Email engagement scoring assigns a numerical value to each subscriber based on their interaction history — opens, clicks, replies, purchases, and recency. Use engagement scores to segment your list into tiers (highly engaged, active, cooling, at-risk, inactive) and adjust sending frequency accordingly. Sending more to engaged subscribers and less to inactive ones directly improves deliverability by boosting positive engagement signals that Gmail and Outlook use for reputation scoring.
Email Engagement Scoring: How to Build and Use Engagement Scores
Why Engagement Scoring Matters
Gmail's spam filtering algorithm weighs recipient engagement heavily. When your recipients open, click, reply, and move messages from spam to inbox, Gmail learns your mail is wanted. When they ignore, delete without reading, or mark as spam, Gmail learns the opposite.
The problem: most senders treat their entire list the same. Same frequency, same content, same everything. This means highly engaged subscribers get the same treatment as people who haven't opened an email in six months. Those inactive subscribers are actively hurting your reputation with every ignored send.
Engagement scoring fixes this by creating a framework to send more to people who want your email and less (or nothing) to people who don't.
Building Your Scoring Model
Step 1: Define Actions and Weights
| Action | Points | Notes |
|---|---|---|
| Email open | +2 | Least reliable due to image blocking/privacy |
| Email click | +5 | Stronger signal — requires deliberate action |
| Email reply | +10 | Strongest email signal |
| Purchase from email | +15 | Revenue-driving engagement |
| Website visit from email | +3 | Measured via UTM tracking |
| Form submission | +8 | Active interest signal |
| Unsubscribe attempt | -20 | Strong negative signal |
| Spam complaint | -50 | Remove immediately |
Adjust weights based on your business. For ecommerce, purchases might be +20. For B2B SaaS, demo requests might be +25.
Step 2: Apply Time Decay
Engagement fades. Someone who clicked every email last month but hasn't opened one in 60 days is cooling off. Apply time decay to prevent stale scores.
Recommended decay: Reduce the total score by 10-20% per week of inactivity. This means:
- A subscriber with a score of 50 who goes inactive hits 25 in 3-4 weeks
- After 8 weeks of inactivity, their score approaches zero regardless of past engagement
Time decay ensures your scoring reflects current behavior, not historical loyalty.
Step 3: Set Tier Thresholds
| Tier | Score Range | Sending Strategy |
|---|---|---|
| Highly Engaged | 50+ | Full frequency, priority for new content |
| Active | 20-49 | Normal frequency |
| Cooling | 5-19 | Reduced frequency (50%), re-engagement content |
| At Risk | 1-4 | Minimal sending, re-engagement campaign |
| Inactive | 0 | Stop sending, sunset campaign, then remove |
These thresholds aren't universal. Calibrate based on your sending frequency and audience. A daily newsletter needs different thresholds than a monthly product update.
Practitioner note: Start with simple scoring and refine. I've seen teams spend months building elaborate models before sending a single scored campaign. A basic model — clicks in last 30 days yes/no — already beats treating the whole list equally. Optimize the model after it's live and generating data.
Implementation by ESP
Klaviyo
Klaviyo has built-in engagement scoring via "Engaged" and "Not Engaged" segments based on open/click activity. You can also build custom scoring using calculated fields and flow logic. Klaviyo's engagement-based segments are a good starting point.
ActiveCampaign
ActiveCampaign's contact scoring feature lets you assign point values to email actions, site visits, and custom events. Scoring rules are configured in Contacts > Scoring. Combine with automations to move contacts between lists based on score thresholds.
HubSpot
HubSpot's lead scoring tool works for email engagement scoring. Create scoring properties based on email engagement, set threshold-based lists, and use workflows to adjust sending behavior per tier.
Mailchimp
Mailchimp doesn't have native scoring, but you can build segments based on engagement activity (opens/clicks in last 30/60/90 days). Use tags and segments as a manual scoring proxy.
Custom Implementation
For teams using transactional senders (SendGrid, Postmark) or cold email tools, build engagement scoring in your CRM or database. Track opens/clicks via webhooks, calculate scores on a schedule, and use score-based segments for sending.
Practitioner note: The ESP's built-in engagement tracking is usually good enough for the scoring model. Don't build a separate tracking layer unless your ESP's data is insufficient. The goal is actionable segmentation, not perfect measurement.
Using Engagement Scores
Frequency Optimization
The primary use: adjust sending frequency by tier. Highly engaged subscribers get every email. At-risk subscribers get one email per week maximum. This improves aggregate engagement metrics, which improves sender reputation.
Content Differentiation
Send your best content to your most engaged subscribers first. They're most likely to interact, generating positive signals. Then send to broader segments. This creates an engagement cascade that lifts overall campaign performance.
Re-engagement Triggers
When a subscriber drops from "Active" to "Cooling," trigger a re-engagement automation. Don't wait until they're fully inactive — re-engagement works better when there's still residual interest.
Sunset Decisions
Engagement scores make sunset policy decisions data-driven. Instead of "remove anyone who hasn't opened in 90 days," you sunset "anyone whose engagement score has been below 5 for 30 consecutive days." This accounts for the full engagement picture, not just one metric.
Deliverability Monitoring
Track average engagement score across your sending pool over time. A declining average score signals list fatigue or content problems before deliverability metrics show the damage.
Practitioner note: One client improved their Gmail inbox placement from 72% to 91% in three weeks by implementing engagement-based sending. They didn't change their content, authentication, or ESP. They just stopped sending to their bottom 30% of engaged subscribers. Gmail saw better engagement signals and promoted their reputation.
Common Mistakes
Scoring opens equally to clicks. Opens are unreliable (Apple Mail Privacy Protection, image blocking). Weight clicks 2-3x more than opens.
No time decay. Without decay, a subscriber who was active two years ago still looks engaged. Decay is essential.
Too many tiers. Five tiers is plenty. More creates complexity without actionable differences in strategy.
Scoring but not acting. Engagement scores are useless if you don't change sending behavior based on them. The score drives the segmentation, the segmentation drives the strategy.
The Bottom Line
Engagement scoring is the most impactful list hygiene practice you can implement. It transforms your sending from "blast everyone" to "send to people who want to hear from you." The deliverability improvement is often dramatic and immediate.
If you need help building an engagement scoring model for your specific ESP and audience, schedule a consultation — I'll architect a scoring system tailored to your business metrics.
Sources
- Gmail: Bulk Sender Requirements
- Klaviyo: Engagement-Based Segmentation
- ActiveCampaign: Contact Scoring
- Return Path: Engagement and Deliverability Research
v1.0 · March 2026
Frequently Asked Questions
What is email engagement scoring?
A system that assigns points to subscribers based on their email interactions. Opens, clicks, replies, and purchases add points. Inactivity subtracts points (decay). The resulting score segments subscribers into engagement tiers that guide sending frequency and content strategy.
Why does engagement scoring matter for deliverability?
Gmail, Outlook, and Yahoo use recipient engagement as a major reputation signal. If most of your recipients ignore your emails, mailbox providers assume your mail isn't wanted and route it to spam. Engagement scoring helps you send primarily to people who interact — boosting the positive signals that maintain high sender reputation.
How do you build an engagement scoring model?
Assign point values to actions (click = 5 points, open = 2 points, purchase = 10 points). Apply time decay (reduce score by 10-20% weekly for no interaction). Set thresholds for tiers (above 50 = highly engaged, 20-50 = active, 5-20 = cooling, below 5 = at risk). Adjust sending frequency per tier.
Want this handled for you?
Free 30-minute strategy call. Walk away with a plan either way.