E-commerce peaks
How much revenue is at stake on your peak day?
Enter your normal order volume, the expected rush and where you stand on safeguards. The check estimates how much revenue an hour of downtime puts at risk, how much load has never been tested and which measures are likely to fit before the date.
- Free
- No sign-up
- about 3 minutes
- Inputs stay in your browser
What you get
- Countdown to Black Friday or your own date
- Revenue at risk per hour of downtime
- Tested vs. untested load
- Risk score from 0 to 100
- Measures sorted by lead time
Methodology
How the check works
The calculation uses only your inputs. None of it leaves your browser.
- Revenue at risk
- Orders per hour on a normal day times the expected rush times the average order value. This estimates potential sales exposure during one hour of complete downtime. It does not determine whether orders are permanently lost or completed later.
- Untested load
- Expected rush divided by the highest load your store has demonstrably handled. Anything above 1 means this part of the peak has never been experienced under real conditions.
- Risk score
- Eight safeguards weighted 8 to 15 points: load test and database 15 each, external services and alerts 12 each, caching, deployment freeze and on-call 10 each, emergency plan 8. “Partly” counts half. Up to 8 more points for untested load. 25 and above is moderate, 50 high, 75 critical.
- Lead time
- Every measure has an assumed planning lead time: load test and database scaling 21 days, external limits, caching and emergency plan 10 days, alerts 7, on-call 5 and deployment freeze 3 days. Actual timing depends on scope, access, approvals and the people available. Anything that no longer fits under this assumption is marked as too late.
- Date
- Black Friday is the Friday after the fourth Thursday in November, Cyber Monday three days later. Ideally you start four to eight weeks before the peak.
Next step
Know where the limit is before the peak
In the Peak Readiness Audit, I review user journeys, infrastructure and processes and deliver a risk matrix, a prioritised action plan and a concise runbook for the sales event.
Questions about this tool
Isn’t auto-scaling on our cloud instances enough?
Auto-scaling mainly helps stateless web servers. When the database, caches, or ERP and PIM integrations become the bottleneck under load, more compute often just means more waiting requests and a higher bill.
How early should we start preparing?
Four to eight weeks before the peak. That leaves time for a realistic load test, fixing the bottlenecks and a second measurement.
Does the check apply to Shopify or Shopware Cloud?
Partly. On SaaS platforms, the provider scales the infrastructure. Your risks then sit with apps, integrations and third parties, not server operations.
Are my numbers stored?
The calculation runs in your browser. The latest result summary and result link are temporarily stored in this tab for the booking window. If you share them by e-mail, you decide whether to send the message.
