Websites your team can still maintain in year three.
Corporate sites, storefronts, landing pages and customer portals built on mainstream stacks, with a performance budget enforced in the pipeline and a handover your team can act on.
Overview
The maintenance problem this addresses
Most sites we are asked to rebuild are not badly designed. They are unmaintainable. Publishing rules sit buried in theme templates, nobody was trained on the CMS, and the original vendor has moved off the stack. The rebuild is the symptom rather than the requirement.
So we optimise for what happens after launch: mainstream frameworks with an active hiring market, a content model your marketing team can extend without us, and a performance budget that fails the build rather than the quarter.
Scope of work
What the service covers
Multi-language and multi-region marketing sites, with a content model your team can extend without a developer.
Headless or platform-native storefronts, with search, catalogue and checkout tested at the volume you expect in two years.
A reusable page system for campaigns, so marketing publishes in a day rather than raising a ticket.
Authenticated customer, dealer or partner portals with role-based access and audit trails.
Outcomes
What changes after handover
Your marketing team edits pages, publishes campaigns and adds sections. We are not a dependency for routine work.
A budget enforced in the pipeline means a regression fails the build rather than being noticed a quarter later.
The marketing practice sits in the same team, so conversion paths are designed in rather than retrofitted.
Standard frameworks, your repositories, your hosting account. Your next vendor will not need a rebuild to make a change.
What we enforce in your pipeline
These are automated gates rather than intentions recorded in a document. A change that breaks one of them fails the build and does not reach production.
A speed limit agreed with you in discovery and checked automatically on every change
A performance target quoted in the proposal and measured once, on a desktop connection, before launch
Image handling in the publishing pipeline, so an oversized upload is converted and compressed rather than served as it is
Whoever updates the site next unknowingly adding two megabytes to the homepage
An agreed limit on marketing and analytics tags, with additions reviewed before they go live
Scripts accumulating until the site is slower than the one it replaced
Accessibility checks on every change, plus a manual audit at handover
A single accessibility review in the week before go-live
What your team holds on the final day
Pages, sections and campaign templates your marketers extend without a developer.
Reusable blocks with usage notes, so new pages are correct by default.
Benchmarked before cutover and monitored for sixty days afterwards.
One for editors, one for whoever inherits the repository.
Source code, cloud accounts and domain registrations are registered in your name throughout, as on every Candela engagement.
Delivery process
How engagements are run
Content audit, analytics review, discussions with the people who publish, and a sitemap both sides sign off.
Content model, design direction, component inventory, stack decision and the performance budget agreed in writing.
Two-week sprints with a staging URL open to your team from week two.
Load testing, accessibility and SEO checks, redirect map, then a cutover on a date you approve with rollback prepared.
Support window, an analytics review at thirty days, and iteration against real behaviour.
Tools and platforms
The delivery stack
We standardise on React with TypeScript because it is the easiest stack for you to hire into after handover, not because it is the newest option available. Where your team already runs something mainstream, we build on that instead.
We avoid: newly released frameworks with small hiring markets
See the full technology page →In practice
A storefront rebuilt to handle a growing catalogue
The store was slow on mobile, the checkout was losing customers, and search degraded as the catalogue grew. We rebuilt on a headless platform with a dedicated search layer, moved merchandising into the CMS, integrated inventory and payments, and added automated speed checks before every release.
questions
Yes. If you hold brand guidelines or a design system, we build to it. If the design is the problem, we will say so in discovery rather than after the build.
We benchmark rankings, traffic and indexed URLs before anything moves, then build a redirect map covering every live URL, including those that only matter to one paid campaign. After cutover we monitor crawl errors and rankings for sixty days and correct regressions as part of the engagement rather than as a change request. Migrations dip briefly. They should recover within a month.
Fixed cost where the scope is clear, which after discovery it usually is. A landing page system sits at the lower end, a corporate site in the middle, and a headless storefront or portal at the upper end. If discovery shows you need something smaller than you asked for, we scope it smaller.
Every build includes a support window after launch. Most clients then move to a monthly retainer for iteration, campaign pages and platform updates, but it is optional, and it has to be, because the handover is built so you can stop.
Find out whether your site needs rebuilding or unblocking
Send us the URL and tell us which CMS it runs on. We will review it and tell you in a call whether the problem justifies a rebuild or whether it can be resolved within what you already have. Available as a standalone review, with no obligation to continue.