Quick answer
Here’s a concrete identity resolution example. A person visits your website Monday morning from a work laptop. They browse three product pages and leave without buying. Tuesday they open a marketing email on their phone during lunch. Thursday they call your support line to ask a question. Friday they buy on their home tablet. Without identity resolution, this looks like four separate people. With identity resolution, it becomes one customer profile with a five-channel touch history that powers accurate attribution, personalization, and modeling.
Walking through the example
Monday, work laptop. The visitor lands on your website. A first-party cookie assigns them anonymous ID A. They browse three product pages. Your analytics record page views tied to ID A. No login, no email captured yet.
Tuesday, phone. They open a marketing email on their phone. The email contains a tracking pixel that records the open and links it to their email address (E1) plus a mobile device ID (D1). Your platform now has E1 linked to D1.
Wednesday, work laptop. They return to your website and this time they subscribe to a newsletter with email E1. The website now knows anonymous ID A belongs to email E1. Identity resolution merges the Monday browsing history into the E1 profile.
Thursday, phone call. They call support from phone number P1. Your support system captures the call and asks for verification. They give email E1. Support records the call tied to E1. Identity resolution links P1 to E1.
Friday, home tablet. They purchase on a home tablet, checking out with email E1 (deterministic anchor). The purchase includes payment info that anchors a customer ID C1. Identity resolution now has: E1 = C1 = P1 = D1 = A, and a home tablet device ID D2 also gets linked because it authenticated as E1.
What identity resolution produced
One customer profile with:
- Five identifiers (A, E1, D1, P1, C1, D2)
- Five touchpoints across three channels
- A purchase history and product interest signals
- A support interaction log
- Enough data to attribute the purchase across the full journey and predict lifetime value
Without identity resolution
The same person shows up as four to five separate records. Marketing wastes budget re-acquiring the same person. Attribution credits the last touch (probably the email) at the expense of the earlier discovery. Lifetime value modeling underestimates them because their purchase history is split. Support has no context on their prior interactions.
Where this scales
At 1,000 customers, the loss from missing identity resolution is manageable. At 100,000+ customers across multiple channels, the loss compounds into millions of dollars annually in wasted marketing spend and misattributed revenue.
Common follow-up questions
How long does the linking process take?
Deterministic links happen instantly. Probabilistic links happen on batch cycles (hourly to daily). Full customer graphs update continuously in most modern systems.
What if the person uses different emails at different points?
Identity resolution can still link them if other identifiers overlap (same phone, same devices, same purchase history). Multiple email addresses on one profile is common.
Does this example require enterprise tooling?
No. Basic identity resolution handles this specific example. More complex use cases (cross-device without login, multi-brand data, offline-to-online linking) require more sophisticated tooling.