4-email winback (day 7/14/30/60), segmented by churn reason. Discount logic, sunset survey for ghosts, SMS variants for high-value accounts.
Someone canceled last month. Your inbox already has 12 cancellation notifications this week and you're not going to manually write a thoughtful email to each one. So they go cold, the seat lapses, and three months later they're using a competitor. This tool builds a 4-email winback sequence sent at day 7, 14, 30, and 60 post-cancel — segmented by why they actually left. The price-driven cancel gets different copy than the feature-gap cancel, which gets different copy than the 'switched to competitor' cancel, which gets different copy than the ghosted user who never said why. Each email has a job — week 1 is empathy, week 2 is product news, week 4 is offer, week 8 is sunset survey. Plus discount-offer logic so you don't burn margin reactivating customers who'd have come back at full price anyway, and SMS variants for the high-ACV accounts where an email won't land.
Brief used: “We're a $89/mo design-feedback tool for product designers (think Loom for design reviews). Common cancel reasons: (1) 'too expensive for solo use', (2) 'we use Figma comments now', (3) 'team didn't adopt it', (4) lots of ghosts who never replied to the cancel survey. Time since churn: write for day 7, 14, 30, 60. Original price tier: $89/mo. High-value account threshold: anything that was on the $290/mo team plan.”
## 4-Email Winback Sequence --- ### Email 1 — Day 7 — Empathy, no offer **Tone:** Curious, not needy. The job here is to get a reply, not a re-subscribe. #### Variant A — Price-driven cancel **Subject:** quick question about the price **Preview:** No pitch, no offer — just trying to learn. **Body:** Hey [first name], You canceled [Product] last week and mentioned price was the issue. I'm not writing to pitch you back — I just want to understand. Was it 'I don't get $89/mo of value from this' or 'I get the value but it's not in the budget right now'? Those are very different problems on our end and the answer changes what we build. If you have 30 seconds to hit reply, I'd genuinely appreciate it. — [Founder name] **CTA:** [Reply, no button] #### Variant B — Feature gap / 'we use Figma comments now' **Subject:** Figma comments — fair call **Preview:** Curious what made it click. **Body:** Hey [first name], You mentioned moving to Figma comments — fair call, that's the obvious default if you're already living in Figma all day. Quick question — was it that the workflow was finally good enough in Figma, or that the cost of switching tools wasn't worth the extra features we had? Trying to figure out where we still add real value vs where we're a nice-to-have. No reply needed if you're heads-down, but if you have 30 seconds I'm reading every response. — [Founder name] **CTA:** [Reply] #### Variant C — Team didn't adopt **Subject:** the adoption thing — I want to understand **Preview:** Adoption failure is on us, not on you. **Body:** Hey [first name], You wrote 'team didn't adopt it' on the way out and that's the one I take personally. Adoption is the founder's problem, not the buyer's. Was it that the tool didn't fit the team's actual workflow, or that nobody had time to switch how they already work? I'd love to know which — because the answer changes whether we should fix the product or fix the onboarding. — [Founder name] **CTA:** [Reply] #### Variant D — Ghost (no reason given) **Subject:** no pitch, just curious **Preview:** Two-line reply would help me a lot. **Body:** Hey [first name], You canceled [Product] last week and didn't fill out the exit survey, which is totally fair — I never fill those out either. If you have 30 seconds though, what made you leave? Two-line reply is fine. Helps me figure out what to build next. — [Founder name] **CTA:** [Reply] --- ### Email 2 — Day 14 — Ship news, no offer **Tone:** Sharing, not selling. The job is to remind them you're alive and shipping. **Subject:** what shipped since you left **Preview:** Three things you might've wanted. **Body:** Hey [first name], Since you canceled two weeks ago we shipped a few things you specifically asked for in the trial / mentioned in support: - [Feature 1] — the thing where comments were getting buried; fixed. - [Feature 2] — Figma sync got faster (the lag was real). - [Feature 3] — you can now [thing]. Not pitching — just thought you'd want to know. The door is still open whenever. — [Founder name] **CTA:** [See what shipped] (links to changelog) *Note:* For the 'ghost' segment, swap 'specifically asked for' to 'customers asked for' since you don't know what they wanted. --- ### Email 3 — Day 30 — Offer (segmented) **Tone:** Direct, no fake urgency. The job is to give a reason to come back now. #### Variant A — Price-driven cancel (offer fires) **Subject:** 30% off if you come back this month **Preview:** Anchor: $89 → $62 for the next 3 months. **Body:** Hey [first name], A month ago you left because $89/mo wasn't working. If price is still the blocker, here's an offer I can only run once: 30% off for 3 months — $62 instead of $89 — if you reactivate by [date 30 days out]. After that it goes back to full price, no extensions, no second discounts. Reason for the offer: I'd rather have you back at $62 than at $0. [Reactivate at $62/mo] — [Founder name] *P.S. If even $62 is too high, just hit reply — there's a solo tier I can move you to that we don't advertise.* #### Variant B — Feature gap / Figma cancel (NO discount, value re-pitch) **Subject:** what Figma comments still don't do **Preview:** Not a discount — a use case you might be hitting now. **Body:** Hey [first name], No discount in this email — I don't think price was your problem. One thing that comes up from people who left for Figma comments and then came back: when you're getting feedback from stakeholders who aren't in Figma (PMs, founders, marketing), the comment-thread-in-Figma flow falls apart. That's the wedge where we still beat Figma every time. If you've hit that, the door's open and your old account is still there. No reactivation friction. [Sign back in] — [Founder name] #### Variant C — Team didn't adopt (NO discount, onboarding-led) **Subject:** if you want to try again, I'll do the onboarding myself **Preview:** 30-min team kickoff, on me. **Body:** Hey [first name], A month ago you said your team didn't adopt the tool. I want to fix that. If you're open to it, I'll personally run a 30-min onboarding session with your team — I'll demo the parts that usually click, set up your project templates, and stay in your Slack for the first 2 weeks. Free, no pitch, no obligation to re-subscribe. [Book 30 min] or just reply with 3 times that work. — [Founder name] #### Variant D — Ghost (soft discount) **Subject:** 20% off + one ask **Preview:** Anchor: $89 → $71. Plus a question. **Body:** Hey [first name], Soft offer: 20% off your next 3 months if you reactivate by [date]. The ask: if you do come back, would you tell me why you left in the first place? It helps me more than the MRR does. [Reactivate at $71/mo] — [Founder name] --- ### Email 4 — Day 60 — Sunset survey + door open **Tone:** Goodbye-but-not-really. No offer. **Subject:** last email, promise **Preview:** One question, 10 seconds. **Body:** Hey [first name], This is my last winback email — I won't send another. Two things before I stop: 1. If you have 10 seconds, hit reply with the one thing that would have kept you. Doesn't matter how long ago you canceled — I read every one. 2. The door is open. If you ever come back, your old project history is still there, no rebuild needed. Thanks for the time when you used it. — [Founder name] **CTA:** [Reply with one thing] --- ## SMS variants (high-ACV only — $290/mo team plan accounts) **After Email 1 (day ~8):** 'Hey [first name] — [Founder] from [Product]. Saw your team canceled last week. Not pitching, just want to understand what broke for you. 2 mins to chat this week? — [Founder]' **After Email 3 (day ~31):** 'Hey [first name] — quick one. I sent an email with a comeback offer last week. Wanted to make sure it didn't get buried. Worth a call before it expires? — [Founder]' --- ## Churn-reason tagging guide (for your CS/ops team) When a cancel comes in, tag it in your CRM/billing with ONE of: - `churn_price` — they said price or budget anywhere in the reason. - `churn_feature_gap` — they said missing feature, switched to competitor, or 'just use [other tool]'. - `churn_team_adoption` — they said team didn't adopt, didn't use it enough, low usage. - `churn_ghost` — no reason given or exit survey skipped. The sequence above auto-routes based on this tag. Tag the cancel within 24h or the day-7 email goes out as the ghost variant by default.
Static example — your run uses Claude live on your specific brief.
Subscription founders watching MRR churn that's not getting any attempt at recovery, growth leads who have a 'win back' folder of stale drafts, customer-success teams handling annual renewals who need a non-pushy cadence for non-responders. Not for: free product users (this is for paying churn) or first-week-trial churn (different sequence, different psychology).
A 4-email winback sequence sent at days 7, 14, 30, and 60, each with a clear job and segmented variants: (1) day 7 — empathy/curiosity, no offer, variants for price/feature-gap/competitor/ghost, (2) day 14 — what's shipped since they left, no offer, (3) day 30 — discount offer with anchor logic so the discount only fires for segments that justify it, (4) day 60 — sunset survey for ghosts and 'door always open' close for everyone else. Each email includes subject line, preview text, body, CTA, and a 2-line note on tone. Plus SMS variants for accounts above your high-ACV threshold (text after email 1 and email 3 only), and a churn-reason tagging guide so your team knows which segment to file each cancellation into.
You've never run winback. MRR churn is a leaky bucket and nobody emails the ones who leave. Get a sequence live in a week using the export.
You raised prices and a wave of customers churned 'on price'. Use the price-segment variants to test offer-vs-no-offer and learn which churners will come back at the new price anyway.
Your CS team can't personally write every lapsed enterprise account. Use the high-ACV SMS variants plus the email cadence to keep the touchpoints warm without burning rep time.
Annual subs that didn't renew get treated like total churn but they're more recoverable. Use the 30-day discount window to win them back before they fully migrate to a competitor.
The anchor logic exists for exactly this reason. The discount only fires for the price-driven cancel (where it's a price problem you can solve) and the ghost segment (where you have no signal and the cost of giving 20% off is lower than the cost of permanent churn). Feature-gap and adoption-failure cancels do not get a discount — those are not price problems.
Then most of your sequence will run on the ghost variant, which is built to extract the reason in email 1. Over time you build a tagged dataset. Six months in, you'll have segmented data and the sequence gets sharper. The first run is mostly a research exercise.
60 days is when winback ROI typically drops below the cost of sending. Past day 60 you're spamming, and the door-open framing in email 4 does more than another 'come back' email at day 90. Re-engage them in 6 months with a product-news email if you ship something material — that's a different sequence.
You can run it manually if you have under 30 cancellations a month — just schedule the emails by hand. Past that, you'll want Customer.io, Loops, or even a simple Postmark template setup with delays. The sequence is tool-agnostic; the segmentation is what matters.
Yes. You get an anonymous preview instantly with no signup. Drop your email and you unlock 3 full-length runs per month for Churn Winback Sequence — no credit card. Unlimited runs are $49 one-time, or $19/mo for every tool.
Paid ($49 one-time) unlocks unlimited runs for Churn Winback Sequence, longer outputs from Claude Sonnet, full exports, and priority generation. $19/mo unlocks every tool on JustNeeda.
Free runs render in-browser and can be copy-pasted. Paid unlocks copy-to-clipboard, Markdown, and plain-text exports — and history of every run tied to your account.
No. Every run hits Claude live with your specific input. We don't reuse outputs across users. Your input stays private to your session and account.