Reaching a target audience has become increasingly challenging for advertisers. With the average American owning 22 internet-connected devices, the journey from ad exposure to conversion is more fragmented than ever.

An identity graph is the key to effectively reaching your audience throughout this journey. It recognizes consumers across channels and devices, and unifies these identifiers into a single profile. These profiles help advertisers understand and engage with their target audience. When you target an audience based on this profile rather than individual signals, you create a more holistic marketing strategy that resonates with your audience and drives them to convert. You also have greater control over how often a single user is exposed to your ad. This control leads to smarter budget allocation and helps avoid negative associations with your brand due to overexposure.

The IQM Identity Graph covers the US, with a scale of 250M+ US individuals (~75% of US population) and 102M+ US households (~78% of US households). We refresh the graph daily to account for minor changes based on signals we observed in the previous 24 hours, and monthly to ensure its IP addresses, devices, and cookies are still active and reachable.

Here’s how the IQM Identity Graph works: IQM collects billions of data signals from various sources. We make connections between these sources based on pseudonymized identifiers and probabilistic and deterministic data. This process allows us to figure out which signals likely came from the same user and assign a person-level identifier to them. We call this identifier an IQM ID, which allows you to target the individual across their devices via your programmatic ad campaigns.

The IQM Identity Graph goes beyond the individual user by also spotting when multiple IQM IDs are usually linked to the same signals such as IP addresses. This pattern helps us decide when multiple individuals likely come from the same household. Their IQM IDs are grouped together and given one IQM household-level identifier (HH ID).

The IQM Identity Graph is fully privacy compliant by design. IQM doesn’t track any device or cookie with “Do not track” settings enabled, and ingests data that’s compliant with strict privacy standards (e.g., HIPAA, CCPA), making it inherently suited for regulated verticals such as healthcare, government, and finance.

When you create an audience with first-party data, or enable other audience-targeting features, you also agree to comply with applicable data laws and regulations included but not limited to GDPR, CCPA, and HIPAA.

The IQM Identity Graph is natively connected to both our DSP and DMP. It gives you full visibility and control throughout the campaign lifecycle (planning, targeting and activation, measurement) without relying on third-party identity layers:

Pseudonymization is a de-identification process. It replaces personally identifiable information (PII) with a pseudonym. This means that the result of the pseudonymization process is an identifier that is unique and acts as a placeholder for sensitive information.

While pseudonymized data has been de-identified, it can be reversed and re-identified. This is different from anonymized data, which has been changed in a way that makes re-identification impossible.

An IQM ID is a pseudonymized identifier that’s universal and tied to devices and cookies in the IQM Identity Graph. Each ID can be thought of as related to an individual person and associated with demographic, psychographic segments. Note that an IQM ID’s segment attribution and links to devices and/or cookies may change over time as we receive more signals and learn more about a given user.

An IQM household identifier (HH ID) is a collection of IQM IDs. These IQM IDs are grouped together under a single IQM HH ID when those individuals likely come from the same household. This is determined by reviewing when multiple IQM IDs are usually linked to the same signals such as IP addresses. An IQM HH ID also has segment attribution as the household level (e.g., household income, family size and mix, etc.)

The IQM HH ID is especially important as we keep the cookieless future in mind. By linking IP addresses to verified households, precise targeting is possible even when cookies or device IDs are unavailable.

A number of the IQM Identity Graph’s source signals are hashed identifiers such as hashed phone numbers, household addresses, and emails. This means that any formerly personally identifiable information (PII) was replaced with a unique string of characters. This process makes it difficult to determine the original phone number, household address, or email, which helps protect users’ privacy and security.

Deterministic data is a user’s known information based on information they’ve provided, such as a record of their email address. It’s ideal for accuracy, but scale is often limited.

Probabilistic data is information that we can infer about users based on similarities and patterns in observed signals. It has the potential to be less accurate, but it provides much greater scale.

The IQM ID Graph employs both probabilistic and deterministic data-linking methodologies to deliver better audience accuracy and higher match rates.