B2B web design pricing in the UK.

The price bands, honestly

Under £3,000 buys template customisation: your logo and copy in a pre-built theme, fine for validating a startup, structurally unable to encode your buyer's journey. £4,500 to £25,000 is the studio band: real discovery, custom design, proper engineering, where most mid-market B2B businesses should buy. £25,000 to £80,000 and beyond is the larger-agency band, where you pay for team size, workshops and account management as much as output.

Web Hero prices in the studio band deliberately: websites from £4,500, software from £18,000, with every scope itemised so you can see what each pound buys. The senior-only model means none of it funds juniors learning or account managers relaying.

It is worth being clear that the bands describe what you are buying, not who is better. A £2,500 template site is the correct purchase for a business testing whether a market exists, and paying £20,000 to find that out would be a mistake. The failure is not buying cheaply; it is buying in the wrong band for the job — a template for a business whose website is its main sales channel, or a large-agency engagement for a company that needs eight pages and a contact form.

Why is there such a wide range at every band?

Because "a website" describes an outcome, not a quantity of work, and two projects with the same page count can differ by a factor of five in effort. The spread within a band is mostly content structure, integrations and how much strategy sits in front of the design.

The other reason is that agencies price their own overheads. A studio with senior people and low overhead can quote a number that a larger agency structurally cannot, because the larger agency is funding account management, new business teams and office space. Neither is dishonest. It does mean that comparing a small studio quote to a large agency quote is partly comparing business models rather than projects.

What can I actually get for under £3,000?

A well-chosen template, configured properly, with your own copy and photography, on a platform you can edit yourself. That is a genuinely useful thing and considerably better than no website. What it cannot do is encode anything specific about how your buyers decide, because the structure was designed before anyone knew who you sell to.

The trap at this level is not the build cost, it is the false economy of buying twice. If your site is a primary sales channel, a £2,500 build that gets replaced in fourteen months costs more than the £6,000 one that did not need replacing — and the replacement usually arrives at the least convenient time.

What moves a project up a band

Four things, in order of impact: the size and structure of the content (a 10-page site versus a 200-SKU specification catalogue), integrations and software features (portals, quoting tools, CRM and ERP connections), the depth of strategy and research, and brand work bundled alongside. Page count alone is a weak predictor; data structure is a strong one.

A useful way to think about it: designed templates cost money, pages are close to free. Twelve pages that use four templates is a small project. Twelve pages that each need their own layout is three times the design work for the same word count. When a quote seems high, ask how many distinct templates are being designed rather than how many pages are being delivered — that is the number driving it.

• Content scale and structure (collections, products, multilingual)
• Software features: logins, portals, calculators, integrations (from £18,000 with us)
• Research depth: buyer interviews, search and AI-answer mapping
• Brand and messaging work alongside the build
• Copywriting, if the studio is producing it rather than you

Does a multilingual site cost double?

No, but it costs meaningfully more than people expect, and almost none of it is design. The build work is a manageable increment: URL structure, language switching, hreflang, and a CMS that models translations properly rather than duplicating pages.

The real cost is ongoing and organisational. Every future content change now happens in every language, and either you have translators in the loop or the secondary languages quietly rot. Multilingual is a commitment to a process, not a feature you buy once, which is why it is often right to launch in one language and add the second when the demand is demonstrated.

Should copywriting be part of the quote?

It should be explicitly in or explicitly out, and it very often is neither, which is how projects stall. Assume that if the proposal does not say who writes the words, nobody has planned to, and the answer will turn out to be you, at the point when the design is waiting on it.

Studio-produced copy typically adds a meaningful line to the quote and removes the most common cause of delay. Client-produced copy is cheaper and works well when a named person has the time booked. The failure case is client-produced copy assigned to someone with a full-time job and no deadline.

Monthly costs nobody puts on the pricing page

Hosting on modern infrastructure runs £0 to £50 a month for most B2B sites. Care and updates vary by stack: plugin-heavy platforms need constant patching, modern builds need little. Growth is the real ongoing line: credible UK retainers run £1,000 to £5,000 a month for SEO (ours starts at £1,000) and £850 upwards for AI visibility. Budget the engine, not just the car.

The costs that surprise people are the small recurring licences rather than the hosting: a CMS plan, a form or booking tool, a search provider, an analytics tier above the free one, transactional email. Individually trivial, collectively a few hundred pounds a month, and rarely itemised in a build quote. Ask for the running-cost list before you sign, not after.

Then there is the cost of no ongoing budget at all, which is the most expensive option and never appears on a spreadsheet. A site with no owner and no monthly allowance decays for two years and is then declared tired, at which point the business pays for a full rebuild it would not have needed. A modest retained budget is not an upsell; it is what stops you buying the same thing twice.

Do I need an SEO retainer straight after launch?

Not necessarily a full one, but you do need someone watching the first ninety days. That period has the highest concentration of cheap wins available in the site's life: fixing what the redirect map missed, responding to the search queries that now land on you, and correcting the two or three structural things that are obviously wrong once real people are using it.

Whether that is a £1,000 a month retainer or a few days of paid attention depends on how much of your pipeline the site produces. What does not work is launching and looking again in a year.

Keeping quotes comparable

Ask every bidder the same four questions: what research is included, who exactly does the work, what does the scope itemise, and what happens in the 90 days after launch. Answers expose whether a £6,000 and a £16,000 quote are actually for the same project. Our brief-writing guide in this series gives the full question set.

Check what happens when the scope changes, because it will. A proposal with a stated change process — how variations are priced, who approves them, what the hourly or day rate is — is more honest than one that implies nothing will ever change. Fixed-price quotes with no change mechanism are not protection; they are an argument scheduled for month two.

Watch for the two numbers that are usually missing entirely: the running costs listed above, and the cost of the thing the proposal defers. "Phase two" is doing a lot of work in some quotes, and a project that looks cheaper because half of it has been moved out of scope is not cheaper, it is later.

If you want an evidence-based starting point, the £1,500 Action Plan audits your current site and prices a prioritised roadmap, which you keep whether or not you hire us to deliver it.