UK Tech Recruitment AI Visibility Index 2026: which recruiters appear in buyer-selection answers

Last updated: 28 September 2026 · First published: 8 September 2026

UK Tech Recruitment AI Visibility Index 2026

Which recruitment firms appear in answer-engine responses when a UK employer needs technology talent? This is a one-day snapshot captured on 28 September 2026 across 53 supplier-selection questions and 1,436 usable answers from ChatGPT, Gemini and Google AI Overview. We tracked 40 firms. Thirty-five were named at least once; five received no selection-answer mentions and are deliberately not identified in the public distribution.

The result is not a best-agency league table. It is a measure of answer-engine naming against one shared denominator. Harvey Nash was named in 18.94% of the 1,436 usable answers, or 272 answers. Even that lead means it was absent from more than four out of five recorded answers.

The numbers in brief

• Harvey Nash: 18.94% of selection answers, or 272 of 1,436.
• Full 40-firm cohort median: 2.68%, or 38.5 of 1,436 answers.
• Supplier-domain source record: 721 of 1,436 answers, or 50.21%.
• Spencer Stuart led technology executive search: 157 of 495 answers, or 31.72%.
• Harvey Nash led contract and project talent: 82 of 451 answers, or 18.18%.

What we measured

The research design contains 75 questions. The public leaderboard uses the 53 questions where a buyer is selecting a supplier. A separate set of 22 informational questions generated 618 usable answers and is reported separately from the leaderboard. The selection set covers permanent technology recruitment (18 questions), technology executive search (18) and contract and project talent (17).

We record the answers returned by the engines. Naming rate is the percentage of 1,436 usable selection answers that name a firm. The measurement receipt contains the definition, full published cut and limits.

The combined selection baseline contains 516 usable ChatGPT answers, 530 Gemini answers and 390 Google AI Overview answers. Together they form the 1,436-answer denominator used throughout this report. This is answer-weighted measurement, not a market-share estimate or an equal-platform vote. A one-day capture also does not create independent trials: it records what each platform returned to the fixed question set on that date. The useful repeat is therefore a like-for-like rerun using the same supplier-selection questions, the same eligible-firm rules and the same platform coverage. That makes later movement comparable without treating a platform count as evidence of buyer demand or commercial performance.

Who the engines name

HHarvey Nash

RRobert Walters

SSpencer Stuart

HHarnham

EEgon Zehnder

HHeidrick & Struggles

Russell Reynolds Associates

KKorn Ferry

AAruba Exec

IIntelligent People

HHays Technology

LLa Fosse

RRichard Wheeler Associates

XXcede

UUnderstanding Recruitment

Across all 40 tracked firms, the median is 38.5 of 1,436 answers, or 2.68%. Across the 35 positive firms, the median is 4.11%. Beyond rank 15, the ceiling is 4.32%: Burns Sheehan and Exec Capital share that rate, followed by Christian & Timbers at 4.11%. The complete 35-firm positive distribution is in the measurement receipt.

The full cohort has 40 firms. Five zero-score firms are excluded from the named public distribution so the report does not turn a zero into a claim about any individual business. The receipt preserves the cohort count and the anonymous zero-score rule.

The service changes the shortlist

Permanent technology recruitment is the relevant purchase when a founder, CTO, CPO or people leader is building engineering, product, data, AI, design or cloud capability. In its 490 usable answers, Harnham led at 18.16%, followed by Richard Wheeler Associates at 15.31% and Harvey Nash at 12.24%.

Technology executive search is a different decision: a board or founder appointing a CTO, CIO, CPO, VP Engineering, chief data leader or another senior technology executive. In 495 usable answers, Spencer Stuart was named in 157, or 31.72%. This is a service-specific result, not evidence that a firm is the right search partner for every leadership brief.

Contract and project talent addresses interim, contractor and flexible delivery capacity. In 451 usable answers, Harvey Nash led at 18.18%, followed by Robert Walters at 13.08% and Hays Technology at 12.64%. The result does not establish that every recruiter offers a managed statement-of-work service.

Where the answers draw evidence

The frozen capture records source URLs used in returned answers. It distinguishes recorded source URLs from a claim about how an engine retrieved or ranked the web. URLs with query parameters, including YouTube watch links, are retained in the frozen capture rather than normalised away.

Of the 1,436 usable supplier-selection answers, 721 (50.21%) contain a supplier-domain source record and 538 (37.47%) contain a comparison or listicle URL. LinkedIn and Reddit appear in a combined 123 answers (8.57%), while 10 (0.70%) contain a YouTube URL. These groups overlap: one answer can contain an official firm page, an editorial comparison and a LinkedIn post. They must not be added together as though they are separate slices of the answer set.

Three practical patterns stand out. First, official firm pages are useful only when they explain a specific bought service, role level and geography; a generic homepage gives a buyer little to test. Second, leadership content and employer guidance sit alongside firm pages in the evidence environment, so a clear executive-search or contractor page matters as much as a brand claim. Third, a buyer should inspect the exact answer and URLs behind the relevant service group before treating any ranking as a procurement shortlist.

This is also why a source table is not an outreach list. A comparison article may be useful because it explains the difference between a retained executive mandate and a contingent recruitment brief. A community discussion may be useful because it exposes a buyer concern about interview design or contractor compliance. Neither fact means a firm should ask an editor, author or participant for a generic mention. The useful next step is to add accurate, independently checkable material where the buyer’s question is actually being answered.

How to use the three service views

The three service groups are deliberately not a funnel. They are different purchases made by different people under different constraints.

For permanent technology recruitment, begin with the team plan. A CTO building a first data platform team needs a different market from a CPO adding product managers or a scale-up adding several software engineers. The growth result shows which names appeared in that slice of questions; it does not tell the buyer whether a firm has the right candidate network for their stack, location, stage or compensation model. Ask for evidence about the role family and team stage before judging any comparison.

For technology executive search, make the appointment definition explicit. A CTO for a founder-led software business may need to build the engineering organisation and set a product platform direction. A CIO in a complex enterprise may instead be accountable for a broad operating estate, risk and transformation. A league-table position cannot resolve that distinction. A buyer should test the search firm’s stated role coverage, the partner who will run the mandate, its relevant market experience and how it evaluates leadership context.

For contract and project talent, separate an individual contractor requirement from interim leadership and from a broader managed delivery requirement. The dataset records recruitment names in questions about flexible capability; it does not certify a supplier’s contract structure, compliance process or delivery accountability. That detail belongs in a direct service review. A practical brief should specify the duration, skills, location, employment-status constraints and whether the buyer is purchasing named individuals or an outcome-based service.

The common operating instruction is simple: start with the service group that fits the work, then look at the captured answer and its sources. A firm with a lower overall naming rate can still be more relevant to a narrowly defined leadership or technical requirement than a higher-ranked generalist. Conversely, a high-ranking name may have little to say about a buyer’s particular hiring motion.

What a technology hiring leader can do next

First, write the purchased unit in one sentence. “We need a permanent VP Engineering for a Series B product business” is a usable brief. “We need recruitment help” is not. The first statement tells a prospective partner, an internal stakeholder and a source reader what evidence is relevant.

Second, inspect the official pages for the firms that appear in the matching service group. Check whether the page names the roles, the business stage and the geography that matter to the brief. Then use the source record as a verification path: read the comparison, community discussion or video in its original context before accepting its implication.

Third, separate a measurement result from a commercial conclusion. A higher naming rate can make a firm easier for an answer engine to recall. It does not establish better candidate quality, lower cost, faster delivery or a stronger working relationship. Procurement still needs the normal checks on remit, ownership, process, compliance and references.

For teams building their own evidence cycle, our guide to GEO and SEO explains why accurate, specific pages matter more than adopting a fashionable label. The relevant action is to improve the clearest missing evidence for a real buyer question, then measure the same question set again after enough time has passed to observe a change.

What this means for recruitment firms

Start with the service a buyer is purchasing. A permanent technology page should explain the role families, company stages and geography the firm serves, supported by relevant placement examples. An executive page should describe the leadership mandate, search process and partner responsibility. A contract page should make clear whether the offer is individual specialists, interim leadership or managed delivery, with the corresponding commercial and compliance boundaries.

Then connect those service pages to useful evidence. A dated case study can explain the brief, the constraints and the work delivered without implying that one successful assignment predicts every future result. A hiring guide can help an employer choose an approach before it introduces the firm. Comparison content should disclose ownership and use explicit criteria, including situations where another approach may fit better.

Finally, prioritise the weakest relevant service group rather than the overall league table. Inspect the exact questions where the firm was absent, identify which sources supplied the answer, and decide what factual contribution is missing. Improve that evidence before seeking coverage, then repeat the same question set and compare the naming counts against the same denominator.

What this can and cannot tell you

These findings are directional, not definitive. This report can show:

• Which eligible firms were named in this dated selection-answer set.

• How often they were named against the same denominator.

• How the three service groups differ.

• Which captured sources deserve closer inspection.

• It cannot prove a causal link between a firm’s activity and its naming rate.

• It cannot show revenue, placements, hiring quality, fees, candidate experience or cultural fit.

• It cannot tell a firm that it should appear in every question or service group.

Do not chase a blended rank. Split the buyer brief into permanent technology recruitment, technology executive search or contract and project talent. Read the exact answer, assess the official service evidence, and use a source as a research queue rather than a reason to ask for generic coverage. Our AI search tool, agency and agent guide explains the difference between measurement and operational work; AI search agents for SMBs explains the recurring evidence cycle.

Sources

• Measurement receipt: methodology, cohort policy and final figures.
• Harvey Nash
• Robert Walters
• Spencer Stuart
• Harnham
• Egon Zehnder
• Heidrick & Struggles
• Russell Reynolds Associates
• Korn Ferry
• Aruba Exec
• Intelligent People
• Hays Technology
• La Fosse
• Richard Wheeler Associates
• Xcede
• Understanding Recruitment
• Burns Sheehan
• Exec Capital
• Christian & Timbers
• Oliver Bernard
• Robert Half
• OpenSource
• Wilson Grey
• Adria Solutions
• Signify Technology
• Trust in SODA
• OHO Group
• Daversa
• Riviera Partners
• Client Server
• Optima Search Europe
• Renovata
• The Up Group
• Morgan McKinley
• SR2
• Head Resourcing

The official-site links identify the suppliers’ own public presence. They are not presented as a source of the measured ranking; the measurement receipt and frozen answer capture are the sources for that. They give a buyer a direct route to verify whether a supplier describes the service and geography relevant to the particular brief.

FAQ

Which firm was named most often?

Harvey Nash, in 272 of 1,436 usable supplier-selection answers, or 18.94%.

Does the overall table select a firm for my brief?

No. The permanent, executive and contract results use different question groups and answer denominators.

Why are five firms not named in the public distribution?

They had no supplier-selection mentions. The report discloses the count but keeps the identities anonymous.

Will the results stay the same?

No. They are a dated 28 September 2026 snapshot. Repeat the exact measurement before treating movement as meaningful.

If you need to understand the questions and captured evidence that shape your market, start with an AI search diagnostic.