When the money stops
Stripe Payments Failing: Find Which of Three Breaks You Have Before Anyone Touches the Code
The sales report says zero since 9:40 this morning. That number is a symptom, not a diagnosis. A payment can fail at the card, at Stripe, or in the software that was supposed to hear about it, and each one has a different first move.
November 20, 2026 · 11 min read
- 09:38 deploy goes live with new environment settings
- 09:40 live events fail signature check, 400 returned
- 10:55 owner reports no new orders since 09:40
- 11:20 Event deliveries tab and server logs saved
- 11:34 live signing secret restored on the server
- 11:41 handler now records event IDs first
- 11:58 missed events resent from the Dashboard
- 12:15 orders reconciled; test charge and refund pass
The short answer
When Stripe payments fail, first decide which break you have: cards are being declined, Stripe charged the card but your software never processed the event, or your code never reached Stripe at all. Check the Stripe Dashboard's payments list and the webhook destination's Event deliveries tab before changing anything. Stripe retries live-mode webhooks for up to three days, so missed orders can usually still be recovered.
On this page
- Zero sales can mean three different breaks
- Sixty minutes, in the order that protects revenue
- The webhook is a delivery truck nobody watches
- Subscription renewals that fail without a sound
- Four fixes that cost more than the outage
- Replaying the events your software missed
- When to handle it yourself, and when to call us
Zero sales can mean three different breaks
Owners call this "Stripe is down." Sometimes it is. Just as often, Stripe is fine and something on either side of it has broken. The fastest way to the right fix is to sort the failure into one of three bins before anyone opens a code editor, and the Stripe Dashboard is enough to do the sorting.
| What you see | Where to look first | What it usually means |
|---|---|---|
| Customers say their card was refused; the Dashboard shows failed payments | Payments list in the Stripe Dashboard, the decline reason on each one | The card or the bank said no; Stripe and your code may be fine |
| Stripe shows the payment succeeded; your app shows no order, no access, no receipt | Your webhook destination's Event deliveries tab | Stripe charged the card but your software never processed the event |
| Nothing at all in Stripe for the last few hours | Your own server logs and error tracker, then Stripe's status page | Your checkout never reached Stripe: a broken deploy, a rotated or revoked key, a dead server |
| Subscriptions marked past due in bulk | Billing settings and recent invoice.payment_failed events | Renewals are failing and retries are running on their schedule |
The second row is the dangerous one, because the money arrives and the customer still feels robbed. They paid and got nothing. If a checkout page on your website is refusing orders outright, the checkout and payments page covers the storefront side. This page is about the plumbing behind it: subscriptions, webhooks and the custom software that listens for Stripe.
Sixty minutes, in the order that protects revenue
Panic wants to fix things. The first hour is for finding out and freezing the scene, so the fix later is quick and nobody pays twice. Do these in order.
- 1
Open Stripe's status page
Stripe runs a public page titled "Stripe system status" at status.stripe.com. If it reports a problem, your job shrinks to telling customers and waiting. If it looks normal, the cause is probably closer to home.
- 2
Read one failed payment from end to end
Pick a single customer who complained. Find the payment in the Dashboard, read its status and decline reason, and note the exact time. One real case beats a hundred guesses.
- 3
Check the webhook destination
Open the destination's Event deliveries tab. A column of failures means Stripe is knocking and your server is not answering properly. Note the response code Stripe recorded.
- 4
Ask what changed
A deploy yesterday, a new server, a renewed domain certificate, a rotated key, a plugin update. Payment breaks almost always follow a change. Write the list down with times.
- 5
Save the evidence
Export or screenshot the failed deliveries and copy the server error logs somewhere safe before anyone restarts or redeploys. Logs that rotate away cannot explain anything tomorrow.
- 6
Tell customers something true
A short note on the site or in email: payments are being looked at, nobody will be charged twice, orders placed will be honored. Our outage communication page has the wording.
If the trail leads to a deploy that went out just before the failures started, rolling back usually comes before diagnosing; our page on a software update that broke everything explains how to do that without making it worse.
The webhook is a delivery truck nobody watches
Custom software rarely learns about a payment from the checkout page. It learns from a webhook: Stripe sends an event to an address on your server, and your code fulfills the order, opens the account or emails the receipt. When that address breaks, customers are charged and nothing happens next. Stripe's webhook documentation, read in November 2026, sets out the rules that matter in a crisis.
- Retries are generous but finite. Stripe says it "attempts to deliver events to your destination for up to three days with an exponential back off in live mode," as of November 2026. Sandbox events are retried only "three times over the course of a few hours."
- A redirect is a failure. Stripe treats redirect responses, 4xx and 5xx codes, TLS errors and timeouts as failed deliveries. A new www redirect or an expired certificate can break webhooks with no code change at all.
- Speed matters. Stripe says your endpoint "must quickly return a successful status code (2xx) before any complex logic that could cause a timeout." A handler that emails, writes to three tables and calls a CRM before answering is a timeout waiting to happen.
- Order is not promised. Stripe "doesn't guarantee the delivery of events in the order that they're generated," and endpoints "might occasionally receive the same event more than once." Stripe says to log processed event IDs.
- Each endpoint has its own signing secret, and the test and live secrets differ. Deploying with the test secret in production makes every live event fail verification.

A webhook failure is an integration failure, so it gets the same treatment as any broken connection: find the response code, fix the cause, replay what was missed. Our API integration repair page walks through 401, 403 and 429 responses, which turn up when a key or permission has changed underneath a webhook handler.
Subscription renewals that fail without a sound
A subscription business can lose a month of revenue in silence. Nobody sees a broken page. Renewals simply fail, one card at a time, while the retry schedule runs. Stripe's Smart Retries documentation, read in November 2026, says that "For new Stripe accounts, Stripe enables Smart Retries by default and configures it for up to 8 retries within 2 weeks." Retry policy windows offered, as of November 2026, are "1 week, 2 weeks, 3 weeks, 1 month, or 2 months."
Hard declines are not retried
Stripe does not retry when the issuer returns a hard decline code. Its list includes incorrect_number, lost_card, stolen_card and authentication_required. Those customers need to update their card; no schedule will rescue them.
No card, no retry
Stripe also does not retry when no payment method is available, for India-issued cards, or when a Connect account is disconnected. A batch of customers with no card on file is a data problem, not a Stripe outage.
Custom schedules are short
Stripe says you can configure up to three retries, each a set number of days after the last, as of November 2026. Follow failures through the invoice.payment_failed event; its attempt_count shows how many tries have run.
The last attempt is final
"After the final payment attempt, Stripe makes no further payment attempts." The subscription then follows your setting: canceled, marked unpaid, left past due, or paused where eligible.
Bank debits play by other rules
For ACH Direct Debit, Stripe retries only for insufficient funds, and at most 2 retries within 40 days, as of November 2026.
If your app cuts off access the moment a renewal fails, a quiet retry problem becomes a loud support problem by lunchtime. Check what your code does on past due before you check anything else. WooCommerce stores with renewal trouble have their own fix in our subscriptions not renewing guide.
Four fixes that cost more than the outage
- Do not delete the webhook endpoint and create a fresh one. Stripe says that if a destination "has been disabled or deleted when we attempt a retry, we prevent future retries of that event." Deleting it throws away the retries that would have recovered your orders.
- Do not set the timestamp tolerance to zero to quiet a replay warning. Stripe's libraries default to a tolerance of 5 minutes, as of November 2026, and Stripe says plainly: "Don't use a tolerance value of 0."
- Do not refund and recharge customers by hand while events are still arriving. Duplicate events are normal, and a handler that does not record event IDs will fulfill or charge twice.
- Do not roll the signing secret in a hurry with no overlap. Stripe lets you delay the old secret's expiration for up to 7 days, as of November 2026. Use the overlap, or the fix becomes the next outage.
If you suspect the secret or your server was actually compromised, the job changes. Rolling the secret is right, and so is reading our page on the first hours of a data breach, with your attorney and insurer involved early.
Replaying the events your software missed
Once the endpoint answers properly again, Stripe's own retries will deliver much of the backlog. For the rest, Stripe's guide to managing webhook endpoints and its webhook docs give two manual paths, with limits worth writing on a sticky note, as of November 2026.
Dashboard Resend
- Works for up to 15 days after the event was created
- Good for a handful of specific orders
- No developer needed to click it
Stripe CLI resend
- Works for up to 30 days after the event was created
- Stripe returns only events created in the last 30 days
- Suits a batch, run by someone who knows the handler
Replay only after the handler records event IDs, or every resend risks a second fulfillment. Then reconcile: compare Stripe's successful payments for the outage window against orders in your system, and contact anyone who paid and received nothing. An honest email sent the same day does more for trust than a refund sent next week. For a picture of how a clean Stripe setup is wired from the start, our Stripe payments page shows the parts.
When to handle it yourself, and when to call us
Handle it yourself when Stripe's status page shows an incident, when the failures are hard declines from a few customers, or when the Event deliveries tab shows one obvious cause you can fix in settings, such as an endpoint URL that now redirects. Questions about Stripe's side of the line belong with Stripe's own support. You do not need a developer to read a decline reason.
Call us when the webhook handler is custom code nobody on staff understands, when the person who wrote it is not answering, or when orders and payments no longer match and you cannot tell how far back. We respond the same business day (our hours are on the contact page). Repairs are from $575, depending on complexity, with a written estimate before work starts; emergency pricing explains what moves the number. A broken payment connection is restored the same day once we have access to Stripe, the server and the code.
What we do, in order: read the deliveries and logs, fix the endpoint, add event-ID logging if it is missing, replay within Stripe's windows, reconcile, and test a live charge and refund. It is part of our emergency software repair work, and longer-term rebuilds fall under web app development and API integrations. If the whole app is down, not just payments, start with the web app down checklist. DataCram is a San Antonio studio building for the web since 1999 (about us, our work); call (210) 346-0848, or see every scenario on the emergency repair hub.
Questions owners ask in an emergency
Stripe's webhook documentation says it attempts delivery for up to three days with an exponential back off in live mode, as of November 2026. Sandbox events are retried three times over a few hours. If the destination is disabled or deleted when a retry comes due, Stripe stops retrying that event, so fix the endpoint rather than deleting it.
Usually the webhook that tells your software about the payment failed. Open the destination's Event deliveries tab in the Stripe Dashboard and read the response code. Redirects, 4xx and 5xx codes, TLS errors and timeouts all count as failures to Stripe. Fix the cause, then resend the missed events within Stripe's windows.
Yes, within limits. As of November 2026, Stripe's Dashboard Resend works for up to 15 days after the event was created, and the Stripe CLI resend command works for up to 30 days. Make sure your handler records processed event IDs first, so a resend never fulfills an order twice.
Stripe does not retry hard declines such as lost_card, stolen_card or incorrect_number, payments with no payment method available, India-issued cards, or disconnected Connect accounts. Those customers need to update their card. Stripe's Smart Retries handles the rest, set by default on new accounts to up to 8 retries within 2 weeks, as of November 2026.
No. Stripe warns that without verification an attacker could send fake events to trigger order fulfillment, account access or record changes. Verification failures usually come from the wrong signing secret, mixing test and live secrets, or a framework that altered the raw request body. Fix that instead.
DataCram's software and integration repairs are from $575, depending on complexity, with a written estimate before work starts. We respond the same business day, and a broken payment connection is restored the same day once we have access to Stripe, the server and the code. Stripe's own fees are separate and set by Stripe.
Keep reading
- The website offerEmergency Website Repair: Who to Call When the Site Is Down, and What Happens After You DoWebsite down and no one to call? A same-business-day reply, fixed WordPress repair prices, from $575 for other platforms, and the checks before we sign off.
- The software offerEmergency Software Repair: Triage, Containment and a Written Root Cause for Broken Custom SoftwareCustom software or a web app broken today? Triage, stop the damage, restore, find the root cause, and a written estimate first. From $575, depending on scope.
- The mobile offerEmergency App Repair: An iOS and Android Developer for Crashes, Store Trouble and Expired CertificatesApp crashing, rejected or pulled from the store? Crash triage, store fixes, certificates and the access we need, from $575 with a written estimate first.
- Undo now, investigate laterSoftware Update Broke Everything? Roll Back First, Then Find Out WhyA release broke production. Roll back to the last good version, check what a rollback cannot undo, then diagnose away from the live site. The first hour.
- Silence from APNsPush Notifications Stopped Working on iOS: Expired APNs Certificates, Stale Tokens and the First HouriOS push notifications stopped? Learn whether you use an APNs certificate or a token, what expired, and what your server's error says, before you rebuild.
- Yesterday's build still existsNext.js or React Site Down After a Deploy: Roll Back First, Debug SecondA Next.js or React site broke after a deploy. How Vercel's Instant Rollback works, what it will not undo, and how to read a failed build without guessing.
- Everything in Emergency RepairSee the section
From the Insights desk
- Next.js or PHP? The Better Question Is Who Patches Your SiteApache, NGINX and PHP all shipped urgent fixes between November 11 and 19, 2026. PHP is not the villain, and Next.js has patch days too. The real difference is who installs the fix, how fast, and how much of your site runs code at all.
- The Optimizely Hack Was Not an Outage. It Was a Delivery.On November 16, a flaw in Optimizely's sign-up flow let an attacker slip card skimmers, wallet swappers and iPhone exploits into the testing scripts of 178 organizations, the BBC and Zoom among them. How it happened, what it did, and the fixes, in order.
Get the payments flowing again
Tell us what Stripe shows and what your app shows. We reply the same business day with what we think broke, what we need access to, and a written estimate.
Which of these sounds like you? (pick any)
Rather talk now? Call (210) 346-0848.