Find subscribers who have gone quiet and send them a re-engagement flow, with a control group that gets nothing
A Danish retailer used a vendor's explainable AI to find paid loyalty members who had been active and then went dormant, sent them a personalised flow, and compared them with a group that received no flow.
Evidence: The author reports this. We have not checked it beyond reading the source.
The business problem
Subscribers who quietly stop using a service cancel later, and a general churn score does not say who to contact or why.
What was tried
Define dormancy from behaviour, here moderate activity over the previous 60 days followed by 30 days of inactivity. Use a model whose reasoning you can read so each segment can be judged. Send the dormant segment a personalised re-engagement sequence, and hold back a control group that receives nothing. Compare retention, site engagement and orders after 30 days. The retailer ran about ten similar predictive flows in six months, each with a control group.
What was reported (positive)
After 30 days retention was 29 percentage points higher than the control group, web engagement was up 39 percent and orders up 20 percent. The article says the retention gap is absolute, not a 29 percent relative gain.
Limitations
Published by the software vendor's case study and not independently audited. It leaves out sample size, how people were assigned, retention rates in each group, confidence intervals, discounts, campaign cost and profit, and the message content is private. The company is a large retailer with a loyalty programme, so the method transfers to a small subscription business but the scale does not. The write-up itself says AI was used to help draft it.
What you need
Behavioural data per subscriber, a way to send a targeted flow, and enough subscribers to hold back a control group.
Sources
- Industry Contents (case study of a Subsets customer; vendor-supplied) ↗ Vendor case study, published October 14, 2026
Source published: October 14, 2026. Last reviewed here: October 14, 2026. Spot a mistake? Tell us.
Related workflows
Have Claude plan and prompt an AI page builder to redesign a landing page, then A/B test it with 99 percent significance before rolling it out
A testing-tool company let Claude write the strategy, copy and builder prompt for a landing page, built it in an AI page builder, and tested it against the current page, where it converted 34 percent better.
Run a sales-email triage agent on OpenClaw for 30 days and plan for the five things that break: memory bloat, silent crashes, captcha stalls, model drift and daylight-saving schedules
A founder of an agent-hosting company describes a single OpenClaw agent on a small server that triaged sales email for a month, with a day-by-day account of what failed and how each problem was fixed.
Give your coding agent an email marketing playbook: an open-source Claude Code skill for audits, flows, copy and compliance checks
A founder of several email businesses published a free, open-source knowledge base and Claude Code skill that lets an agent audit an email account, build welcome, cart and win-back flows, draft copy and check compliance.
Research each prospect before a discovery call with Claude, and keep outreach and trust-building steps human
A sales leader who helps technical B2B founders pastes a prospect's name, company and website into Claude before a call to prepare, links Claude to Notion and Gmail, and avoids automated outreach.