Reviews: client feedback and an honest rating of data sources
We publish reviews only from organisations that have actually commissioned a study from us, and only after checking that the project took place. Anonymous praise, reviews written in exchange for a discount and rewritten testimonials do not appear on this page. Further down the page, you will find our own editorial ratings of the free and paid data sources we use in desk research, with the strengths and weaknesses of each.
Anyone can submit a review through the form on this page, but a review is published only after we have confirmed that the reviewer's organisation was our client and that the person was involved in the project. We publish critical reviews as well as positive ones, as long as they concern the work and contain no personal attacks or confidential project details.
We do not rewrite reviews. We may remove personal data, confidential client information or offensive language, and we mark such edits. We never offer payment, discounts or other benefits in return for a review. A reviewer can ask us to withdraw a published review at any time, and we will do so without asking for a reason.
Reviews of the data sources we use
Most of our studies start with desk research, so the quality of the underlying sources matters as much as the quality of our fieldwork. Here we rate the sources we use most often on four criteria from 1 to 5: coverage, freshness, ease of use and reliability. The ratings are our editorial judgement, and no source has paid for inclusion.
Eurostat
Eurostat is the statistical office of the European Union and our first stop for any comparison across member states. Its harmonised surveys mean that a figure on online shopping, internet use or enterprise AI adoption is measured in the same way in every country, which is rarely true of national sources. Short-term indicators, such as the monthly consumer confidence series, are published within weeks, but structural business statistics often trail the reference year by a year or more. The database browser is powerful but assumes you know the dataset code you are looking for.
Free, documented and citable, with metadata explaining how each indicator is collected.
Finding the right table takes practice, and some breakdowns are suppressed for confidentiality.
OECD Data
The OECD publishes comparable economic, social and sectoral statistics for its member countries and selected partner economies, together with analytical reports on topics such as productivity, digital adoption and consumer policy. For international readers, it is valuable because it places European countries alongside the United States, the United Kingdom, Japan and other major economies in the same framework. Its data are carefully reviewed and well documented. Coverage of smaller EU economies is less complete than Eurostat's, and many series are annual and published with a delay.
Strong methodological notes and analytical publications that explain what the numbers mean.
The data explorer has a learning curve, and dataset structures change from time to time.
World Bank Open Data
The World Bank's World Development Indicators provide long annual series for almost every country: population, GDP per capita, internet use, trade and many other macroeconomic measures. We use them for the basic size and income context of a market, for example population and GDP per capita in Bulgaria and the Netherlands. The interface is simple and every indicator has a stable code and a clear definition. The figures are compiled from national and international sources, so they are only as current and as accurate as those inputs, and the most recent year can be revised.
Covers non-EU markets that European sources do not.
Macroeconomic level only; nothing on specific product categories or consumer behaviour.
Google Trends
Google Trends shows how interest in a search term changes over time and across regions, scaled from 0 to 100 relative to the peak in the selected period. It is quick, free and current, which makes it useful for spotting seasonality, comparing the relative interest in two brands or checking whether a topic is growing. It does not show absolute search volumes, it is based on a sample of searches, and the same query can return slightly different curves on different days. Low-volume terms in smaller markets often show no data at all.
Easy side-by-side comparison of several terms and regions in one chart.
Sampling and rounding make small markets and niche terms unreliable.
Business registers
National business registers, such as Companies House in the United Kingdom and the commercial registers of EU member states, are the primary record of which companies exist, who owns and runs them and, in many countries, what their filed accounts show. For competitor analysis and B2B sampling, they are indispensable. Access differs widely: some registers offer free search and bulk data, others charge per document or publish accounts only as scanned files. The EU's Business Registers Interconnection System links national registers, but the depth of information still depends on each country.
Allows complete lists of companies in a sector, which is the basis for B2B sampling frames.
Accounts are filed with a delay, and small companies may publish only abbreviated figures.
TED
TED (Tenders Electronic Daily) is the EU's official platform for public procurement notices, covering contracts above the EU thresholds from every member state. We use it to see what public bodies buy, which suppliers win, at what contract values and how often, which is valuable for B2G market sizing, competitor analysis and preliminary market consultations. New notices appear continuously and can be searched by country, sector code and buyer. Contracts below the EU thresholds are mostly published only on national portals, and award data are sometimes incomplete.
Award notices reveal winning suppliers and contract values, which few other sources show.
Data quality depends on what each contracting authority enters, so fields can be blank or inconsistent.
Statista
Statista is a paid aggregator that collects statistics from official bodies, industry associations, company reports and its own surveys, and presents them as ready-made charts. It is fast and easy to search, which makes it a convenient way to find out whether data on a topic exist at all. For a report, however, we always trace a figure back to its original publisher, check the definition and the date, and cite that source rather than the aggregator. Some figures are forecasts or modelled estimates, and the method behind them is not always fully transparent.
Useful for discovering which original sources exist on a niche topic.
Modelled estimates and forecasts sit alongside measured data and can be mistaken for them.
Online panels
Online access panels are pools of people who have agreed to take part in surveys, recruited and managed by specialist panel providers. They are the most common source of respondents for consumer surveys because fieldwork is fast and quotas can be filled precisely. They are opt-in, non-probability samples, though, so some groups, such as older people or people who rarely go online, are under-represented, and quality varies by provider. Fraudulent and inattentive respondents are a real risk, which is why we clean every panel dataset before analysis.
Suitable for complex questionnaires, images, video and choice tasks such as conjoint.
Quality varies between providers and requires speeder, straightlining and attention checks.
Tell us what you need to find out
No subscription, no commitment. We send a tailored proposal with scope, method and timeline.
- No price lists — a proposal built around your question
- Proposal within 1 working day
- Bulgarian, English and Dutch