Agent 05 · Sentiment Intelligence · StewardWise AI
Your Donors Are Telling You
They Are Leaving. Is Anyone Reading the Signals?
In every email they open without clicking. Every survey response with distance. Every event they stop attending. 1,247 stakeholder communications analyzed in 90 days. 13% disengaged. 3 flagged for immediate action.
He told you he was leaving. In a survey response. Six months ago. Did anyone see it?
- 1,247 stakeholder communications analyzed per 90-day window.
- Behavioral pattern shift detection: opens-only flagged as emotional disengagement.
- Explicit language signals flagged within 72 hours of submission.
- Channel recommendation: when email is no longer effective, escalate to phone.
1,247
Communications analyzed
72 hours
Explicit signal flag time
2 types
Behavioral and explicit signals
90 days
Rolling analysis window
Your Donors Are Communicating Their Disengagement in Every Interaction. Most Organizations Never Read It.
Donor disengagement leaves a trail. It is not always explicit. Robert Kim did not call to tell you he was leaving. He answered a survey question with six words that flagged exactly where his relationship with your mission stood: "I am not sure my gifts are making a difference." That response was submitted six months ago. It sat in a spreadsheet with 246 other survey responses. No one read it with the lens of retention risk.
Sarah Chen did not say anything. She just stopped clicking. For three months she has opened every email your organization sends. Prior pattern: clicked 80% of links. Current pattern: opens only. That behavioral shift is pre-lapse behavior. It is not visible in a CRM note. It is visible only to a system that tracks the ratio of opens to clicks over time and knows what a declining click-through rate signals about emotional engagement with the mission.
Sentiment Analysis reads both signal types simultaneously across every stakeholder communication in your 90-day window. Behavioral signals like Sarah's opens-without-clicks pattern. Explicit signals like Robert's survey language. Both are pre-lapse indicators. Both require different responses. Both are invisible without a system that is reading every communication with retention risk as the frame.
1,247 Communications. Two Donors. Two Signal Types. Both Pre-Lapse.
Sentiment Analysis · 90-Day Rolling Window · Full Stakeholder File
1,247
Communications analyzed
3
Flagged · Immediate action
14
Watching · Signals emerging
92%
Healthy · Positive sentiment
Sarah Chen
Email open rate
75%
→
74%
Click-through rate
80%
→
3%
Event attendance
100%
→
0%
Survey response
Always
→
None
AI Interpretation: Sarah is still aware of the organization. She opens every email. But she has emotionally disengaged from the mission narrative. Opens-without-clicks for 90 days means email is no longer effective. Phone or in-person contact required now.
Robert Kim
Survey Response · Q4 Donor Survey · October 2025
"I am not sure my gifts are making a difference."
Submitted 6 months ago · Sat in survey spreadsheet unread until now
Explicit disengagement language: uncertainty about impact, language of disconnection from mission outcomes.
No acknowledgment received after submission. No follow-up in 6 months.
Giving pattern: missed usual January window. Risk score now 44 and declining.
Action Required: Personal impact report delivery within 72 hours. Specific attribution connecting Robert's gifts to program outcomes. Direct acknowledgment of his survey response. Phone preferred over email.
Channel Recommendation · When Email Is No Longer Effective
Sarah Chen · Behavioral Disengagement
Email open rate stable. Click-through collapsed. Email is no longer reaching her emotionally. Recommended channel: phone call from Development Director within 5 days. If no response: in-person meeting request via LinkedIn.
Robert Kim · Explicit Disengagement
Survey language indicates impact uncertainty. He told you what he needs: proof his gifts matter. Recommended channel: personal phone call acknowledging the survey response directly, followed by impact report within 72 hours.
Two Signal Types. Both Pre-Lapse. Only One Is Visible Without a System.
Behavioral signals require a system that tracks patterns over time. Explicit signals require someone reading every communication with retention risk as the frame. Most organizations catch neither until the gift stops.
Behavioral Signal · Invisible Without Pattern Tracking
The donor does not say anything. The behavior changes.
Opens-without-clicks. Event declines after consistent attendance. Smaller gift at the usual time. Survey skipped after always responding. No single data point is alarming. The pattern across 90 days is the signal. A human reading a CRM cannot see it. A system tracking ratios over time can.
Sarah Chen opened 6 emails in 90 days. Clicked zero times. Prior pattern: 80% click-through. The silence is the signal.
Explicit Signal · Invisible Without the Right Lens
The donor says something. Nobody reads it as a retention risk.
A survey response with distance language. An email reply with ambivalence about the mission. A comment at an event that sounds positive but contains uncertainty. The words are there. They are just not being read by anyone who is looking for pre-lapse language. The signal is explicit. The interpretation requires a system trained to recognize it.
"I am not sure my gifts are making a difference." Six words. Submitted six months ago. Sitting in a spreadsheet. Unread.
1,247 Communications. Every Signal. Nothing Missed.
1,247 Communications Analyzed Per 90-Day Window
Every email opened or ignored. Every survey submitted or skipped. Every event RSVP accepted or declined. Every click-through ratio shift. 1,247 data points per rolling 90-day window, analyzed simultaneously for both behavioral pattern shifts and explicit disengagement language. No communication falls through the reading.
Behavioral Pattern Shift Detection
Opens-without-clicks is not a single email metric. It is a pattern. Sarah Chen's click-through rate dropped from 80% to 3% over 90 days while her open rate stayed stable. That specific pattern, high opens combined with collapsed clicks, is one of the strongest behavioral indicators of emotional disengagement before a gift lapses. Sentiment Analysis detects it before anyone on your team notices.
Explicit Signal Flagging Within 72 Hours
Robert Kim's survey response was submitted six months ago. Sentiment Analysis would have flagged it within 72 hours of submission. The language "I am not sure my gifts are making a difference" is explicit disengagement text. The system recognizes it, flags it, and surfaces the recommended response before six months pass and the relationship deteriorates past the point of easy repair.
Channel Recommendation When Email Fails
Email is not always the right channel for re-engagement. When Sarah Chen's behavioral pattern shows email is no longer effective, Sentiment Analysis recommends the escalation: phone call from the Development Director within 5 days, followed by in-person meeting request if there is no response. The channel recommendation is part of the alert, not a separate decision.
Sending Surveys Is Not the Same as Reading Them
Most nonprofits send donor surveys. Most of those survey responses are read for satisfaction scores and testimonial language. Sentiment Analysis reads every survey response with retention risk as the frame. The same response that reads as neutral or mildly positive in a satisfaction context can contain explicit disengagement language that signals a lapse within 60 days. The frame changes what you see.
Reactive Fundraising vs. Intelligence-Driven Retention.
The difference is whether you see the signal before or after the gift stops.
Reactive fundraising responds to lapse after it happens. Intelligence-driven retention detects the signal 90 days before the lapse and intervenes while the relationship is still retrievable. Robert Kim told you he was leaving six months ago. The question was never whether the signal was there. It was whether anyone was reading it.
From Unread Signals to Acted-On Intelligence
Without Sentiment Analysis
- Robert Kim's survey response sits in a spreadsheet for 6 months. Unread as a retention signal.
- Sarah Chen's click-through collapse is invisible because no one tracks the ratio over time
- 1,247 communications sent and received. Zero analyzed for disengagement patterns.
- Both donors lapse. Both lapse surprises the development team.
- Reactivation emails sent at below 15% response probability
- Development team continues sending email to donors for whom email is no longer effective
- The signals were there. Nobody was reading them as retention intelligence.
With Sentiment Analysis
- Robert Kim's survey response flagged within 72 hours. Personal impact report delivered within the week.
- Sarah Chen's opens-without-clicks pattern detected at 30 days. Phone escalation recommended.
- 1,247 communications analyzed per 90-day window. Every signal surfaced and categorized.
- Both donors receive the right response at the right time before the gift lapses.
- Intervention at the behavioral signal stage, not after the lapse
- Channel recommendation tells the team when to stop emailing and start calling
- The signals were there. Now someone is reading them.
He Told You He Was Leaving. In a Survey Response. Six Months Ago.
Did anyone see it?
Your Donors Are Communicating Their Disengagement Right Now. This Is What Reads It.
Get started and see Sentiment Analysis surface the behavioral and explicit signals across your donor file in the first 90-day window.
Get Started