A web design consultant who will tell you not to redesign, when that is the right answer

Three situations bring people to a website consultant. They need very different things, too.

You are about to commission a build and the quotes make no sense. One agency says 8,000, another says 60,000, and the briefs read as if they describe entirely different projects. You cannot tell who is right. Because the thing being priced was never properly specified. Simple as that.

Or you are mid-build and something feels off. Timelines slipping, the design looking generic, questions getting vague non-committal answers. You need someone who can read the actual work and tell you honestly whether this is normal turbulence or a project heading for the rocks.

Or the site is live and underperforming. Traffic flat, conversion poor, everyone with an opinion about what to change. Redesign gets floated as the answer (because redesign is always the answer nobody can argue with).

In that third case, the honest read is often that a redesign will not fix it. That is worth paying to hear too. Genuinely.

I have been building for the web for over twelve years. I co-founded Seahawk Media, where we have shipped more than 12,000 sites, and I am still an operator rather than someone who stopped building and started selling. I run HostList, a directory of about 25,000 hosting companies. I run Deluxe Astrology, which serves 91,000-plus pages in 30 languages. The site you are reading is built on Astro and Supabase and publishes in eight languages from an admin I wrote myself.

That matters for one reason. When I tell you a stack decision will cost you later, it is because I am running that decision in production somewhere (not because I read about it in a blog post).

The most useful thing I can do on this page is be straight about when I am not the answer.

An agency is right when you need a team to execute a defined scope, you have the budget for a full delivery engagement, and you want one accountable supplier from design through launch. Agencies are genuinely good at this. I co-founded one.

In-house is right when the website is a continuously evolving product rather than a project, and there is enough ongoing work to keep someone busy every week. If you are shipping changes weekly, a contractor relationship becomes friction. It just does.

An independent consultant fits the gap between those two. You need senior judgement but not sixty days of it. You need someone with no delivery work to protect, so the advice can be "keep what you have" without anyone losing revenue by saying it. You need translation between a technical team and a commercial one. Or a second opinion on a quote before you sign (which is one of the most common reasons people come to me).

The structural difference is incentive. An agency advising you on whether to rebuild is being asked whether it would like a large project. Most are honest anyway, but you are asking a question that works against their interest. I have no build capacity to fill, which is precisely why I can say no without it costing me anything.

So, honestly: which of those three sounds like your situation right now?

Advisory calls and sessions, 200-400 GBP per hour

For a specific decision: reviewing a quote, choosing between two agencies, sanity-checking a technical proposal, deciding whether a platform migration is worth it. Usually a single session, sometimes a handful.

Design and technical audit with a written brief, 5,000-15,000 GBP

Two to four weeks. I review the existing site or the proposed direction and produce a document you can hand to any agency: what is wrong and how much each thing matters, what to build, the stack recommendation with reasons, a realistic budget band, and a scope you can put out to tender. Clients typically use this to run a procurement process, and it usually pays for itself in the quotes it corrects. Sometimes quite dramatically.

Embedded advisory, 3,000-10,000 GBP per month

Ongoing senior input alongside your team or your agency. Design review, architecture calls, hiring input, holding the standard when nobody internally has the seniority to push back. Runs anywhere from three months to years.

If the answer turns out to be a full build, I can bring that through Seahawk or point you at the right agency. But the advisory relationship is priced and delivered separately, so the advice is never a sales route.

Design is where most people start, and it is genuinely important. But it is one of five things that decide whether a site works.

Positioning first. Most sites that "look dated" are actually saying nothing specific. A visual refresh on unclear positioning produces an attractive site that still does not convert. That is the most common expensive mistake in this field. Full stop.

Then architecture. Whether the structure matches how buyers decide, rather than how the company is organised internally. Org-chart navigation is the single most common structural fault I see.

Then the technical foundation. Core Web Vitals, rendering strategy, whether the CMS suits the people who edit it. A site that editors find painful stops being updated, and a stale site fails regardless of how it looks.

Then search and AI visibility. Whether the thing is findable, and increasingly whether it is citable by AI Overviews, ChatGPT, and Perplexity. That is now a materially different discipline from classic SEO. Worth treating separately.

Design last, on top of those four. Not because it matters least, but because design decisions made before the other four are settled are essentially guesses.

HostList runs about 25,000 hosting company pages on Next.js and Supabase. The interesting problem there was not design. It was quality gating: publishing at that scale without producing the thin content that gets programmatic sites filtered out of the index. Genuinely hard to get right.

Deluxe Astrology runs 91,000-plus pages across 30 languages. The hard part was translation quality at scale and hreflang that stays bidirectionally correct across every locale, which is where most multilingual sites quietly break.

This site is the smallest and the most useful as a reference, because you can inspect it yourself. Around 3,100 URLs in eight languages on Astro and Supabase, published from a custom admin, with a build-time SEO linter that fails the deploy if metadata or structured data regresses (which is the sort of thing that sounds boring until your rankings drop). More at selected work.

Four cases. Honestly.

Your budget for the whole website is under about 5,000. Spend it on the build, not on advice about the build. A good template, executed carefully, is the right answer at that level and I will say so without any awkwardness.

You have already decided what you want and need someone to execute it. That is a delivery brief, not a consulting one. You want an agency or a freelancer, and you will pay less for exactly what you asked for.

You want a consultant to validate a decision that has already been made politically. I will give you my actual read, which is a poor outcome if what you needed was agreement.

You need a full-time person. If the work genuinely fills a week, every week, hire someone. I will help you write the job description if that is useful.

Three things make thirty minutes genuinely useful. Your site, or the brief if there is not a site yet. The decision you are stuck on, stated plainly, whether that is which agency, which platform, or whether to start at all. And your realistic budget range. Because advice that ignores budget is entertainment rather than counsel.

You will leave with a straight answer about what the problem actually is. If that answer is that you do not need a consultant, I will say so on the call. Not after an invoice.

• Web consultant London, the same advisory, geography-specific →
• Website redesign without losing what you have →
• Bespoke web design in London, when the answer is build →
• Choosing a London web design agency →
• Selected production work →