Identity Resolution and Alternative IDs

Understanding who your customer is across multiple devices, sessions, and channels has always been a challenge in digital advertising. Identity resolution addresses that challenge, and as the industry moves away from third-party cookies, alternative identifiers have emerged as a key piece of the solution.

What Is Identity Resolution?

Identity resolution is the process of linking disparate data points (a mobile device ID, an email address, a loyalty card number, a browser session) to a single, unified customer profile. The goal is to recognize that the person who browsed a product on their phone in the morning and completed a purchase on a desktop in the afternoon is the same individual.

This matters for advertising because fragmented identity leads to:

• Duplicate ad exposure (showing the same ad to the same person multiple times on different devices)
• Broken attribution (failing to connect ad touchpoints to a conversion)
• Poorly calibrated frequency caps
• Inaccurate audience segmentation

Deterministic vs. Probabilistic Matching

Identity resolution methods fall into two broad categories:

Deterministic Matching

Deterministic matching uses a known, consistent identifier, typically a hashed email address or a phone number, to link records with high confidence. When a user logs into an app with their email, that signal is deterministic: the system knows with certainty who that user is.

Deterministic matching is accurate but limited in reach. It only works when users are authenticated, which covers a fraction of total digital activity.

Probabilistic Matching

Probabilistic matching uses statistical inference to link records that share common signals (IP address, device type, browser fingerprint, behavioral patterns) even without a definitive identifier. It extends reach significantly but introduces uncertainty, and its accuracy degrades over time as signals shift.

Most real-world identity graphs combine both approaches, using deterministic anchors where available and probabilistic inference to extend coverage.

The Third-Party Cookie Problem

For over two decades, the third-party cookie served as the de facto identity layer of the open web. It allowed advertisers to track users across sites, build cross-site behavioral profiles, and serve retargeted ads. As browsers restricted or eliminated third-party cookies, this mechanism broke down.

The consequences are significant:

• Cross-site audience segments become harder to build and maintain
• Retargeting relies more heavily on authenticated signals or publisher first-party data
• Frequency management across sites becomes unreliable
• Attribution models that depended on cookie-based tracking require rethinking

Alternative IDs: A New Identity Layer

The industry has responded to cookie loss by developing alternative identity solutions. These fall into several categories:

Email-Based IDs

Solutions such as Unified ID 2.0 (UID2) and RampID use hashed or encrypted email addresses as persistent identifiers. When a user authenticates on a publisher’s site, by logging in or accepting email-based consent, their email is hashed and matched against the same hash held by advertisers and DSPs. This creates a cross-site identifier that does not rely on cookies.

These solutions require user consent and publisher participation, and their reach is proportional to how widely they are adopted across the ecosystem.

Publisher First-Party IDs

Many publishers have developed their own authenticated identity layers, assigning persistent IDs to logged-in users. These are highly accurate within a publisher’s own properties but do not carry across unrelated sites without a shared identity graph.

Contextual Signals

Rather than identifying the user, contextual targeting focuses on the content being consumed. Advanced contextual solutions use natural language processing to assess the relevance of a page to a given topic or product category, enabling targeting without any user-level identifier at all.

Device Graphs and IP-Based Approaches

Device graphs attempt to link devices to households or individuals using a combination of deterministic and probabilistic signals. IP-based targeting has also seen renewed interest as a household-level proxy, though it carries significant limitations in shared-network environments.

Privacy and Consent Considerations

Alternative IDs are not inherently privacy-safe. Their compliance depends on how they are implemented. Key considerations include:

• Whether user consent is collected before an identifier is created
• Whether users can opt out and have their identifier deleted
• Whether the identifier enables cross-site tracking in ways users would not expect

The most robust solutions are built around explicit consent and offer clear user controls.

How Arago Helps

Arago integrates with modern identity solutions to help local retailers and regional advertisers maintain addressability in a post-cookie environment. By supporting authenticated identifiers and first-party data activation, Arago enables campaigns that are both effective and aligned with current privacy standards.