{ "schema": "aiblocks-page/1", "id": "subscription-churn-watcher", "type": "agent", "name": "Subscriber cancellation and retention reviewer", "summary": "Review when subscribers cancel and what they said. Compare earlier changes and propose a focused retention fix.", "headline": "Subscriber cancellation and retention reviewer", "subhead": "Share dated cancellations, customer reasons, subscriber counts and sent-email records. It reviews the weak stage, checks the last change and prepares relevant flow or page drafts. You get a measured baseline and proposed changes. Thin samples stay hypotheses, missing reasons block a cancel-flow draft and timing alone does not reveal why someone left.", "job": { "id": "keep-customers", "name": "Keep customers coming back" }, "level": "intermediate", "thinking": "high", "tested": true, "updated": "2026-11-03", "first_published": "2026-10-21", "inputs": [ { "type": "spreadsheet", "required": true, "label": "Spreadsheet", "what": "a spreadsheet with its data, formulas and charts" }, { "type": "performance_sheet", "required": false, "label": "Performance sheet", "what": "ad or store performance export for a closed date window, with definitions" }, { "type": "voc_reviews", "required": false, "label": "Voc reviews", "what": "customer words from reviews, tickets, surveys or calls, with source and date" }, { "type": "store_capture", "required": false, "label": "Store capture", "what": "saved evidence from a storefront visit: screenshots, page text and what was observed, with dates" }, { "type": "product_data", "required": false, "label": "Product data", "what": "one product's facts: price, variants, ingredients or materials, how it is used, photos by job" }, { "type": "brand_brain", "required": false, "label": "Brand brain", "what": "the brand's facts, voice, products, offers (with effective dates), claims register, personas and look" }, { "type": "offer", "required": false, "label": "Offer", "what": "a live offer with price, terms, effective_from/to and a current flag" } ], "outputs": [ { "type": "audit_report", "required": true, "label": "Audit report", "what": "findings with evidence, severity, missing inputs and next actions" }, { "type": "plan", "required": true, "label": "Plan", "what": "a dated plan: what to do, in what order, with a decision rule for each step" }, { "type": "email_flow", "required": true, "label": "Email flow", "what": "an email/SMS flow: trigger, timing, messages, segments" } ], "you_give": [ "Last month's cancels with date, order number and reason", "Revenue per customer by month and a first-order subscription flag, if you have them", "Each email you sent to subscribers, with the cancels and starts that followed", "Your save rate from the cancel portal", "Your products, brand file and offer limits, for the emails" ], "you_get": [ "A score out of 100 in four parts, against last month", "Cancels by days since the first order, with the biggest spike named", "A verdict on last month's fix, helped, no change or worse", "One fix, written and ready for you to approve", "The one number to watch next month" ], "use_when": [ "You sell subscriptions and want one monthly check on where people leave.", "Churn moved and you do not know which part changed.", "You made a change last month, such as new billing emails, and want to know if it helped." ], "skip_when": [ "You have only a few hundred subscribers or fewer: the day-by-day chart is too thin, so treat the result as a rough guess or run the Subscription churn auditor alone.", "You sell one-off orders with no repeat billing: use a plain repeat-purchase check in your shop reports instead.", "You want emails sent or your subscription app changed for you: this agent only lists changes, so make them yourself in your own tools." ], "output_sample": { "input": [ "Last month's cancels with date, order number and reason", "Each email you sent to subscribers, with dates", "Your save rate from the cancel portal", "Cancel reasons in your customers' own words (optional)", "Last month's saved summary (optional)" ], "prerequisites": [ "An AI that can read files, such as Claude Code, if you want it to keep last month's summary", "Your brand brain and offer, only if it needs to write emails (optional)" ], "output": [ { "name": "This month", "contains": "Your churn score out of 100 and the four part scores, each shown against last month." }, { "name": "Last month's fix", "contains": "Whether the part you fixed helped, made no change or got worse, with the numbers." }, { "name": "Where subscribers leave", "contains": "Cancels grouped by days since the first order, with the biggest spike named." }, { "name": "The fix chosen", "contains": "The weakest part, the skill that ran for it and why it was picked." }, { "name": "Changes to approve", "contains": "Each change, where it goes and why, with the most important first." }, { "name": "Watch next month", "contains": "The one number that shows whether the fix worked." }, { "name": "Files and missing", "contains": "Each step's saved file and every line marked MISSING." } ] }, "output_shape": "THIS MONTH\n Churn score /100 vs last month /100\n Reminder and receipt /25 | Save rate /25 | Campaign cancels /25 | Value gap /25\n\nLAST MONTH'S FIX\n : | \n\nWHERE SUBSCRIBERS LEAVE\n day 0-1: | days 2-6: | days 7-14: | around 28: | first rebill: | biggest spike: