---
title: "Identity resolution | Snowplow Signals"
description: "Snowplow Identities links cookies, logins, device IDs or any identifier you collect to one Snowplow ID, so a Signals profile follows a customer across devices."
source: https://signals.snowplow.io/products/identity-resolution
---

[Product](https://signals.snowplow.io/products) / Identity resolution 

# Recognize the same customer across devices. 

Real-time identity resolution links any identifier you collect, whether a cookie, a login, a device ID or your own loyalty number, to one customer.

[Start free](https://signals.snowplow.io/start) [Talk to an engineer](https://signals.snowplow.io/start?intent=engineer) 

The real engine · no card · no sales call

The problem

## One customer shows up as several profiles. 

A customer browses on their phone, signs in on their laptop and buys in your app. Your app sees three different people, none with the full history.

How it works

## Every identifier, resolved to one person. 

The pipeline resolves each event to a Snowplow ID before any attribute, trigger or warehouse row reads it. Your app reads attributes, what you want to know about a customer, keyed on that ID in 6ms, in-region.

* cookied\_7f3a2c
* user\_idu\_481
* email\_sha2569f86d0…
* idfv4C1E-77A0
* gaid38b2-91fe
* loyalty\_idL-2201
* phone\_sha256c1d02e…
* crm\_id0048172
* device\_iddv\_33
* account\_idacct\_90f
* store\_card7712-0049
* cookied\_c210e4

**12** identifiers 

One customer

One Snowplow ID

Works with 
* Snowflake
* BigQuery
* Databricks
* AWS
* Google Cloud

Capabilities

## Resolved as it happens, never guessed. 

01 

### Real time

Identities resolve as the pipeline processes them, so attributes count one customer while the visit is still going.

02 

### Deterministic

It links only the identifiers you choose in Console, such as built-in IDs, a hashed email or your own customer ID, and never matches on probability. A unique ID such as a login stops two people on a shared device from merging.

03 

### One ID in Signals and your warehouse

Every enriched event carries the Snowplow ID, so Signals attributes keyed on it and your warehouse tables count the same customer. A dbt package builds the mapping tables.

[Snowplow Identities →](https://docs.snowplow.io/docs/identities/) [See it in Snowplow Console →](https://signals.snowplow.io/products#console) 

Identity resolution

## Frequently asked questions. 

### What happens to browsing on a new device before a customer signs in?

When they sign in, identity resolution merges that device's anonymous ID into their Snowplow ID, so the browsing before sign-in joins the rest of their history.

### Which platforms does it cover?

Web, iOS and Android, signed in or anonymous.

### Where does the identity graph run?

In a managed database in the same AWS or GCP cloud as your pipeline. An erasure API removes a person's entries for a right-to-erasure request.

### What happens without Identities?

Attributes key on the IDs in your events, such as the cookie or the user ID, so one person can show up as several profiles.

## Start with one attribute. 

Add the SDK, define one attribute, and watch it update as you browse and search your own product. The free tier runs on the real engine: no card, no sales call.

[Start free](https://signals.snowplow.io/start) [Talk to an engineer](https://signals.snowplow.io/start?intent=engineer) 

Next up: [Event collection →](https://signals.snowplow.io/products/event-collection) and [Real-time attributes →](https://signals.snowplow.io/products/real-time-attributes).