Understanding AudiencePoint’s Recency Engagement Score (“Spruce”)
AudiencePoint’s Spruce Engagement Score adds a dimension that your own send history cannot provide: how actively a subscriber engages with brands via email. Someone who is consistently active across email is more likely to engage with your next send than someone who isn’t. That signal clarifies the decisions your other data leaves open: who belongs in this send, who to hold back, and who is worth a reactivation attempt.
Why your own data is not enough
Every sender works from the same limited view: what happened when you sent something. If a subscriber opened, your ESP recorded it. If they did not, you recorded silence. The problem is that this view has two significant blind spots.
The first is false positives. Mail Privacy Protection and similar proxy features pre-fetch email content on a user’s behalf, registering an open before the person ever sees the message, or decides whether they want to. Some ESPs attempt to filter these out, but detection varies widely by platform and typically catches only the most obvious signals. The result is a layer of noise that most senders are still building decisions on top of.
The second is false negatives. A subscriber who has been quiet on your list may be actively engaging with email from other brands. From your vantage point, they look dormant. From a cross-brand view, they are still engaging.
AudiencePoint operates across a shared data pool of engagement signals observed across senders. That scale makes two things possible: filtering out machine-generated opens with a high degree of accuracy, and surfacing genuine engagement activity that your own sends have never captured.
The result is a score built on observed human engagement rather than your brand’s history alone.
What you will see
Each subscriber in your audience receives a score from 0 to 5. Higher scores reflect more recent verified engagement activity. Lower scores reflect less recent activity or no verified activity within the scoring window.
| Score | What it means |
|---|---|
| 5 | Real open and click recorded in the last 14 days |
| 4 | Real open or click recorded in the last 60 days |
| 3 | Real open or click recorded in the last 120 days |
| 2 | Real open or click recorded in the last 240 days |
| 1 | Real open or click recorded in the last 360 days |
| 0 | No real activity in the last 360 days, but does have real activity in the last 720 days |
Subscribers who cannot be scored fall into one of two additional categories: Unmatched, meaning insufficient engagement history exists in the data pool to produce a score, or Invalid, meaning the address cannot be reached. Neither is the same as a low score.
What the score tells you about your list
Not every subscriber in your list can be scored. When AudiencePoint evaluates an audience, each address falls into one of three categories.
Matched. Enough signal exists in the data pool to assign a score. These subscribers are evaluated and returned with a rating of 0 to 5.
Unmatched. The address is real, but there is insufficient engagement history to produce a score. This is uncertainty, not confirmation of inactivity.
Invalid. The address cannot be reached. It may be syntactically broken, a known spam trap, an abandoned inbox, or otherwise undeliverable.
The distinction between Unmatched and a low score matters. A 1-star subscriber recorded a real open or click within the last 360 days. An Unmatched subscriber is an unknown. They may be active, inactive, or somewhere in between. Treating them identically will produce different outcomes, and neither should be treated the same as Invalid.
Looking at your score distribution as a whole tells you something about the condition of your list: what proportion is highly active, what proportion is fading, and what proportion cannot be evaluated at all.
What the Engagement Score is not
The Engagement Score is not an inbox placement signal, an address verification result, or a measure of how a subscriber has responded to your specific program. It measures how recently the subscriber has demonstrated verified engagement with email across brands. A high score does not guarantee engagement with your next send, but it indicates they are a stronger candidate than one who has not.
How the score works
“Verified” means human-initiated opens and clicks. Proxy and automated events are excluded.
The scale reflects recency, not frequency. A subscriber who opened 40 times two years ago scores lower than one who clicked once last week. Recent behavior predicts future behavior more reliably than volume of past activity. That is the core principle the scale is built on.
The 5-star threshold requires both an open and a click within 14 days.
Every score band reflects observed engagement within a defined time window. A 0-star subscriber has enough history in the data pool to be scored, but no verified engagement has been recorded in the last 360 days. That is a known dormant, which is different from an Unmatched address, where no scoring history exists.
Where the score changes decisions
The Engagement Score is most useful when it informs three types of decisions.
Protecting deliverability
Inbox providers pay close attention to how your subscribers respond. High engagement signals that you are sending to people who want to hear from you. Low engagement, over time, signals the opposite, and inbox placement suffers as a result.
The score gives you a basis for pulling lower-scoring subscribers out of regular high-volume sends, preserving your sender reputation while you decide what to do with them. Suppressing a 1-star address from a major campaign is not the same as losing a subscriber. It is protecting the deliverability of every other message you send.
The same logic applies to IP warming. When establishing a new sending IP, the window for building a positive reputation is narrow. Mailing only your highest-scoring subscribers during that period gives you the strongest possible signal to start from.
Reactivation
A poorly targeted reactivation send can exacerbate deliverability issues rather than resolve them. Mailing a large block of low-scoring addresses can produce complaints and hard bounces that damage your program before any recoverable subscribers have a chance to re-engage.
The score brings precision to that decision. A 2-star subscriber has been quiet for a while but has shown genuine engagement in the last 8 months. A reactivation attempt is a reasonable investment. A 1-star subscriber’s last recorded engagement anywhere in the data pool may be as far back as 360 days. They may still be worth an attempt, but it should be low volume, closely monitored, and separate from any send that carries deliverability risk you cannot afford.
High-value sends
Not every email deserves the same audience. Product launches, high-value offers, and time-sensitive campaigns perform better when they reach subscribers who are already likely to act. Routing these sends toward 4- and 5-star subscribers is not about excluding everyone else. It is about ensuring your most important messages reach the people most likely to act on them.
The score is one part of the picture
Engagement scoring answers one question: how recently has this person been active with email? It does not answer whether the address itself is safe to send to.
An address can score well on engagement and still carry risk. A subscriber flagged as a known spam trap, a role address, or a high-complaint address exposes deliverability, a factor that a good engagement score cannot offset. Verification and engagement scoring work together. They answer different questions, and both matter.
The engagement score is one input into a complete picture of your audience. Used alongside address verification and inbox placement data, it gives you the basis for decisions you can defend: who is reachable, who is active, and who is worth the send.
