Contact Engagement Predictions

See AI-predicted engagement likelihood for contacts

Contact Engagement Predictions

Mailnaptic's predictive analytics engine uses machine learning to estimate how likely each contact is to engage with your future emails. These predictions help you prioritize high-value contacts, re-engage at-risk subscribers, and make smarter sending decisions.

How Predictions Work

The prediction model analyzes each contact's historical behavior across several dimensions:

• Email engagement — Open frequency, click patterns, time-to-open, and engagement recency.
• List tenure — How long the contact has been subscribed and their activity trajectory over time.
• Campaign interaction depth — Whether they just open emails or also click links, visit your site, and convert.
• Frequency sensitivity — How their engagement changes relative to the volume of emails they receive.
• Web activity — If site tracking is enabled, page visits and custom events are factored into the prediction.

Engagement Scores

Each contact receives an engagement prediction score from 0 to 100, updated daily. The score is divided into four tiers:

• Highly Engaged (75–100) — These contacts consistently open, click, and interact with your emails. They are your most valuable subscribers.
• Engaged (50–74) — Solid engagement with room for growth. These contacts open most emails and occasionally click.
• At Risk (25–49) — Engagement is declining. These contacts may be losing interest or experiencing inbox fatigue.
• Disengaged (0–24) — Minimal or no recent engagement. Consider a re-engagement campaign or list hygiene action.

Viewing Predictions

Engagement predictions appear in several places within Mailnaptic:

• Contact profile — Each contact's detail page displays their current engagement score with a trend indicator showing whether it is rising, stable, or declining.
• Contact list — Sort or filter your contact list by engagement score to quickly identify your strongest and weakest subscribers.
• Segment builder — Use the "Predicted engagement" filter to create segments based on score ranges or tiers.

Using Predictions in Campaigns

Leverage engagement predictions to optimize your campaigns. Target "Highly Engaged" contacts for product launches and exclusive offers. Create dedicated re-engagement flows for "At Risk" contacts before they disengage completely. Exclude "Disengaged" contacts from regular campaigns to improve overall deliverability metrics.

Prediction Accuracy

Predictions become more accurate as Mailnaptic collects more behavioral data. New organizations or contacts with fewer than five email interactions will see less precise scores. The model recalibrates nightly to incorporate the latest engagement signals.