◆
Peak is not an event
The day everyone arrives at once is handled by arrangement, not by a team sitting up overnight hoping.
Retail is judged on a handful of hours. Everything else is preparation for them. The infrastructure question is not whether you can handle an average Tuesday — it is whether the exceptional day passes without anyone noticing it was exceptional.
Used for storefronts, checkout, catalogue, fulfilment and campaign surfaces.
What this looks like in practice
The day everyone arrives at once is handled by arrangement, not by a team sitting up overnight hoping.
A landing page goes live when marketing decided it should, not when the next release window opens.
A bad change is undone in seconds. That changes what your team is willing to try.
You pay in proportion to the trading you did, not to the peak you provisioned for and mostly did not use.
The traditional answer to peak is to provision for it permanently and absorb the waste. It works and it is expensive in a way that compounds quietly.
The reliable sign that this has worked is cultural rather than technical: nobody books the evening out. The trading peak arrives, the site is the same as it was, and the team finds out it happened by looking at the revenue rather than the alerts.
The part of the journey where failure is measured directly in money, treated accordingly.
Every promotion its own surface, live when it was meant to be and gone when it is over.
The applications your store teams use improve at the same pace as the storefront.
Large, frequently-changing catalogues that stay fast as they grow.
Our peak day used to be an all-hands event. Last year two of us checked it from our phones.
That is the day this is for. Capacity follows demand rather than being guessed at in advance, so the exceptional day does not require an exceptional effort.
Yes. Campaign surfaces go live on marketing’s schedule, with the same record and the same ability to undo as anything else.
Most teams stop needing one, because the cost of a mistake drops to seconds. Whether you keep one becomes a choice about risk appetite rather than a technical necessity.
It rises with the trading and falls afterwards, in proportion. You stop paying year-round for a capacity you need for a fortnight.
Run regulated workloads with a complete record of every change, and answer an auditor in an afternoon rather than a quarter.
Carry sensitive workloads with separation that is structural, and keep clinical and research software moving.
Serve the public on infrastructure that stands up to scrutiny, procurement and a service that cannot be allowed to fail.
Publish the moment it matters, carry an audience that arrives all at once, and keep an archive that stays reachable.
Bring plant data somewhere useful, keep the software near the line dependable, and stop treating the factory as offline.
Carry metering and grid data at full fidelity, and run customer systems that hold up when everyone calls at once.
Begin with one workload. Keep what you have running. Decide whether to continue once you have seen it carry something real.
No migration required to begin. Nothing to tear down if you stop.