Performance &audits
Find the bottleneck, fix it, and prove the improvement.
We profile real users rather than guessing from a lab score, then prioritise fixes by effort against measured impact. You leave with the numbers, the reasoning, and a team that knows what to watch.
What the engagement involves
A lab score tells you what could happen on a fast laptop on a good network. Field data tells you what is actually happening to your users. We start from the field data ’ real-user Core Web Vitals, the slowest routes weighted by traffic, and the waterfalls behind them ’ and work down until the remaining fixes stop being worth the effort.
Each finding arrives with a measurement, an estimated effort and a recommended fix, ordered by impact. We implement the high-value items inside the engagement rather than handing you a report, because a report that says "reduce your JavaScript" is not worth the audit fee.
Who it suits
This fits sites and applications where speed is measurably costing you conversions, rankings or support tickets, and teams who have tried the usual fixes without moving the numbers. It is the wrong engagement if you have no field data and no way to collect it ’ instrument first, audit second, or the exercise becomes guesswork delivered with confidence.
How we run it
Every engagement follows the same sequence. The steps below are in delivery order — each one is independently useful, so an engagement can stop after any of them with something finished.
01
Measure real users
Field data first: what your actual visitors experience, on their devices.
Lab scores are rehearsals. We start with real-user distributions — the 75th percentile on a mid-range Android over 4G — because that is where your revenue lives.
02
Profile, never guess
Traces and waterfalls before recommendations.
Every claim in the report links to a trace, a filmstrip or a query plan. "Probably the images" is not a finding; a 2.1-second render-blocked hero is.
03
Fix by impact per effort
A ranked list where each item carries its expected gain and cost.
The plan reads like triage, not a wishlist. Quick wins first for momentum, structural fixes sequenced so each one is independently shippable.
04
Prove it stuck
Before-and-after measurement with regression monitoring left running.
We re-measure after every fix against the same methodology and leave the monitors in place. Performance rots without a budget watching it — so we install the budget.
When performance & audits is the wrong answer
You want a perfect score regardless of cost
Chasing 100 on every page burns budget users will never feel. We optimise to the point where humans stop noticing — past that, the money belongs in features.
The bottleneck is the business model
If visitors bounce because the offer is unclear or the pricing is wrong, no amount of millisecond-shaving moves revenue. Fix the proposition first.
Nobody will act on the report
An audit nobody implements is an expensive PDF. If there is no engineering capacity lined up for remediation, wait until there is — then call us.
Questions we get about this kind of work.
01Do you implement enterprise systems or just advise on them?
02Will you tell us not to buy something?
03How do you price enterprise work?
04Can you take over a project that has already gone wrong?
Ready to talk about performance & audits?
Most enquiries get a reply within 2 working days. If it is not a fit we will say so rather than waste your time.