Schedule a call

Services

Web Development

A website is infrastructure. It has uptime, certificates, DNS, backups and a maintenance burden, and it goes wrong in the same ways everything else does.

What this covers

  • Website design and build
  • Migration off page-builder platforms
  • Performance and Core Web Vitals
  • Technical SEO foundations
  • Hosting, DNS, and TLS
  • Ongoing maintenance and updates

We did this to our own site first

This site used to run on Webflow. We moved it onto a static build we control, and the exercise turned up the sort of thing that accumulates quietly on any site nobody is auditing.

There was no sitemap and the robots.txt file was empty. Not one page carried a canonical tag. Five subscribe forms had no destination at all and had been silently discarding anyone who used them. Two links in the navigation led to pages that did not exist. The about page's main heading said "Get in touch".

None of that was visible from the front, and all of it was costing something. It is the honest argument for the service: the problems worth fixing on most sites are not the ones anyone looks at.

What you get

The same build this site runs on, which is the most useful thing we can tell you about it — you are looking at the deliverable rather than a portfolio screenshot.

A static site. Pages are built once and served as files, so there is no database to breach, no plugin to patch, no admin login to brute-force, and nothing to go down under load. It is the fastest thing a website can be and the least there is to attack.

A quality gate in the build itself. Twenty-nine automated checks run on every deployment and fail it rather than warn: a missing canonical tag, a duplicate H1, an image without dimensions, a form with no destination, a heading level skipped, an internal link that goes nowhere. The rules cannot quietly stop being followed, because nothing ships when they are not.

Hosting on a global edge network, with certificates and renewals handled, and a deployment history you can roll back to.

Forms that deliver, on an account in your name, with submissions stored so a mail problem cannot lose an enquiry.

The source, in your repository. Every change is a commit with an author and a date, and any of them can be reverted.

The platform decision, made honestly

Page builders are genuinely good at what they are for. They let someone non-technical change copy without waiting for anyone, and for a business where that matters more than anything else, that is the right trade and you should keep it.

What you give up is control of the output. The markup is generated, the performance ceiling is set by the platform, and the parts that matter for search sit behind whatever settings the builder chose to expose. You also cannot leave easily, because the design and the hosting are the same product.

A static build inverts that. The output is yours, it is fast because there is very little of it, and moving hosts is a deployment rather than a rebuild. The cost is that changing a page is a developer task, not a drag-and-drop one. Neither is the right answer in general. It depends entirely on who needs to change the site and how often, and that is worth settling before anyone builds anything.

Fast is a feature, and it is mostly about restraint

Speed is a ranking factor, and well before that it is the reason people leave. Most sites are slow for unremarkable reasons rather than exotic ones.

Images shipped at several times the size they display at. Fonts loaded from three sources when one would do. Analytics, chat widgets, heat mapping and a tag manager, each added for a reason nobody remembers, all executing before anything renders. Whole frameworks downloaded to run a menu.

The work is largely deletion. It is also the least glamorous part of the job and the part with the most reliable payoff.

Findable is a build decision, not a plugin

A good deal of what is sold as search optimisation is repair work on decisions made during the build. Canonical URLs, heading structure, internal links, structured data and a working sitemap are cheap to get right at the start and tedious to retrofit across a site that already exists.

That is the argument for the build gate above. It is not a clever trick — it is the ordinary observation that a rule a person has to remember is a rule that eventually gets forgotten, and that the forgetting is invisible until an audit years later turns up a site with no canonical tags on any page.

We know how that looks from the inside, because it is what we found here.

You should own all of it

The same principle we apply to managed IT applies here, and it matters more than it sounds: your domain, your DNS, your hosting account and the source of your site belong in your name.

This is the most common way businesses get stuck. The site was built by someone who registered the domain on your behalf, hosts it in their account, and holds the only copy of the source. Nothing is wrong until you want to change supplier, at which point you find that leaving was made expensive by an arrangement nobody deliberately chose.

It is worth checking who holds those four things for your current site. The answer should take about a minute to establish, and if it does not, that is the answer.

What this is not

Worth saying plainly, because a mismatch found after signing is expensive for both of us.

It is not a site your team edits themselves. A static build has no admin login, and that is deliberate — it is most of where the speed and the security come from. Changes come to us and we make them, which is part of what the maintenance covers rather than an extra. For a site that changes a few times a month, that is a better arrangement than it sounds: nothing breaks because someone pasted formatting from Word, and the site cannot drift out of the shape it was built in.

If you need to publish several times a week without waiting for anyone, we are the wrong fit and we would rather say so now. That is a content-management product and there are good ones.

It is not e-commerce. Taking payments brings card handling, inventory and a different compliance surface. It is a separate conversation rather than a page count.

It is not a web application. Customer logins, accounts, dashboards and anything with a database behind it is software, with the ongoing burden that implies. A brochure site and an application are different products that happen to arrive through a browser.

It is not brand design from scratch. We will build something clean and coherent, and work within the brand you have. Creating a visual identity, original photography and illustration are a designer's job and we would rather tell you that than approximate it.

Launch is the start of the maintenance

Sites decay. Certificates expire, dependencies pick up vulnerabilities, forms stop delivering, a supplier changes an endpoint, links rot as other people reorganise their sites.

The failure that costs most is the silent one. A contact form that stops delivering looks exactly like a quiet month, and nobody investigates a quiet month. We know because we found precisely that on our own site — an inherited form endpoint rejecting every submission, with no account behind it and nobody watching.

So the maintenance question is worth asking before the build question: who is responsible for this in a year, and how would anyone find out it had broken? For an existing managed IT client this is straightforward, because it is the same job as everything else we look after.

Common questions

Do you only do this for managed IT clients?

No — it is available on its own, and hosting and ongoing maintenance are included rather than sold back to you afterwards. It does fit naturally alongside managed IT, because a website has the same uptime, certificate, DNS and maintenance needs as everything else and benefits from the same people watching it. But you do not have to buy the rest to get this.

Can we update the site ourselves?

No, and it is worth establishing that early. A static build has no admin login, which is where much of the speed and security come from. Changes come to us and we make them, as part of the maintenance rather than as an extra. For most business sites that is a better arrangement than it sounds — nothing breaks because someone pasted formatting out of Word, and the site cannot gradually drift out of the shape it was built in. If you need to publish several times a week without waiting for anyone, we are honestly the wrong fit and would tell you so rather than sell you something that will frustrate you.

Can we see an example of your work?

You are on it. This site is the deliverable — the same static build, the same build-time quality gate, the same hosting. Run it through any speed or SEO tool you like rather than taking our word for it, and compare that against whatever your current site scores.

Should we move off our current platform?

Not automatically. Page builders are good at letting non-technical staff make changes without waiting for anyone, and if that is the thing you most need, keeping it is the right call. Moving makes sense when the platform's ceiling is costing you — performance you cannot fix, markup you cannot control, or settings the builder does not expose. The honest version of this conversation starts with who needs to change the site and how often.

Will a new site improve our search rankings?

It removes the technical obstacles, which is not the same thing. Canonical tags, heading structure, a working sitemap, internal links and page speed are foundations — with them broken, good content underperforms; with them fixed, you have a fair chance rather than a guarantee. Anyone promising rankings is promising something they do not control.

Who owns the site when it is finished?

You do — the domain, the DNS, the hosting account and the source. We think that should be true of any supplier, and it is worth checking against whoever built your current site. Where those four things sit is what decides whether changing supplier later is straightforward or expensive.

What happens after it launches?

Something needs to keep watching it. Certificates expire, dependencies pick up vulnerabilities, forms quietly stop delivering, and links rot as other people reorganise their sites. The dangerous failures are the silent ones — a contact form that stops working is indistinguishable from a quiet month, and nobody investigates a quiet month.

The other things we do

  • Managed IT Services

    We take responsibility for your systems. Unlimited helpdesk, proactive maintenance, and status monitoring, so your team stops losing hours to technology that should just work.

  • Compliance & CMMC

    CMMC Level 2 readiness and Microsoft GCC High / Azure Government environments for the defense supply chain, built so evidence of every control is ready to produce.

  • Cyber Security

    End-to-end protection built on a zero-trust approach: we map normal activity and act on the outliers, rather than waiting to be told something has gone wrong.

  • Cloud Solutions

    Microsoft Azure, Azure Virtual Desktop, and Microsoft 365 — designed, migrated, and run by engineers who hold the certifications for all three.

  • Backup & Disaster Recovery

    Backups you have actually tested and a disaster recovery plan that has actually been rehearsed. It takes ten years to build a business and one bad day to lose its data.

  • AI Services

    Practical AI inside the tools your team already uses — with the same care about where your data goes that we apply to everything else.

Get in touch

Talk to us about web development

Tell us what you're dealing with and we'll respond as soon as possible.

We don't share your data. View privacy policy.