The SEO Works review

Strong — 4 of 5 on this rubric’s scale, from Poor (1) to Excellent (5). This dimension carries 25% of the total score.

Benchmark — across the 16 other agencies evaluated in this discipline, this dimension runs Strong 8 · Adequate 7 · Weak 1. The typical agency here scores Strong, and none scores higher than this one.

What this dimension measures: What the firm actually does, described concretely enough to evaluate: content restructuring for extractability, entity and knowledge-graph work, schema, source and citation-footprint building, presence on the corpora assistants actually draw from, llms.txt and crawler access. Be sceptical of anything indistinguishable from ordinary SEO sold at a premium under a new name: if the described method is only 'good content and schema', say so plainly in the verdict rather than crediting the label.

Scores high — Named, specific techniques tied to how retrieval works score 4-5.

Scores low — 'We optimise for AI' with no described mechanism scores 1-2.

What we found — Five separate service pages (GEO, AEO, AI SEO, AIO, free LLM audit) describe named techniques: restructuring content into modular direct-answer blocks with the question as the heading, FAQ/Article/HowTo/Organisation/LocalBusiness schema, an llms.txt file, crawler access for AI bots, and digital PR to build brand mentions on the corpora assistants draw from (they state a mention is as useful as a link). The AIO page names the surfaces audited: ChatGPT, Gemini, Perplexity, AI Overviews. What lifts this above a relabelled SEO deck is that three AI-specific artifacts are observable on their own domain rather than merely described: /llms.txt (Yoast-generated), /llm-info.html (a canonical-facts page written for model ingestion), and /entitymap.html, a machine-readable EntityMap v1.0 JSON dated 5 June 2026 declaring 21 entities and 38+ relationships. That is real entity/knowledge-graph work, not a slogan. Held to Strong rather than Excellent because the majority of the stated method - schema, heading hierarchy, question research via AnswerThePublic and AlsoAsked, backlinks, NAP consistency, digital PR - is ordinary SEO and featured-snippet practice sold under a new label, and the AEO page's own FAQ concedes AEO and GEO 'overlap a lot and create the same outcome'. That is the high band above, which is why it scored Strong.

On the record — “They publish a machine-readable EntityMap v1.0 JSON, generated 5 June 2026, declaring 21 entities and 38+ relationships and marking its own status as self-declared - direct evidence of entity/knowledge-graph work applied to themselves.” seoworks.co.uk ↗

On the record — “AI search is a real named service line, not a single marketing page: separate pages exist for GEO, AEO, AI SEO, AI Optimisation (AIO) and a free LLM audit, each with its own described method.” seoworks.co.uk ↗

Evidence: verified — checked against a named source you can open; the links under Sources below are where to check it yourself.

Adequate — 3 of 5 on this rubric’s scale, from Poor (1) to Excellent (5). This dimension carries 20% of the total score.

Benchmark — across the 16 other agencies evaluated in this discipline, this dimension runs Excellent 1 · Strong 6 · Adequate 8 · Weak 1. The typical agency here scores Adequate, and 7 of them score higher than this one.

What this dimension measures: How the firm proposes to prove anything, in a setting where outputs are non-deterministic and vary by user, session and region. Look for a described tracking approach with its limits acknowledged: prompt sets, sampling frequency, variance handling, and how a citation is attributed.

Scores high — A stated method that names non-determinism as a limit scores 4-5.

Scores low — Vague 'we track your AI visibility' scores 2-3. A firm presenting single-prompt screenshots as measurement scores 1-2 — one answer on one day is an anecdote, and treating it as a metric is the characteristic error of this category.

What we found — They name what they count - citation frequency, answer share versus competitors, brand sentiment in AI answers, AI Overview and snippet placements, and LLM referral traffic - which is more than 'we track your AI visibility', and referral traffic is genuinely checkable in a client's own analytics (the self-case-study attributes its numbers to Google Analytics). But no prompt set is published anywhere, no sampling frequency, no sample size, no attribution rule for what counts as a citation, and no acknowledgement that model outputs are non-deterministic or vary by user, session and region. Tooling is described only as 'enterprise-grade tools and our own custom-built software' and 'in-house technology' - never named, so a buyer cannot assess it. Not scored lower because nothing on the site presents a single-prompt screenshot as a metric. That is between the two bands, which is why it scored Adequate.

On the record — “The AIO page names the surfaces audited (ChatGPT, Gemini, Perplexity, AI Overviews) and the KPIs tracked (citation frequency, brand sentiment, answer share, referral traffic), but names no tool, prompt set or sampling method.” seoworks.co.uk ↗

On the record — “Their headline '2000% increase in AI referral traffic' is their own brand's result and, in the case study itself, resolves to a move from roughly 5 to over 100 AI referrals per month over about a year.” seoworks.co.uk ↗

attribution — working out which marketing touch actually caused a sale. Good practice names its model and its blind spots; bad practice quotes each ad platform’s self-graded numbers, which overlap and overclaim.

Evidence: vendor stated — the agency’s own claim, recorded as theirs rather than ours.

Adequate — 3 of 5 on this rubric’s scale, from Poor (1) to Excellent (5). This dimension carries 20% of the total score.

Benchmark — across the 16 other agencies evaluated in this discipline, this dimension runs Strong 5 · Adequate 7 · Weak 4. The typical agency here scores Adequate, and 5 of them score higher than this one.

What this dimension measures: Weighted unusually high because this category's defining risk is overclaim, not opacity. No vendor controls what an assistant says. Also flag results claims presented without the prompt set or timeframe that produced them, and any implication that the firm has privileged access to a model's ranking.

Scores high — A firm that states plainly what it cannot guarantee scores 4-5.

Scores low — Guaranteed rankings, guaranteed citations, guaranteed 'first result in ChatGPT', or claimed partnerships with model providers score 1-2 and must be named in red_flags.

What we found — Across five service pages, the free-audit page and five case studies I found no guaranteed ranking, no guaranteed citation, no 'first result in ChatGPT', and no claimed partnership or privileged access to any model provider - which is the main thing this dimension is guarding against, and they clear it. The AI SEO page notes billions still use traditional search; the AIO page says results need patience and that AI models are constantly evolving. What keeps this off the top band is the absence of any plain statement of what they cannot guarantee, plus results published without the conditions that produced them: 'ChatGPT is now recommending your competitors' framing sits alongside '93% of our clients are already appearing in AI overviews' with no timeframe, sample size, prompt set or definition of 'appearing', and the AI case studies are percentage deltas ('+287% AI Overview citations', '2000% increase in AI referral traffic', '73% citation rate', '60.6% share of voice') with no baselines and, on four of five, no timeframe. That is between the two bands, which is why it scored Adequate.

On the record — “No guaranteed ranking, guaranteed citation, guaranteed placement in ChatGPT, or claimed partnership with a model provider appears on any of the five AI service pages or the free-audit page.” seoworks.co.uk ↗

On the record — “The AIO page names the surfaces audited (ChatGPT, Gemini, Perplexity, AI Overviews) and the KPIs tracked (citation frequency, brand sentiment, answer share, referral traffic), but names no tool, prompt set or sampling method.” seoworks.co.uk ↗

Evidence: verified — checked against a named source you can open; the links under Sources below are where to check it yourself.

Strong — 4 of 5 on this rubric’s scale, from Poor (1) to Excellent (5). This dimension carries 15% of the total score.

Benchmark — across the 16 other agencies evaluated in this discipline, this dimension runs Strong 9 · Adequate 7. The typical agency here scores Strong, and none scores higher than this one.

What this dimension measures: Attributable client work. Be lenient on volume and strict on attribution: this category is young enough that a short roster is expected, but a long roster of unnamed logos is not evidence. Where a firm's OWN brand is demonstrably visible in AI answers, note it as a real if partial signal — it is at least a claim about themselves they cannot make privately.

Scores low — Named clients with described engagements score 4-5; anonymised or absent scores 2-3.

What we found — One AI-search engagement is fully named and described: DHL, where destination content for UK-to-international shipping routes was expanded and structured for both crawlers and LLMs, reported as +121% ChatGPT referral traffic, +287% AI Overview citations and +108% AI Overview mentions quarter on quarter, and stated to have won Use of Search - B2C (SEO) (Large) at the UK Search Awards (award claim is vendor-stated; I did not read the awards site first-hand). The other four AI case studies are sector-anonymised (seating manufacturer, pool rental company, coffee brand, marketing company), which is not attribution. Per the rubric's own instruction, their own brand's AI presence counts as a partial signal and it is unusually legible here: a self-case-study documents going from roughly 5 to over 100 monthly AI referrals over about a year, and the published entitymap, llm-info and llms.txt files are the method applied to themselves in public. Named logos elsewhere on the site (Roland, Zippo, NHS, Just Eat, Tata Steel, University of Sheffield) are not attributed to AI search work and were not credited here.

On the record — “DHL is named as an AI search / GEO client, with the engagement described (structuring international destination content for crawlers and LLMs) and results given as +121% ChatGPT referral traffic, +287% AI Overview citations and +108% AI Overview mentions quarter on quarter.” seoworks.co.uk ↗

On the record — “Their headline '2000% increase in AI referral traffic' is their own brand's result and, in the case study itself, resolves to a move from roughly 5 to over 100 AI referrals per month over about a year.” seoworks.co.uk ↗

attribution — working out which marketing touch actually caused a sale. Good practice names its model and its blind spots; bad practice quotes each ad platform’s self-graded numbers, which overlap and overclaim.

Evidence: partly checkable — corroborated in part against the sources below; the remainder rests on the agency’s own account.

Strong — 4 of 5 on this rubric’s scale, from Poor (1) to Excellent (5). This dimension carries 10% of the total score.

Benchmark — across the 16 other agencies evaluated in this discipline, this dimension runs Excellent 3 · Strong 3 · Adequate 3 · Weak 7. The typical agency here scores Adequate, and 3 of them score higher than this one.

What this dimension measures: Note contract length: long lock-ins are hard to justify in a category this young, where nobody can evidence a twelve-month outcome yet, and a firm offering short or month-to-month terms should be credited for it.

Scores low — Published prices, ranges or defined packages score 4-5; ranges on request score 2-3; bespoke-only scores 1-2.

What we found — Four named packages with monthly figures are published at /seo-packages/: Springboard GBP 1,500-3,000, Growth GBP 3,000-4,500, Growth+ GBP 4,500-6,000, Enterprise GBP 6,000+. That page is also a vendor-authored UK cost guide and its general market ranges are marketing, not disclosure - but the named tiers with their own numbers are real published pricing. Clutch independently lists a $1,000+ minimum project size and a $100-149 hourly band, consistent with those tiers. Two gaps keep it off the top band: the AI search line itself carries no price at all (every GEO/AEO/AIO page routes to 'request a quote' or the free LLM audit), and no minimum term or contract length is published anywhere, so the firm earns no credit for short or month-to-month terms in a category too young to justify a long lock-in.

On the record — “Four named retainer packages are published with monthly figures: Springboard GBP 1,500-3,000, Growth GBP 3,000-4,500, Growth+ GBP 4,500-6,000, Enterprise GBP 6,000+. No AI search / GEO price is published and no contract term is stated.” seoworks.co.uk ↗

On the record — “AI search is a real named service line, not a single marketing page: separate pages exist for GEO, AEO, AI SEO, AI Optimisation (AIO) and a free LLM audit, each with its own described method.” seoworks.co.uk ↗

Evidence: verified — checked against a named source you can open; the links under Sources below are where to check it yourself.

Adequate — 3 of 5 on this rubric’s scale, from Poor (1) to Excellent (5). This dimension carries 5% of the total score.

Benchmark — across the 16 other agencies evaluated in this discipline, this dimension runs Strong 7 · Adequate 9. The typical agency here scores Adequate, and 7 of them score higher than this one.

What this dimension measures: Whether the firm acknowledges that the surfaces move, and describes how it adapts: monitoring for model and interface changes, revisiting method, published research or original testing.

Scores high — Evidence of ongoing testing and updated method scores 4-5.

Scores low — A static playbook presented as settled practice scores 2, since nothing in this category has been stable long enough to be settled.

What we found — They do acknowledge the ground moves: the AI SEO page says models are constantly evolving and commits to continuously adapting strategy, and the AIO page notes AI Overviews shift faster than traditional rankings. There is some evidence of ongoing work rather than a frozen playbook - the EntityMap is versioned (v1.0) and stamped with a generation date of 5 June 2026, and they describe building in-house LLM-readiness tooling. But I found no published original testing on AI search: the blog through August 2026 is practical guidance (social media and SEO, Google Ads experiments, SEO ROI) and their downloadable reports cover the UK SEO market and Instagram content, not generative-engine behaviour. No monitoring process for model or interface changes is described. That is between the two bands, which is why it scored Adequate.

On the record — “AI search is a real named service line, not a single marketing page: separate pages exist for GEO, AEO, AI SEO, AI Optimisation (AIO) and a free LLM audit, each with its own described method.” seoworks.co.uk ↗

Evidence: partly checkable — corroborated in part against the sources below; the remainder rests on the agency’s own account.

Strong — 4 of 5 on this rubric’s scale, from Poor (1) to Excellent (5). This dimension carries 5% of the total score.

Benchmark — across the 16 other agencies evaluated in this discipline, this dimension runs Excellent 2 · Strong 5 · Adequate 6 · Weak 3. The typical agency here scores Adequate, and 2 of them score higher than this one.

What this dimension measures: Independent, verified reviews. Low weight and low expectations: most firms here are under two years old, so an absent review base is a fact about the category's age rather than a finding about the vendor.

Scores low — A substantial verified base scores 4-5; a handful scores 2-3; none located scores 2.

What we found — Clutch profile read first-hand: 4.8 overall from 58 reviews, of which 55 are marked verified. That is a substantial verified base by this category's standards and by any standard - a function of the firm being 17 years old rather than an AI-search startup. The material qualification is that essentially none of it is about AI search: the review base predates the GEO/AEO line and speaks to traditional SEO, PPC and account management. Google and DesignRush ratings appear in search results and on the firm's own llm-info.html page but were not read first-hand, so no numbers from them are cited here. Glassdoor is employee feedback and was excluded.

On the record — “Independent review base read first-hand on Clutch: 4.8 from 58 reviews, 55 verified, $1,000+ minimum project size, $100-149/hr.” clutch.co ↗

On the record — “AI search is a real named service line, not a single marketing page: separate pages exist for GEO, AEO, AI SEO, AI Optimisation (AIO) and a free LLM audit, each with its own described method.” seoworks.co.uk ↗

Evidence: verified — checked against a named source you can open; the links under Sources below are where to check it yourself.