Custom or template: where the real difference sits
A site built on a theme and a page builder ships fast and costs little. The trade-off usually appears later, in maintenance and flexibility: code you did not write and have to understand before changing it, and a chain of plugins to keep current. If your need fits inside that, it is a good choice, and I will tell you so.
Custom isn't justified for everyone: if three static pages are enough, tell me and I'll say so too. It becomes justified as soon as the site carries something at stake: a conversion funnel to optimise, business data to handle, search visibility to win in a competitive market, or a requirement that will keep evolving for years.
Speed and search, handled during the build
A slow site loses visitors before it is read, and Google knows it: Core Web Vitals are a ranking signal. These constraints are far cheaper to respect while building than to retrofit afterwards: which is why they are part of the work, not an optimisation package sold separately.
Concretely: server-side rendering so the content is readable by a crawler without executing JavaScript, images converted and sized automatically, self-hosted fonts, schema.org structured data, consistent metadata and breadcrumbs, and an automatically generated sitemap.
What a web project actually covers
Beyond the pages themselves, a serious web project includes the parts that decide whether the site stays alive: a content structure your team can actually update, analytics that answer a real question rather than filling a dashboard, and legal compliance, consent, privacy policy, terms, handled before launch rather than after a complaint.
Hosting stays in your name, on your account. It's a detail until the day you want to change provider, and then it's the only thing that matters.
Redesign: the case that needs the most care
A poorly prepared redesign can cost a significant share of organic traffic; a well-prepared one can improve it. The difference isn't the budget, it's three precautions taken before launch: inventory every existing URL, redirect each of them to its real equivalent, and record current search positions so what moves afterwards can be measured rather than guessed.
When another solution will fit better
- Projects without a decision-maker. If every page needs three approvals, the announced timelines won't hold.
- Taking over undocumented code without an audit phase. On legacy work, no timeline commitment before a prior audit.
- Design without development, and search work without development. I build what I design.
- Content writing. I build the structure that makes content findable; I don't write your copy.
