIRST

What a customer is.

WHY

A person, an account, an organisation and a billing entity are different things, and systems disagree.

WHAT TO BUILD

An identity mapping, resolving records across systems.

WHAT TO DECIDE PER FIELD

Which system is authoritative.

WHAT TO RETAIN

The source values alongside the resolved one.

WHAT ATTRIBUTES TO INCLUDE

Identity and contact

  • Acquisition: when, and from where
  • Commercial: revenue, products, tenure
  • Behavioural: usage and engagement
  • Status: active, lapsed, churned

WHAT TO MODEL OVER TIME

Status and segment, since history matters.

WHAT TO BE CAREFUL WITH

Personal data, which this concentrates Retention obligations Access control, which matters more here than elsewhere

WHAT TO TEST

That counts reconcile against each source system.


Was this answer helpful?
Back

Are you happy with your experience? Leave us a review on Trustpilot.


Trustpilot