What Identity Stitching Is
Identity stitching is the process of joining the separate records one person creates across devices and systems into a single profile. A person might be an anonymous visitor on your website, a lead in your CRM and a signed-in user in your app. Each system gives them a different identifier. Stitching is how you learn that all three are the same person.
Without it, every system reports its own slice of the journey, and none of them can say which campaign produced a customer.
How Identity Stitching Works
Stitching depends on an identifier that appears in more than one record:
- User ID: an ID your own systems generate, passed at sign-up or login on the web and in the app
- CRM contact ID: the lead's ID, carried into the app through an onboarding link
- Hashed email or phone: contact details turned into a one-way code so systems can match without sharing the raw value
When two records carry the same identifier, they are joined, and the earlier anonymous activity is attached to the known person.
Deterministic Versus Probabilistic Stitching
Deterministic stitching joins records only when they share an exact identifier. It is accurate and easy to explain. Probabilistic stitching infers a match from signals such as device type or network, which is less reliable and harder to defend under privacy rules.
Most teams should start deterministic: choose one ID, pass it everywhere, and accept that unmatched records stay separate.
Identity Stitching in Attribution
Stitching is what turns cost per lead into cost per customer. When a website visit, a CRM deal and an app payment share one ID, you can credit the payment to the ad that started the journey. It also closes cross-device gaps, for example a form filled on a laptop and an app installed on a phone.
Linkrunner matches deterministically first. For cross-device journeys, it can connect records through your customer user ID and hashed email or phone, never IP address.
Common Mistakes
- Choosing the ID too late. If the website, CRM and app each invent their own, they cannot be joined afterwards.
- Inconsistent formatting. Upper and lower case or stray spaces in an email break exact matches.
- Relying only on email. People often sign up in an app with a different address from the one on the form.
- Stitching without consent. Joining records across systems needs a lawful basis, recorded consent where required, and a deletion path.