How to choose an SEO agency in the UK
The things that actually predict whether SEO works for you sit under the logo and the monthly fee: who does the work, how senior they are, and how honestly they measure it.
Start with the question most buyers skip: who actually does the work?
The person who impresses you in the sales meeting is very often not the person who will touch your account. In a lot of agencies the model is built around that gap. Senior people win the work, then it is handed down to junior executives who are learning the craft on your site, supervised loosely and reporting up through an account manager whose job is to keep you calm rather than to do the SEO. None of that is visible in a proposal, which is why it is the first thing worth asking about directly. So ask it plainly. Who will be doing the technical work, the content strategy and the analysis, week to week? How senior are they, and will I speak to them or to a layer in between? A good answer is specific and names people. A weak answer talks about 'the team' and 'our process' without ever telling you whose hands are on the account. At Rogue Logic there are four of us and no juniors, so the person who assesses your site is the person who runs it and the person who takes the call when something is not working. That is a deliberate structure, not a boast, and you should expect any agency to be equally clear about theirs.
Why seniority beats headcount
SEO is often sold on scale: bigger team, more content, more links, more activity. But most of the work that moves the needle is judgement, not volume. Reading server logs, untangling a render path, scoping a redirect map for a migration, deciding which fixes are worth doing and which are not, working out which of a hundred possible pages actually deserves to exist. These are decisions, and the cost of getting them wrong on a live site is high. A large team running a checklist on autopilot will happily generate activity while missing the two or three changes that would have mattered. This is why a small, senior team can outperform a much larger one. When roughly forty years of combined experience sits behind the priority order, you get told what is not worth fixing as readily as what is, and you get a plan built around your site's real constraints rather than a template. When you are comparing agencies, try to separate the size of the team from the seniority of the people who will do your work. Ask how many years the actual practitioners have, and whether the person setting priorities is the same one executing them. A confident agency will show you its working rather than ask you to trust the brand.
How to read the measurement they promise
SEO reporting is where good intentions and quiet dishonesty are hardest to tell apart, because almost any agency can produce a chart that goes up. The trick is knowing which numbers are real. Rankings for a handful of easy, low-demand terms can improve while the traffic and enquiries that pay your bills do not move at all. Ask which keywords they will be accountable for, what search demand actually exists behind them, and how they connect a ranking change to clicks, and clicks to enquiries. If the answer stops at position tracking, that is a flag. Attribution deserves the same scrutiny. Ask how they will separate the effect of their work from everything else happening in your market, and how they treat branded search, which often flatters a report without the agency having earned it. Honest measurement starts from your own Search Console, analytics and, where relevant, log data, rather than a third-party tool's estimates. It reports on the pages and queries that matter to your business, and it is candid when a movement is noise rather than progress. Before you sign, ask to see a real, anonymised report from a live client so you can judge whether it would actually help you make decisions.
What a serious engagement looks like in the first months
Good SEO is front-loaded with understanding, not activity. Early on you should expect diagnosis: a proper look at what demand exists in your market, where you sit against it now, what is technically holding the site back, and where the genuine content and authority gaps are. That work should produce a clear account of where you can realistically compete today and where you cannot yet, with the reasoning shown rather than asserted. If an agency wants to start publishing content or building links in week one without that groundwork, they are selling motion, not outcomes. This is the thinking behind our visibility model: before the work starts, we publish an account of where demand exists, what each part of it is worth to you, and the specific conditions that would have to hold for you to own it. You do not have to adopt that exact approach, but you should expect something equivalent from anyone you hire. The point is that you can see the plan, understand why each recommendation earns its place, and hold the agency to a shape of work you agreed to, rather than discovering the strategy retrospectively in a monthly deck.
The questions worth asking in the meeting
A few direct questions will tell you more than any proposal. Who specifically will do the technical work and the content, and how senior are they? Can I see the actual reporting a comparable client receives? Which keywords and pages will you be accountable for, and how do you tie those to enquiries? What would you tell me not to bother fixing, and why? What does leaving look like, and what do I keep? The last one matters more than it sounds, and we come to it below. Pay as much attention to the shape of the answers as to their content. You are listening for specificity and for candour, including the willingness to say a term is out of reach for now or that a fashionable tactic will not help you. Vague, uniformly positive answers usually mean either that the person in front of you does not do the work, or that the work does not bear close inspection. An experienced practitioner will happily go a level deeper than you expected, because the detail is where their advantage actually lives.
Read the contract for how you leave
Everyone reads a contract for what they are buying. Fewer people read it for how they get out, which is where the real risk sits. Long minimum terms, notice periods measured in months, and deliverables you never take ownership of are all ways of making it expensive to leave an engagement that is not working. The healthier arrangement is a rolling relationship you stay in because it earns its place, with a plan, content and technical changes that are yours to keep whichever way the relationship goes. So ask what happens on the day you give notice. Do you keep the strategy, the content and the analytics setup, or does it walk out with the agency? Are there tools or tracking you would lose access to? A confident agency has no reason to lock you in, because its retention comes from results rather than paperwork. If the commercial terms feel designed to trap rather than to reassure, take that as information about how confident the agency really is in the work.
Match the agency to the problem you actually have
There is no single best SEO agency, only the right fit for your situation, and the situations differ more than the marketing suggests. A thin site that needs authority and depth, a technically broken site leaking crawl budget and dropping pages, a strong site being out-published by competitors in the content gap: these are three different jobs, and an agency strong at one may be ordinary at another. Before you shortlist, get honest about which of these is really your bottleneck, because it changes who you should be talking to. Whichever it is, the underlying test is the same. Are you buying the judgement of senior people who will show their working and measure honestly, or are you buying activity dressed up as strategy? If you want to see how we answer these questions ourselves, our case studies show the reasoning behind real engagements, our pricing is set out plainly, and the how-we-work page describes the model in full. Use them the way you should use any agency's materials: as evidence to interrogate, not a brochure to admire.