A web agency does not build a site — it builds a surface that has to be found, understood and convert.

A web agency works across four layers: information structure (which pages exist and how they connect), interface (how they look and behave), engineering (how they are built, hosted and loaded) and content (what the pages actually say). Some agencies deliver design only and hand over a mockup; others build and maintain. That difference decides who carries responsibility when something breaks after launch. Advertising, campaign planning and media buying normally sit outside the remit.

One: who owns the code, content and domain when the engagement ends? Two: how are existing URLs handled — is there a redirect plan before old pages are switched off? Three: what mobile load time does the agency commit to, measured how? Four: can the editorial team publish without ordering development time? Five: what does maintenance after launch cover — security updates, follow-up, bugs? Those answers differ between agencies far more than the fee does.

When a site is replaced the addresses usually change. Without a complete map of old URLs to new ones, permanent redirects and a check of titles, descriptions and canonical tags, the site loses the history search engines built up. A web project should therefore open with an inventory of which pages actually receive traffic today, and close with a check that every one of them returns 200 or points to the right place.

One: load time and Core Web Vitals on mobile, measured on real visits rather than lab tests alone. Two: organic entrances per month compared with the same period before launch. Three: the share of visits that end in a form reply or a booked call. A web agency that cannot show those three after three months delivered a design, not a working website.

We go through structure, load time, which pages actually pull traffic and what is at risk in a rebuild. After that it is clear whether you need a new site or a rebuild of parts.

Book a site review →See the SEO work →