ESOMAR Directory — Find a Research Partner Data

It is the buyer-facing index of the insights industry, converted from a search experience into rows. ESOMAR, the global association for market, social and opinion researchers based in Amsterdam, operates the Find a Research Partner directory so buyers can locate member organisations by geography and specialism. Listings present the registered company name, the country the organisation files under, the capability descriptors it claims, and a contact surface for enquiries.

Two properties make the universe worth extracting rather than merely browsing. First, inclusion is gated on live membership status - the directory reflects firms that currently belong, not a frozen register of everyone who ever did. Second, the same organisations self-describe through capability tags, which turns qualitative impressions ('the Dutch market skews quant') into counts. The universe runs to thousands of listed organisations, though the directory declines to publish an exact total on its landing page.

Within Datadory's catalog of 1,744 datasets across 159 viable industries, this slice sits in the renewable electricity pool because insight spend follows capital programs: the energy transition is the largest capital program running, and every market-entry case, offtake study and policy brief behind it starts by hiring someone who understands a local market. The record scores 6/10 on our quality rubric.

One organisation, one row. A record names the firm exactly as the listing displays it - Ipsos, filed under Netherlands, carrying Quantitative, Qualitative as its claimed capabilities. Every row keys on company_name x country, so the entire directory resolves into one addressable table that rolls up cleanly by either half of the key.

The variants matter as much as the reference row. Some listings claim a single tag; some stack several; some publish no capability descriptors at all. That unevenness is not noise to be smoothed over - it is the difference between markets whose firms describe themselves thoroughly and markets where the self-description layer is thin, and it survives intact in the rows instead of being flattened by an intern with a spreadsheet.

Organisations operating in multiple markets can hold separate listings per office, so a global firm is not one ambiguous row but several precise ones - one per filing country. Multi-office presence beyond the primary country folds under additional fields on request.

Three verified core fields. company_name captures the registered name as displayed in the listing, which is the string every downstream join depends on. country records the jurisdiction the organisation files under - the directory's own geographic spine, and the axis most supplier searches start from. services_offered holds the capability descriptors the profile claims, such as quantitative, qualitative, fieldwork or analytics.

The dictionary stays deliberately conservative because the directory's deeper material renders unevenly. Contact blocks, profile prose and membership markers appear on some listings and not others, and the capability taxonomy runs deeper than the headline tags. Anything beyond the three core columns folds under additional fields on request: full contact surfaces where published, descriptive copy, membership identifiers, extended specialism categories, multi-office country presence and brand assets - all enumerated and confirmed at sample scoping rather than promised blind.

Geography: global by construction. Member organisations list across all major markets and are reached through country browsing, so the extract spans the mature research markets of North America and Western Europe alongside the high-growth insight economies of Asia, Latin America, the Middle East and Africa. Depth follows where members choose to file, which makes country counts themselves a finding worth reading.

Temporal: a current-state universe. Each extract reflects the listings live at extraction, and no archival series ships behind the rows - so treat successive extracts as a time series you build, not one you inherit. That suits the questions this dataset answers: who is in the market now, not who was there in a prior era.

Granularity: one record per member organisation, rollable to country or capability tag without losing the underlying rows. Scale runs to thousands of listed organisations; the exact total is not published on the public landing page, so the first extract doubles as the census.

Because the people commissioning renewables research need the same structured view of their suppliers that they already keep on their assets. A developer entering a new market commissions resource assessment, offtake analysis, policy tracking and customer research; a grid operator running a public-engagement program needs opinion researchers who know the region; an investor screening a portfolio wants local evidence partners it can vet. Every one of those engagements begins with the same question - who is credible here - and the ESOMAR directory is the closest thing the industry has to an authoritative answer.

The pairing also cuts both ways. Analysts modelling the insights industry itself use the energy transition as a demand indicator: counting which markets add capability tags like energy, utilities or sustainability research is a measurable trace of where transition budgets land. Rows make that traceable; browsing never will.

Four constituencies, one table. Buyers of research - corporate insight teams, renewable-energy strategy groups, procurement functions - shortlist and vet suppliers by geography and capability. Sellers to the research trade - panel companies, fieldwork networks, software vendors - build account universes and territories off the member roster. Competitive intelligence teams track the supply side over time, diffing successive extracts for new entrants and re-filings. Journalists and academics ground industry-shape claims in counts instead of adjectives.

What they share is the substitution at the heart of this dataset: someone else's curated membership index becomes your queryable table, and the recurring manual task - searching the directory again for the next project, copying listings into a spreadsheet - stops being a task at all.

API, files, or your warehouse. Daily, weekly, or hourly.

The channel follows your stack, not ours. Rows arrive identical regardless of transport - one record per member organisation, keyed on company name and country - so a supplier-longlist workflow and a market-mapping model read from the same definition of a listing.

Cadence is yours to set. A quarterly supplier review and a continuous market-map dashboard pull from the same schema, which is what keeps comparisons honest across deliveries.

Every delivery includes the field dictionary above and validation rows, so the first thing your team does with the extract is check it against evidence rather than hope.

As self-description, not certification. services_offered records what an organisation claims about itself, which is exactly the right basis for market mapping and the wrong basis for quality scoring - a firm tagging both quantitative and qualitative tells you how it positions, not how it performs.

Treat absent tags as absent claims rather than absent capability, and let the gap distribution inform your reading: markets with dense tagging give sharper segments, markets with sparse tagging need scoping calls before segmentation is safe. Both behaviours are visible in the rows, and neither required anyone to read a single listing page.