Key takeaways
- SunMed's month 3 retention on Skio went from 61% to 73% after RetainUP's subscription flows went live on 3 February 2026.
- The rebuilt Skio cancel flow saves 24.8% of sessions against 14.8% for the old one, with no new discount.
- Failed payment recovery went from 52.1% to 77.6%, and net subscriber growth went from minus 48 a month to plus 340.
SunMed sells CBD supplements online and in more than 300 stores. Subscriptions run on Skio, emails on Klaviyo.
The brief was one number. Show retention from the Skio data, then a plan to raise it.
We kicked off on 28 January 2026. Eight subscription flows went live on 3 February. Month 3 retention went from 61% to 73%.
What was SunMed's subscription problem?
SunMed was losing subscribers about as fast as it signed them up. In Q3 2025, net subscriber growth averaged minus 48 a month.
New subscribers came in every month. More left.
The leak was not only new buyers. Klaviyo held 5,090 past subscribers who had cancelled after three or more orders. They liked the product enough to buy it three times, then stopped.
About 1 in 5 renewals led to a cancel flow session. And the retention emails were thin: a heading, an image, a line of text. No benefits. No cross-sell. No reason to stay.
What did the audit find?
The audit and the cancel flow analysis that followed found three leaks: the cancel flow, failed payments, and the gap between order 2 and order 3.
- The cancel flow. It saved 17% of 2,159 sessions. The top reason was "too much product", a third of all sessions. "I no longer use it" came in 492 times and saved 19.
- What a save was worth. Contact support saved the most people, but earned $10.58 per save after. A product swap earned $86.40.
- Failed payments. The first retry recovered 0 of 213 failed payments. The second recovered 36.
- Order 3. Order 2 retention was strong at 90.9%. Order 2 to order 3 was 59.0%. So about 54 of every 100 new subscribers reached a third order.
Three products made 33% of revenue. That told us where to aim the content.
What did RetainUP build for SunMed?
We built eight Klaviyo flows on Skio events, rebuilt the Skio cancel flow, and reset payment recovery, all without touching the price.
The offer
The offer stayed as it was. SunMed did not want to become a discount brand, so nothing in this build added a new standing discount. Every result below comes from what happens after the first order.
Skio setup
In Skio, we rebuilt the cancel flow and switched failed payments to two-phase smart retries. Skio events feed Klaviyo, so each flow fires on what the subscriber just did: skipped, paused, failed a payment, cancelled.
Cancel flow
The new flow opens with the benefits, then offers a skip before it asks for a reason. "Too expensive" gets two answers to pick from. The copy takes the shame out of leaving before it makes an offer. No new discount.
The screen by screen build is in the cancel flow case study. The method behind it is cancel flow routing.
Payment recovery
Retry 1 recovered nothing, so we moved failed payments to Skio's two-phase smart retries. The payment failed flow runs alongside them: seven touches, with 10% off only at the last one.
Lifecycle flows
Eight flows went live six days after kickoff: onboarding, upcoming subscription, renewal, skipped, paused, cancelled, payment failed and winback.
Every email has three blocks: the main message, a supporting block, and a cross-sell. The flows push one habit. Take it every day. A subscriber who takes it daily runs out on time and reorders.
Between 3 February and 6 April 2026, the flows were credited with $48,619. The upcoming subscription email made $37,517 of it, from 8,573 recipients.
What were SunMed's results?
Month 3 retention went from 61% to 73% on the same Skio account, with no new discount.
- Month 3 retention: 61.46% and 62.98% for the September and October 2025 cohorts. 71.69% and 73.24% for January and February 2026.
- Order 2 retention: 90.9% for Q3 2025 cohorts. 94.5% and 94.7% for February and March 2026.
- Save rate: 14.8% on the old flow across 3,507 sessions. 24.8% on ours across 3,543.
- Recovery rate: failed payment recovery went from 52.1% to 77.6%.
- Cancel flow entry: 19.8% of renewals hit the cancel flow in Q3 2025. 15.5% in March 2026.
- Growth: active subscribers went from 6,075 to 7,582, up 25%. Net growth went from minus 48 a month to plus 340. Subscription revenue went from $384K to $453K a month.
One note on the cohorts. January 2026 was a promo month, with about double the usual sign-ups. February was a normal month, and it posted the highest month 3 number: 73.24%.
Not everything moved. Subscription AOV slipped from $88 to $84. Nothing in this build aimed at order value, and that is the next piece of work.
"Our biggest focus was consistency, and I love that RetainUP built on that. Their work is so much better than what we did, I mean, I can't even compare."
Kristen Osborn, CMO, SunMed
What can you copy from SunMed?
You can copy five moves without changing your price: count cancel reasons, price each save, check retry 1, make the upcoming order email sell, and read month 3 by cohort.
- Count your cancel reasons by sessions and saves. If "too much product" leads, put the skip before the reason screen.
- Check what each save is worth after. A save through support was worth $10.58. A swap was worth $86.40. Lead with the one that pays.
- Look at retry 1 on its own. If it recovers nothing, fix the retry timing before you write another email.
- Make the upcoming order email sell. At SunMed it made 77% of all flow revenue.
- Read month 3 retention by cohort, and mark promo months. A promo cohort can flatter or sink the average. Read subscription cohort analysis for the method.
FAQ
How much did SunMed's retention improve?
Month 3 retention went from 61.46% and 62.98% (September and October 2025 cohorts) to 71.69% and 73.24% (January and February 2026 cohorts). Order 2 retention went from 90.9% to 94.7%.
Which subscription app does SunMed use?
Skio, with Klaviyo for email. The cancel flow and the retries were rebuilt inside Skio. The lifecycle flows run in Klaviyo off Skio events. Our Skio retention setup covers the settings.
How long did it take to see results?
Kickoff was 28 January 2026 and the flows went live on 3 February. The first cohorts after launch hit month 3 in April and May. Net subscriber growth was plus 371 in February and plus 340 in March.
Did SunMed use discounts to get these results?
No new standing discount. The cancel flow took no new discount, and the payment failed flow offers 10% only at the last touch.
How was the cancel flow save rate measured?
Both numbers come from Skio's own flows table, counted the same way: 14.8% across 3,507 sessions on the old flow, 24.8% across 3,543 on ours. The full build is in the cancel flow case study.
How do I find where my own subscription program leaks?
Start with the free Subscription LTV Scorecard. It scores your program on 48 checkpoints from public data and lands in your inbox within 24 hours.
Want SunMed's audit on your own data? The Subscription LTV Roadmap™ reads your cohorts, cancel reasons and failed payments, and hands you the fixes in order, for $1,000. Or start with the free subscription audit.
One email a week on subscription retention: the numbers, the strategy, and what to fix first.
Rahi Jain
Rahi Jain is a subscription LTV specialist who helps $1M to $10M DTC brands reduce subscription churn and maximize LTV on Recharge, Skio, Stay AI and Loop.
More about Rahi
