These are the features subscription merchants ask dunning tools for, drawn from competitor feature lists (Churn Buster, Churnkey, Paddle Retain, Baremetrics, Stunning, FlyCode) and from India-specific payment rules. Each is classified with the same honesty as your own list. Being in demand doesn't mean you should build it. Most of these are already offered by someone.
Legend: TS = Table stakes, D = Differentiator, X = Distraction, L = Liability. Priority: P0 = v1 of the Razorpay wedge, P1 = next, P2 = later, — = don't build.
| Platform | Failure events | Update-payment link | Trigger a retry yourself | Native recovery already | Verdict for a third party |
|---|
| Stripe | invoice.payment_failed webhook with attempt_count (verified) | Yes (hosted pages, customer portal) | Yes, pay an invoice via API | Smart Retries, emails, card updater, bundled in the 0.7% fee | Full access, full competition |
| Paddle Billing | Subscription and transaction webhooks (names not verified) | Yes: the update-payment-method endpoint returns the past_due transaction for checkout (verified) | Not verified | 7 retries over 30 days, plus free Retain | Access is fine, but there's nothing left to sell |
| Lemon Squeezy | subscription_payment_failed (returns an invoice object; medium) | Yes, the hosted recovery page | No retry endpoint (medium) | 4 retries over 2 weeks, dunning emails | Messaging-only tool on a shrinking base |
| Razorpay | Subscription states pending and halted (verified); webhook names not verified | Payment links and mandate re-authentication (to verify) | Unverified. Limited by the 24-hour pre-debit rule anyway | "Smart Payment Retries" (details not published), email on failure | Open. Build on messaging and mandates. |
| Shopify subscription apps | Via each app (Recharge, Skio and others) | Via each app | Only the app that owns the contract bills | Per-app retries | One integration per app. Churn Buster already has 15+. |
4.2 Demanded features, classified
A. Recovery core
| Feature | Demand signal | Category | Priority | Notes |
|---|
| WhatsApp recovery message with a UPI pay-now link | High. Indian customers read WhatsApp. | D | P0 | Utility template, merchant's brand, opt-in captured at checkout or via Razorpay's notification consent. |
| SMS fallback (DLT-registered) | Medium | TS | P0 | Register templates under the merchant's DLT entity, or yours as a telemarketer (verify which). |
| Email sequence | Expected | TS | P0 | Cheapest channel, and weakest in India. |
| "Switch to UPI Autopay" link when a card mandate keeps failing | Medium-high | D | P0 | Moves customers to the method that fails least. Failure rates by method are unverified. Measure them yourself. |
| Retry orchestration (timing, count) | Expected globally | TS / L | P2 | Only if the Razorpay API allows it and the pre-debit rule leaves room. Don't promise it. |
| Failure-reason decoding (bank decline, mandate cancelled, insufficient funds, extra authentication needed) | High among operators | D | P1 | Different reasons need different messages. |
| Multilingual templates (Hindi, then Tamil, Telugu, Bengali, Marathi) | Medium | D | P1 | Global tools localise for Europe, not India. |
| Human calling / collections agents | Asked for by large merchants | X / L | — | Gravy's model. Staff-heavy, lower margin, and conduct risk. |
B. Prevention (India's "pre-dunning")
| Feature | Category | Priority | Notes |
|---|
| Mandate-health monitor: mandates expiring, cancelled in the UPI app, or near their maximum amount | D | P0 | New. No global tool models UPI mandates. |
| Heads-up before charges above ₹15,000 ("approve the renewal in your bank app") | D | P1 | Annual plans hit this limit. A failure is avoidable if the customer is warned. |
| Pre-debit notice companion: a merchant-branded WhatsApp message alongside the bank's notice | D | P1 | The customer sees who's charging them and why, which cuts "unrecognised charge" cancellations. Check it doesn't conflict with bank notices. |
| Expiring-card emails (international cards) | TS | P2 | Only matters for merchants with foreign customers. |
C. Proof and reporting
| Feature | Category | Priority | Notes |
|---|
| Holdout group (10% on default handling) | D | P0 | The only honest way to claim incremental recovery. |
| Monthly "₹ recovered" report with the holdout comparison | D | P0 | The retention engine. Email plus PDF is enough at first. |
| Leakage audit (read-only, last 90 days) | D | P0 | The sales hook. Run it by hand at first. |
| Real-time dashboard | TS | P1 | Expected by Scale merchants. |
| Slack or email alerts on large failures | TS | P1 | Cheap to add. |
| Quarterly Indian subscription failure report (anonymised, across merchants) | D | P1 | Content and a data moat. Needs merchant consent in the DPA. |
D. Retention beyond failed payments
| Feature | Category | Priority | Notes |
|---|
| Cancel flows with save offers | D, crowded | P2 | Churnkey, Churn Buster, Paddle Retain. Needs merchant app code. |
| Monthly-to-annual upgrade offers ("term optimisation") | D | P2 | Paddle Retain has it. It lifts merchant LTV, but it's a different buyer conversation. |
Win-back of halted subscriptions after 30+ days | D | P1 | Cheap: same channel, different copy. |
| Pause instead of cancel | TS | P2 | Needs merchant plan support. |
| Churn-reason surveys | TS | — | Stripe includes cancellation surveys. |
| Feature | Category | Priority | Notes |
|---|
| Cashfree Subscriptions / PayU integration | D | P2 (month 15+) | The second rail for $20k+. Same rules, same channel. |
| Indian SaaS on Stripe with India-issued cards | D | P2 | Stripe doesn't retry India-issued cards (verified). A narrow but real gap. |
| Chargebee + Razorpay merchants | X | — | Chargebee has its own dunning. |
| Paddle, Lemon Squeezy, Shopify apps, general Stripe | X | — | See 4.1 and 02. |
F. Trust and operations
| Feature | Category | Priority | Notes |
|---|
| OAuth or partner-app connection instead of raw API keys | TS | P1 | Check the Razorpay partner terms in week 1. |
| Encrypted key storage, audit log, role-based access | TS | P0 | Non-negotiable. Merchants will ask. |
| DPA template, DPDP-ready privacy notice, data-retention settings | TS | P0 | You process the merchant's customers' personal data. |
| SOC 2 | X until Scale merchants ask | — | Too costly before $10k MRR. |
| Holding or routing funds | L | — | Payment aggregator territory under RBI. |
4.3 Net effect
Of about 35 features, about 12 are real differentiators in the Razorpay wedge, and 10 belong in v1 (the P0 rows):
- WhatsApp recovery with a UPI pay link;
- SMS fallback and email;
- the "switch to UPI Autopay" link;
- the mandate-health monitor;
- the holdout group;
- the monthly ₹ report;
- the leakage audit;
- secure key storage and the DPA pack.
Everything global (retries, cancel flows, dashboards) is table stakes you add later, or someone else's product.