Subscription Renewal Rate: Start 90 Days Early
Subscription renewals are decided months before renewal day. India benchmarks, a 90-day plan-ending journey, and how to recover failed UPI AutoPay renewals.

Updated November 2026.
To increase your subscription renewal rate, start long before renewal day. Renewals are decided by what members got out of the plan, so the work starts with their first result and the habit that follows. Begin the plan-ending journey about 90 days before an annual plan ends, split members by whether auto-renew is on and whether they still use the app, and lead every message with their own progress. Treat failed UPI AutoPay or card renewals as a separate problem, because a payment that didn't go through isn't a member who chose to leave.
I'm Raviraj Ramavarapu, founder of Compound Lab, a retention agency for D2C brands and consumer apps. At a health app, I owned the journeys for paid members of its coaching and AI plans, including the plan-ending journey that won renewals. This is how that journey worked, what the data showed, and how I'd build one for an Indian app now.
What is a good subscription renewal rate in India?
For a first renewal, about 46% on monthly plans and 22% on annual plans is the regional median. RevenueCat's State of Subscription Apps 2026 reports first-renewal rates by region, and India and South-East Asia sit below North America on every plan.
| Plan | India and South-East Asia | North America |
|---|---|---|
| Weekly, first renewal | 37% | 55% |
| Monthly, first renewal | 46% | 55% |
| Annual, first renewal | 22% | 26% |
The first renewal is the hard one. In the same report, retention climbs 18 to 30 points from the first renewal to the second, and by the third renewal, weekly plans in every region land between 77% and 81%. Members who renew once have decided the plan is part of their life. Members who don't usually decided much earlier.
Where is a subscription renewal won?
In six moments, and only the last one is near renewal day. I use these six questions to find where a plan is losing members.
| Moment | The question it answers | What to send |
|---|---|---|
| First value | Did they get what they paid for? | A guided first session, straight to the main feature |
| Habit | Is the product part of their routine? | Reminders tied to a time they already use it |
| First win | Have they seen real progress? | Their own result, shown back to them |
| Slowdown | Is their usage starting to fall? | A save before they drift, not a discount |
| Pre-renewal | Can you show what changed since they joined? | A recap of their progress, then the renewal terms |
| Exit | Why did they leave, and what happens next? | A short reason question, then a win-back built on their history |

Most apps put all their effort into the fifth moment. By then, a member who never reached the third has already made up their mind. The first two moments are covered in how to improve app retention in India and free trial length.
How does a 90-day plan-ending journey work?
It starts three months before the plan ends and gives each kind of member a different path. This is how the plan-ending journey at the health app was built.
- Split members 90 days out. Every member whose plan was ending was grouped by two things. Was auto-renew on or off, and were they still active in the app?
- Active members got their progress first. The renewal was framed around what they'd already achieved, with the renewal terms after it.
- Inactive members were brought back first. They got a reason to open the app again before they saw any renewal offer. Asking someone to pay again for a plan they've stopped using rarely works.
- Price and message were tested. A lower-price version ran against the regular price, alongside several message styles. The winners became the default.
- Members who still lapsed moved to win-back. Their message was built from their own months of data, which I describe in win-back campaigns.

The split mattered more than any single message. On one annual plan, members who were active 15 to 60 days before their plan ended renewed at 3.9 times the rate of inactive ones, and about two thirds of members were active. For a renewal, that gap says the real work happened months earlier, in the app. How the app picks which of these members a coach should call is in who wins renewals.
Which members should you leave alone?
Members with auto-renew on. They're set to renew anyway, and a discount only gives away money you'd have collected. The plan at the health app treated each group differently.
| Auto-renew | Activity | What they got |
|---|---|---|
| On | Active or inactive | Nothing. Do not touch |
| Off | Very active | The next plan up at the same price, or a small discount to stay on their own |
| Off | Somewhat active | The same choice, with a bigger discount |
| Off | Inactive | A useful first step, like building this week's diet plan, before any offer |
| Paid manually | Active or inactive | Treated like auto-renew off, by activity |
Auto-renew was rarer than most founders assume. Of the annual plans ending one January, only about 10% had auto-renew on. About 24% had switched it off, and 66% had been paid manually with no mandate at all. Monthly plans held on better, with auto-renew still on for 87% of recurring members against 30% on annual plans. So for most annual members, renewal is a fresh purchase decision, not an automatic charge.
The copy was split the same way. Active members got messages about the habit they'd built and what the plan had added since they joined. Inactive members got short, honest lines about having stopped, written in a few tones, including humour and Hinglish, and tested against each other.
How should monthly and annual plans be handled differently?
Monthly plans need the first month to go well. Annual plans need both the first month and the last three.
- Monthly plans have a short runway. The first renewal comes 30 days after the first payment, so the first value, the habit and the first win all have to happen inside that month. Send a progress recap a few days before each renewal, then the payment notice.
- Annual plans lose people early too. RevenueCat found that 35% of all annual cancellations happen in the first month. Treat month one of an annual plan like a trial, even though they've already paid.
- Annual renewals need the 90-day journey. After a year, members have forgotten what the plan cost and what they got for it. The recap has to remind them of both.
- Expensive annual plans renew less. In RevenueCat's data, high-priced annual plans renew at 24% at first renewal, against 37% for low-priced ones. That's one more reason to test a lower renewal price, as the health app's journey did. It also offered monthly members the annual plan at its monthly equivalent, so they could see what they were overpaying.
Why do UPI AutoPay renewals fail, and what should you do?
Some renewals fail because the member decided to leave, and some fail even though they wanted to stay. Separate the two before you send anything.
| What happened | What it means | What to do |
|---|---|---|
| Mandate cancelled after the bank's notice | A choice to leave | Treat it as a cancellation. Ask why, then win-back |
| Debit failed, mandate still active | Usually a balance or bank issue | Let the retries run, then send a link to pay |
| Store payment failed on Google Play or the App Store | Card or UPI problem inside the store | Use the store's grace period to remind them, with a link to update payment |
| Auto-renew switched off before the date | Undecided, still paid up | The pre-renewal recap, with their progress |
| Plan was never set up to recur | A setup error, not a choice | Fix the recurring setting with your payment gateway |
The Indian rules shape this. Under the RBI's E-mandate Framework, 2026, the bank or UPI app tells the user about each recurring debit at least 24 hours ahead, and the user can cancel that debit or the whole mandate. That notice is a cancel button, so your own recap should arrive before it. NPCI rules also push AutoPay debits out of peak hours, 10 am to 1 pm and 5 pm to 9:30 pm, and allow one attempt plus three retries.
Payment providers add their own steps. Razorpay retries a failed subscription charge on each of the next three days, then halts the subscription, after which you have to send the member a link to clear the dues. On the stores, Google Play gives a failed renewal a grace period, followed by account hold, for up to 60 days in total, and Apple offers a billing grace period of 3, 16 or 28 days. In RevenueCat's data, billing errors cause 32.2% of Google Play cancellations, against 15.2% on the App Store, so Android apps should watch this closely.
Check your own setup before you blame the message. At the health app, some monthly plans couldn't auto-renew at all, because the payment gateway setting that marks a plan as recurring was missing. Renewal rates also differed sharply by platform and segment. In one quarter, iOS members below the plan's age cut-off with auto-renew off renewed at about 3%, while Android members above it renewed at about 27%. Read renewals by platform before deciding the copy is the problem.
What does an annual renewal calendar look like?
One job per step, with the payment rules built in.
| When | Job | Channel |
|---|---|---|
| 90 days before | Split by auto-renew and activity | No message, just the segments |
| 60 days before | Active members: their progress. Inactive: a reason to come back | Push and in-app |
| 30 days before | The year in numbers, and the plan options | Email and WhatsApp |
| 7 days before | Price, date, how to renew and how to cancel | Email and WhatsApp |
| Bank's notice, 24 hours before | Nothing extra. Your recap has already arrived | None |
| Renewal day | Debit outside peak hours. Confirm or start recovery | Push or WhatsApp |
| After a failed debit | Retries, then a link to pay | WhatsApp and SMS |
| After the plan ends | Ask why, then win-back with their own data | WhatsApp and email |

Messages that are about the member's own plan, such as the price, date and payment status, can usually go as WhatsApp utility messages, which cost less than marketing ones. Anything that promotes an offer is marketing. I explain the difference in WhatsApp marketing messages.
What should you measure?
Renewal rate by renewal number, with chosen and failed renewals counted separately.
| Measure | What it tells you |
|---|---|
| First, second and third renewal rate, by plan | Where members drop out |
| Renewal rate, active vs inactive members | Whether the problem is the product or the message |
| Renewal rate by auto-renew status and platform | Whether a setup or store issue is hiding in the average |
| Cancellations vs failed payments | How much churn was never a choice |
| Failed payments recovered, and how fast | Whether retries and payment links work |
| Renewals against a holdout | Whether the journey caused the renewals |
Read it by monthly cohort, and keep a small holdout group out of the plan-ending journey. A good renewal month can come from a strong sign-up month a year earlier, not from anything you sent.
Where I fit
Compound Lab works with D2C brands and consumer apps on everything after the first payment, from the first value and habit to the plan-ending journey, payment recovery and win-back. If your renewals are half your revenue and you're not sure where members drop out, book a 30-minute call and we'll find the moment your plan loses them.