Baseline
Real devices, real network, real data — recorded before anything changes.
Performance Optimisation
We start with traces, not guesses. Bundle graphs, query plans, cache boundaries, render paths. Most products carry two to three times the JavaScript they need.
- Typical duration
- 2 – 5 weeks
- Engagement
- Fixed scope
- Deliverables
- 5
- Phases
- 4
Signals
Signs you needthis work.
Performance work goes wrong when it starts with a list of best practices. It goes right when it starts with a profile. We measure the real thing on real devices, find the two or three decisions actually costing the time, and change those. Usually it is a dependency nobody remembers adding, an N+1 query behind a convenient ORM call, and a render that runs four times.
Lighthouse is green in the lab and the field data disagrees.
The app is fine on your laptop and unusable on a mid-range Android.
Nobody knows why the bundle grew 40% last quarter.
Sequence
How the work runs.
Profile
Traces and query plans, until the cost is attributable rather than suspected.
Reduce
The smallest set of changes that moves the number, shipped independently.
Hold
Budgets in CI so the next quarter cannot quietly undo it.
Included
Lighthouse and field-data baseline
Bundle and dependency audit
Database and query tuning
Caching and edge strategy
Regression budgets enforced in CI
Typical stack
Chrome DevTools
Lighthouse CI
Web Vitals
Bundle analysis
Query profiling
Questions
None before we have measured. After the baseline we will give you a specific target and the changes required to reach it.
Start here
Have a projectthat needs this?
Tell us what you are trying to achieve and we will come back with an honest view on scope, sequence and whether performance optimisation is really the thing you need first.