Web Platform Engineering & Core Web Vitals
Public-facing platforms built and operated to the same standard as the rest of your infrastructure: performance, security, and machine-readability as engineering properties rather than afterthoughts.
Stack & Delivery
What we build on, and how it runs.
Technology stack
How an engagement runs
- 1
Baseline
Measure before changing anything. Field and lab metrics, crawl state, and current bundle weight, recorded so improvement is provable.
- 2
Architecture
Rendering strategy, cache policy, and route structure agreed in writing, because these are expensive to reverse later.
- 3
Build
Static generation by default, budgets enforced in the build, accessibility checked rather than assumed.
- 4
Instrument
Structured data, crawl configuration, and analytics wired so the platform reports on itself.
- 5
Verify
Re-measure against the baseline. Anything that did not move is reported as not having moved.
- 6
Operate
Dependency upgrades, regression monitoring, and a named owner. Platforms decay when nobody owns them.
Our Approach
A public website is production infrastructure. It is the system most often reached by people outside the business, it is indexed and quoted by machines you do not control, and it is usually the least monitored thing a company runs. We build and operate it with the same discipline applied to a server estate: measured baselines, budgets that are enforced rather than aspired to, security headers and dependency management, and instrumentation that proves what the platform is doing rather than assuming it.
What's Included
Enterprise Application Development
Public platforms and client-facing applications built on Next.js and React, statically generated where content allows, with a design system and accessibility treated as build requirements rather than review comments.
Performance Engineering
Core Web Vitals worked as an engineering problem: bundle budgets enforced in the build, render strategy chosen per route, cache and CDN policy set deliberately, and an image pipeline that reserves layout space so nothing shifts.
Platform Security Hardening
Content Security Policy, transport security, and referrer policy configured rather than defaulted, with dependency management on a schedule. The public platform is the most exposed system most businesses run.
Structured Data & Crawl Configuration
Schema markup generated from page content so the two cannot diverge, plus sitemaps, robots directives and canonical strategy — including paginated and filtered routes, where incorrect canonicals silently remove pages from search.
Search & Analytics Instrumentation
Google Search Console, Bing Webmaster Tools, analytics and conversion events configured as observability: evidence of what the platform is actually doing, reported next to performance and availability rather than as a separate marketing exercise.
Platform Maintenance
Dependency upgrades, framework version migrations, performance regression monitoring and uptime checks, with a named owner. Most platform failures are deferred maintenance rather than incidents.
Capability Matrix
What the service covers.
Who This Is For
Built for operations that live on the phone.
Businesses whose site is a sales channel
Where a slow or unreachable platform costs enquiries directly, and nobody currently owns its performance.
Regulated operators
Where the public platform must meet the same security and accessibility expectations as internal systems.
Teams with a platform nobody maintains
Built once by an agency, never updated, dependencies years behind, and no baseline to measure against.
Related Services
What this is usually deployed alongside.
Guides
Written by the engineers who do this work.
IT Support
nginx 502 or 504 on a Node.js or Laravel app: find the failing boundary
A gateway error is a symptom. Separate a dead upstream, a slow handler, an exhausted worker pool and a timeout mismatch before increasing limits.
Read guideCustom Software & CRM
Choosing a technology stack for business software
Stack debates are conducted on technical merit and decided by taste. For a business commissioning software, the properties that matter are how long the platform is supported, who can be hired to work on it, and what happens when the person who built it leaves.
Read guideCustom Software & CRM
Building business applications that scale
Most business applications do not fail under load. They fail under data — a query that was instant against ten thousand rows and unusable against two million. Scaling is mostly a database problem, and mostly solvable without any architectural change.
Read guideCloud & Infrastructure
AWS vs Azure for a growing business
Both are excellent and the comparison as usually written is unusable — hundreds of services against hundreds of services. For a business with a handful of workloads, three factors decide it, and none of them appears in a feature matrix.
Read guideCustom Software & CRM
How a custom CRM actually gets built
Custom software projects rarely fail on code. They fail on scope that was never bounded, data that was worse than anyone admitted, and a launch that assumed people would change how they work because they were told to.
Read guideTalk Through Your Requirements
We typically respond within 4–8 business hours.