Behavioral Intelligence

Static identity data can tell you what an identity looks like. Behavior can tell you how that identity actually exists.

AtData analyzes longitudinal email activity to help organizations understand whether an identity is established, active, consistent, valuable, or exhibiting behavior that deserves additional scrutiny.

That behavioral layer adds context that names, addresses, and point-in-time checks cannot provide on their own.

Static attributes describe. Behavior reveals.

Two email identities can look nearly identical in a customer record. Both may be valid. Both may use legitimate domains. Both may belong to people whose submitted information appears complete.

But their histories may be entirely different. One may have years of observed activity across many environments. The other may have appeared days ago with limited history and a sudden spike in use.

AtData’s behavioral intelligence makes those differences visible. The objective is not to label behavior as inherently good or bad. It is to provide the context organizations need to understand what that behavior means for the decision they are making.

---

Understand identity through time.

AtData builds behavioral intelligence from recurring observations around persistent email identity.

Age and longevity – How established is the identity?

The date AtData first observed an email provides a historical reference point for understanding how long that identity has existed within the network. Longevity can provide useful context because established identities and newly observed identities carry different histories. A new email is not automatically suspicious. It simply has less behavioral evidence behind it.

Recency – Is the identity still active?

Historical existence does not necessarily mean current relevance. AtData can use recent activity to help determine whether an email continues to demonstrate signs of use. That can support very different decisions:

• Re-engaging a dormant customer
• Prioritizing an audience
• Evaluating whether a profile remains current
• Adding confidence to an identity-risk assessment

Velocity – How much activity is occurring?

AtData’s email velocity is a measure based on activity observed over time. Velocity helps provide context around whether an identity demonstrates little activity, consistent use, or unusually high levels of activity. The signal does not make the decision by itself. Its meaning depends on the use case and the other evidence surrounding the identity.

Popularity – How broadly is the identity observed?

Email popularity measures the breadth of sources in which an email has been observed. An identity seen consistently across multiple environments provides different behavioral context from one with a very limited observed footprint. Popularity can contribute to customer-quality, identity-confidence, and fraud-risk decisions without requiring every organization to build that cross-network history independently.

---

The same behavior can mean different things.

Behavioral intelligence is not inherently a growth signal or a fraud signal. Its value depends on the question.

For growth – Find opportunity beyond your own engagement history.

A customer who stopped opening your emails may still be highly active elsewhere. AtData activity intelligence can help organizations distinguish between identities that appear inactive only within their own environment and identities showing limited broader activity.

Use behavioral intelligence to help:

• Prioritize leads
• Identify stronger audiences
• Re-engage active customers
• Compare acquisition quality
• Improve segmentation
• Strengthen churn and propensity models

For risk – Evaluate whether behavior supports the identity being presented.

Fraud frequently looks legitimate at the attribute level. Behavior can provide additional evidence.

AtData uses signals such as age, velocity, popularity, domain behavior, and activity consistency to help identify identities that deserve greater scrutiny. Its analysis of more than four million transactions found several email-related factors correlated with elevated payment-fraud risk, including disposable email usage, email age, IP anonymization, and domain reputation.

Use behavioral intelligence to support:

• Account-opening decisions
• Synthetic-identity detection
• Fake-account prevention
• Promotion-abuse detection
• Transaction risk
• Fraud-model enrichment

---

Behavioral intelligence works best as evidence, not a verdict.

A young email can belong to a legitimate customer.
A highly active address is not automatically valuable.
A low-activity identity is not inherently fraudulent.

That is why AtData treats individual behavioral signals as evidence within a broader decision framework. Organizations can use:

• Individual signals

  Such as Date First Seen or recent activity.

• Derived measures

  Structured indicators built from multiple observations.

• Model outputs

  Scores designed for specific use cases, such as engagement quality or identity risk.

This allows teams to choose the level of abstraction appropriate to the decision rather than relying on a single black-box classification.

Explore Our Methodology

---

Behavior becomes more meaningful at network scale.

Your first-party data can show how an identity behaves with you. AtData adds a broader perspective. Our network processes more than 150 billion deterministic activity signals each month, allowing patterns to develop across a wider set of observed interactions.

That network-level context helps organizations understand whether:

• activity is isolated or widespread
• an identity has accumulated meaningful history
• patterns appear stable or unusual
• a customer who looks inactive internally remains active elsewhere
• emerging behaviors warrant further evaluation

The value is not simply more events. It is a stronger baseline for interpreting the event happening now.

---

Why AtData

Behavioral intelligence anchored to persistent identity.

Behavioral data is most useful when it can be connected to something durable. AtData uses email as that anchor.

More than two decades of identity history, recurring activity, source breadth, and network-level observation allow AtData to place current behavior inside a longer context.

That creates an intelligence layer capable of supporting both sides of the customer equation:
Who deserves more attention? Who deserves more scrutiny?

The answer depends on the decision. The advantage comes from having the behavioral evidence to make it.

---

Don’t just know the identity. Understand how it behaves.

Static data provides a snapshot. Behavior adds the story between the snapshots. AtData helps organizations use that history to make better decisions across acquisition, engagement, identity, fraud, analytics, and risk.