Method

How we know what we say we know.

Talenomix sells evidence quality, which makes our own evidence handling fair game. This page states how the Nordic research universe is defined, how sources are collected and graded, and what we have not established yet.

01

Defining the research universe

The unit of study is every publicly discoverable firm, subsidiary or independent consultancy with a current office or sustained operating presence in Sweden, Denmark, Norway, Finland or Iceland that explicitly sells retained executive search, board search or an equivalent active-search service for senior leadership.

We call the result the publicly discoverable Nordic universe. That is a narrower and more honest claim than saying every legal entity has been found, and the difference matters when a firm asks how complete a map really is.

  • Included: independent boutiques, Nordic offices of international firms, hybrid leadership-advisory firms where search is a material service, sector specialists, and firms combining search with board advisory, assessment or succession work.
  • Excluded: contingent-only recruiters, staffing and temporary-labour agencies without a retained offering, job boards, candidate marketplaces, individual career coaches, internal talent teams, interim-only providers, and firms with no current Nordic presence.
  • Excluded pending corroboration: dormant firms and broken sites with no other evidence of current activity.
02

Discovery sources

Association membership alone is not sufficient coverage, and we can show why: Sweden’s ESK accredits individual consultants rather than firms, while Finland has both a member association and a separate, broader commercial company listing. Any map built from one directory inherits that directory’s blind spot.

So the universe is seeded and cross-checked from several directions at once: association and network directories, national company registries, local-language query matrices per country, current job advertisements, procurement awards, and search-firm client announcements.

  • Sweden: executive search, chefsrekrytering, ledarrekrytering, styrelserekrytering, headhunting, searchkonsult
  • Denmark: executive search, lederrekruttering, bestyrelsesrekruttering, headhunting, direktionsrekruttering
  • Norway: executive search, lederrekruttering, hodejeger, styrerekruttering, kandidatkartlegging
  • Finland: suorahaku, johdon suorahaku, johtajahaku, headhunting, hallitushaku
  • Iceland: executive search Iceland, stjórnendaleit, ráðningar stjórnenda, headhunting Reykjavík
03

Collection rules

The research pipeline is deliberately constrained, and the constraints are the point. A market map assembled by breaking the terms of the sites it came from is not a defensible asset for anybody.

  • robots.txt, published terms and authentication boundaries are obeyed without exception.
  • No logged-in scraping of professional networks, no CAPTCHA bypass, no collection from private profiles.
  • Rate limited to one request per domain every two seconds, with at most four domains in flight.
  • Every record keeps its fetched URL, retrieval time, page date, language, content hash and the visible passage that supports it.
  • The original local-language passage is stored alongside its English paraphrase, so translation never silently becomes the evidence.
  • Blocked, unavailable and excluded sources are logged as such rather than dropped.
  • No personal emails, candidate records or contact datasets are ever committed to the public repository.
04

When discovery stops

Discovery stops on stated conditions rather than when it feels finished: every association and directory seed resolved, every firm deduplicated by domain, legal identity and office network, two consecutive country-level passes adding under 2% new qualified firms, the largest firms and international networks manually cross-checked, and every uncertain inclusion labelled rather than silently accepted.

At least 10% of extracted records are checked by hand against their sources. Where a manual check disagrees with the pipeline, the pipeline is wrong until proven otherwise.

05

How evidence is graded

Every observation carries a supporting passage, a source date and a confidence grade, plus an explicit flag separating what a firm stated from what a researcher inferred. A vendor describing its own product is recorded as a vendor claim and never promoted to an independent finding.

A
A dated, direct statement by the firm itself, retrievable at the cited URL.
B
A strong documentary proxy — a job requirement, procurement record or customer case that implies the practice.
C
A workflow inference from consistent public signals, stated as an inference.
D
A single weak or undated signal. Recorded so it can be re-checked, never used to support a public claim.
06

Claim ledger

Every external claim made anywhere on this site, with its source, retrieval date and grade. The build fails if a claim loses its source.

Claim Source Type Retrieved Confidence
Clockwork advertises AI that generates research strategies, sources talent, rates pipelines and delivers qualified longlists “in under 10 minutes”, alongside candidate rating against built-in benchmark criteria, target-list automation and a client portal. “Rate Candidates With Built-In Benchmark Criteria … all with A.I. in under 10 minutes.” Vendor’s own description — recorded, not independently verified. Clockwork Recruiting (opens in a new tab) Vendor claim 2026-07-25 A
Cluen’s Encore advertises “hidden market” mapping of board seats, referrals and investments, AI-driven candidate intelligence, and client-ready reporting generated “in seconds”. “Map board seats, referrals, and investments to reveal introductions others will never find.” Vendor’s own description — recorded, not independently verified. Cluen (Encore) (opens in a new tab) Vendor claim 2026-07-25 A
Thrive TRM advertises relationship mapping, compensation benchmarks, recycling of prior research, and the ability to “assess and stack rank candidates”. “Assess and stack rank candidates.” Vendor’s own description — recorded, not independently verified. Thrive TRM (opens in a new tab) Vendor claim 2026-07-25 A
Sweden’s ESK accredits individual consultants rather than firms — its own wording is that the authorisation applies at the individual level. A firm-level map built from ESK alone would therefore be incomplete by construction. “Auktorisationen gäller på individnivå.” — The authorisation applies at the individual level. ESK — Föreningen Executive Search Konsulter (opens in a new tab) Association directory 2026-07-25 A
Finland has a national industry association, FEX — Finnish Executive Search Firms — formed by established executive-search firms, with a members section. “FEX – Finnish Executive Search Firms is a national industry association formed by well-known and reputable executive search firms.” FEX — Finnish Executive Search Firms (opens in a new tab) Association directory 2026-07-25 A
A publicly compiled Finnish directory lists 151 firms selling executive search, of which 49 do so exclusively, and puts the 2023 Finnish executive-search market at €99 million, down 4% year on year. The compiler states the data is gathered from public sources. “Listan yritystiedot on kerätty julkisista lähteistä.” — The company data in the list has been gathered from public sources. suorahakuyritykset.fi (opens in a new tab) Public compilation 2026-07-25 B
The AESC member-firm directory is currently an interim alphabetical listing with no country or location filter — the page states an enhanced directory is still being built. Association directories therefore cannot be relied on alone to enumerate Nordic firms. “We are currently building an enhanced Member firm directory. While that work is underway, use this interim page to explore AESC Member firms around the world.” AESC — Association of Executive Search and Leadership Consultants (opens in a new tab) Association directory 2026-07-25 A
The EU AI Act (Regulation (EU) 2024/1689) treats certain AI used in employment, worker management and access to self-employment as high-risk, which is why v1 stays out of candidate ranking and selection. EUR-Lex — Official Journal of the European Union (opens in a new tab) Regulation 2026-07-25 A
EDPB guidance states controllers must rely on a legal basis to process personal data lawfully, and that legitimate interest requires weighing that interest against the rights and freedoms of the individuals concerned. Public availability is not by itself a legal basis. “Data controllers need to rely on a “legal basis” in order to process personal data lawfully.” European Data Protection Board (opens in a new tab) Regulation 2026-07-25 A
Bolagsverket provides free valuable company datasets via API, including organisation data, SNI activity codes and digitally filed annual reports; access requires issued API credentials. “Det kostar ingenting och det krävs inga avtal.” — The API is free and requires no agreement. Bolagsverket (opens in a new tab) Official statistics 2026-07-26 A
Henrison publishes a stated position on automated research: “In a digital workplace, we prioritize human-powered research — curated and powered by humans, not code.” “In a digital workplace, we prioritize human-powered research - curated and powered by humans, not code.” Henrison AB (opens in a new tab) Direct statement 2026-07-26 A
Henrison describes itself as one of Scandinavia’s most experienced recruitment research companies and states that it maps the entire talent landscape, working for hiring managers, talent-acquisition teams and executive-search firms. “map the entire talent landscape” Henrison AB (opens in a new tab) Direct statement 2026-07-26 A
A query to the Norwegian Enhetsregister for NACE 78.100 (recruitment and provision of labour) returned 1,369 registered entities, of which 372 publish an employee count and 110 fall in the 5–49 band. 78.100 is a superset that also contains staffing and temporary-labour businesses, so this is a frame to be narrowed, not a count of executive-search firms. Brønnøysundregistrene — Enhetsregisteret API (opens in a new tab) Official statistics 2026-07-26 A
A query to the Finnish PRH/YTJ open company API for main business line 78100 returned 461 registered companies. As in Norway, the code covers recruitment and labour provision generally, so it bounds the search frame rather than identifying executive-search firms. Finnish Patent and Registration Office — YTJ open data API (opens in a new tab) Official statistics 2026-07-26 A
Swedish and Danish registry endpoints returned authorisation errors on 26 July 2026 (Bolagsverket portal 403, Virk distribution 401). Both are documented as open data, but neither is queryable without credentials, so their frames are recorded as unresolved rather than estimated from directories. Erhvervsstyrelsen — CVR system access (opens in a new tab) Direct statement 2026-07-26 A
Brønnøysundregistrene publishes open APIs for Norwegian legal entities and key figures from filed annual accounts. Brønnøysundregistrene (opens in a new tab) Official statistics 2026-07-26 A
Virk provides the public Danish CVR company-information surface used for basic legal-entity checks; annual-report evidence is reviewed separately. Virk / Danish Business Authority (opens in a new tab) Official statistics 2026-07-26 B
The Finnish Patent and Registration Office publishes open company data for legal-entity lookup; market classifications on this site remain separate from the registry facts. Finnish Patent and Registration Office (PRH) (opens in a new tab) Official statistics 2026-07-26 A
Iceland is maintained as a manually reviewed coverage frame because no comparable open official revenue feed has been established for this work. Iceland Revenue and Customs (opens in a new tab) Official statistics 2026-07-26 C
07

What we have not established

  • How many hours a mandate launch actually costs a Nordic firm today. We have hypotheses and no measurements, and the first paid pilots exist to produce them.
  • Whether the value concentrates in mandate launch, client reporting or firm-memory recovery. Those imply different products, and we would rather find out than assume.
  • Which CRM matters most in this segment, and therefore which integration to build first.
  • Whether firms will export CRM data at all under a pilot agreement. If they will not, the wedge does not work and we would rather learn that in month two than month twelve.
  • What Nordic title and company-structure normalization genuinely requires beyond Swedish, which is the only market we have looked at closely so far.

Privacy · Demo