As of October 9, 2026, the highest modelled risk on StatusBird's board belongs to Cloudflare, at a 90.0% chance of a non-operational reading in the next 24 hours. BigCommerce sits at 63.7%, Recharge at 53.6% and Shopify at 50.1%, while the other 80 or so services StatusBird watches are either normal, elevated, or high but lower down the list.
StatusBird checks each service every 2 minutes from its own probes, and every figure below comes from those readings over the 90 days ending October 9, 2026, plus the risk model that runs on the same data. The full approach is written up on the methodology page.
Which services carry the highest modelled risk right now?
These are the services flagged high risk, with their 90-day record alongside, so you can see where the risk reading agrees with history and where it does not.
| Service | 24-hour risk | 90-day uptime | Major/critical incidents | 90-day downtime |
|---|---|---|---|---|
| Cloudflare | 90.0% | 95.59% | 4 | 5706 min |
| BigCommerce | 63.7% | 99.86% | 2 | 180 min |
| Recharge | 53.6% | 99.96% | 1 | 46 min |
| Shopify | 50.1% | 99.9% | 2 | 124 min |
| Judge.me | 46.7% | 100.0% | 0 | 0 min |
| Vercel | 40.1% | 99.96% | 1 | 54 min |
| Square | 37.4% | 100.0% | 0 | 0 min |
| Anthropic | 35.0% | 100.0% | 0 | 0 min |
| Gorgias | 35.0% | 100.0% | 0 | 0 min |
Cloudflare is the clearest case. It carries a D grade on 95.59% uptime, 5706 min of downtime across 4 major or critical incidents, and its most recent incident started on October 5, 2026 at 16:34 UTC and did not recover until October 9, 2026 at 13:09 UTC, a run of 5556 minutes. A service that was still recovering on the day of the reading is going to score high. You can watch it live on the Cloudflare status page or read the whole window on its reliability history.
Why is Square flagged high risk at 100.0% uptime?
This is the part that surprises most store owners. Judge.me, Square, Anthropic and Gorgias each recorded 100.0% uptime over the 90 days with 0 major or critical incidents, and all four are flagged high risk anyway, Square at 37.4%. The risk reading is not the 90-day average restated. It responds to recent patterns in the raw 2-minute readings, including degraded and partial states that never grew into a major incident and therefore never showed up in the incident count.
The practical reading is this: a 90-day uptime figure tells you what a vendor cost you last quarter, and a risk level tells you what is wobbling today. They answer different questions. A store owner picking a payment gateway should look at the 90-day column. A store owner deciding whether to ship a theme change this afternoon should look at the risk column.
Does high risk mean a long outage?
No, and this is the second number worth checking before you panic. Likelihood and length are separate. Cloudflare's average incident over the window ran 1426 min. Zapier, flagged elevated at 23.8%, averaged 56 min per incident across 3 incidents on 99.87% uptime. Shippo, also elevated at 15.0%, had 14 major or critical incidents, more than any other service in the window, but an average incident length of 144 min.
So a high-frequency service like Shippo is a drip, and an infrastructure service like Cloudflare is a flood. You plan for them differently. The Shippo reliability page shows the pattern of short repeats clearly.
Elevated, not high: the second tier
Below the high group sit the elevated ones. AfterShip is at 28.3%, Chargebee and Deliverr at 25.0%, Klaviyo at 23.0%, Make (Integromat) at 22.1%, Triple Whale at 20.0%, TikTok Ads at 18.6%, and PayPal, Loox, Tidio, WooCommerce, Shippo, Wix, eBay and Adyen all at 15.0%. Klaviyo's elevated reading lines up with a real event: a 66 min incident on October 8, 2026 that started at 15:02 UTC, its only major incident in a window that still finished at 99.95% uptime.
PayPal and Adyen are worth a note because both recorded 100.0% uptime with 0 incidents over 90 days and still show elevated at 15.0%. Stripe does not appear on the risk list at all, and it also finished the window at 100.0% uptime with 0 major or critical incidents.
What should you actually do with a risk reading?
A percentage you look at once is worth nothing. Here is how to turn it into decisions, with the specific services named:
- Hold deploys when your platform is flagged. Shopify at 50.1% and BigCommerce at 63.7% are not days to push a checkout change, a new app install, or a theme update. If something breaks you will not know whether it was you or them.
- Put SMS on the high group, Slack or Teams on the elevated group. In StatusBird, set the alert channel per service. SMS for Cloudflare, Shopify, BigCommerce and Square. Slack or Teams for Klaviyo, Shippo, PayPal and the rest of the elevated tier, so the phone only buzzes for the ones that stop orders.
- Pre-write the one message you will need. For a Shippo or ShipStation event, the message is about label printing and shipping delays, not checkout. ShipStation's single incident in the window ran 254 min on September 25, 2026, long enough that a day of labels was affected.
- Check the risk board before a promo send. Klaviyo at 23.0% is not a reason to cancel a campaign. It is a reason to confirm the send went out rather than assuming it did.
- Reconcile against a flagged dependency, not just your store. If Recharge is at 53.6% and subscription renewals look thin tomorrow, check the vendor before you start digging in your own app settings.
How often does the picture change?
Often enough that a monthly glance is not enough. Across the last 30 days StatusBird recorded 19 major or critical incidents. By month, September 2026 had 11 major or critical incidents and 12.6 hours of combined downtime across 7 services, August 2026 had 16 incidents and 40.4 hours across 9 services, and July 2026 had 14 incidents and 22.1 hours across 8 services. The month-to-month totals move by a factor of three, and the service that dominates moves too: ShipStation in September, Plaid in August, Shippo in July. The outage index tracks that month by month.
If you want these risk readings and the incidents behind them to reach you without checking a dashboard, StatusBird can send them by SMS, email, Slack, Teams or Discord. You can create an account and pick the services that match your stack.
FAQ
Which service has the highest modelled 24-hour risk right now?
Cloudflare, at a 90.0% chance of a non-operational reading in the next 24 hours. It also has the weakest 90-day record in the set: 95.59% uptime and 5706 min of downtime, including a single incident that ran 5556 minutes from October 5 to October 9, 2026.
Does 100.0% uptime mean a service is safe to ignore?
No. Square recorded 100.0% uptime with 0 major or critical incidents over the 90 days and is still flagged high risk at 37.4%. Judge.me at 46.7%, Anthropic at 35.0% and Gorgias at 35.0% are in the same position.
If a high-risk service does go down, how long will it be out?
It depends entirely on the service. Cloudflare's average incident over the window was 1426 min, Shippo's was 144 min across 14 incidents, and Zapier's was 56 min across 3. Likelihood and duration are separate numbers and should be read separately.
Is Shopify risky to build a store on at a 50.1% reading?
The risk figure is about the next 24 hours, not the next year. Over the 90 days Shopify recorded 99.9% uptime with 2 major or critical incidents and 124 min of downtime, the longest being 106 minutes on August 19, 2026. A high short-term reading is a reason to delay a deploy, not to replatform.