The management board asks what percentage of the category the company controls, and the response is silence – because the three largest competitors are companies that do not publish full financial data, do not disclose their turnover, and do not appear in any ranking. The question of how to calculate market share in such a situation is not a question of a mathematical formula, but of a method for reconstructing data that no one makes available. The research approach described below is used when both the denominator and numerator must be built independently from several independent sources.<\/p>\n
The formula itself is straightforward: company sales divided by market size. The problem is that in most B2B categories and many B2C niches, neither of these values is readily available. A company’s own sales are known, but often under a different definition from the one that needs to be applied to the market (net or gross, including services or excluding them, including exports or domestic only). Market size may be unknown because the category does not correspond to any statistical code, while competitors’ market shares are trade secrets.<\/p>\n
Before any figure can be produced, three definitional decisions must be made. In practice, they affect the result more than the quality of the data sources:<\/p>\n
Only after these three parameters have been established does estimating competitor sales make sense, because it is clear exactly what is being estimated. In practice, the category definition is documented in a short document that is revisited in each subsequent measurement cycle – otherwise, comparability between years no longer exists.<\/p>\n
Reconstructing market shares relies on several parallel approaches. Each has different limitations and potential systematic errors, which is why their evidential value emerges when independent estimates begin to converge.<\/p>\n
The starting point is the total market size, established from macro-level data: production and foreign trade statistics, import and export data based on tariff codes, data from industry chambers and associations, and consumption indicators for a critical raw material or component. The market is then allocated among players based on indirect indicators – the number of outlets, employment, production capacity, and participation in tenders. This method provides good control over the total, but does not differentiate individual companies well.<\/p>\n
Here, the sales of each competitor are built separately from operational parameters: number of employees multiplied by revenue per employee typical for the industry, number of production lines multiplied by output multiplied by capacity utilization, number of locations multiplied by average location turnover, or number of customers served multiplied by average contract value. Competitor data then comes from financial statements available, among other sources, in the KRS Repozytorium Dokument\u00f3w Finansowych and foreign registers, recruitment advertisements, public procurement registers, permit documentation and environmental decisions, and materials containing implementation references. The method can be precise for large, formalized entities, but fails with the fragmented tail of the market.<\/p>\n
Market shares can also be reconstructed from buyer declarations. In B2C, this involves quantitative research asking about the brand purchased most recently and purchase frequency, converted into volume. In B2B, interviews with decision-makers<\/strong> are used to examine supplier structures and estimated category budgets. The resulting indicator is the share of respondents’ spending which, with appropriate sampling and weighting, can approximate market shares. Limitations include recall bias and uneven representation of the largest buyers – in markets with highly concentrated demand, one omitted customer can materially shift the entire result.<\/p>\n Intermediaries are an important source of information: distributors, wholesalers, installers, integrators, and service networks. Conversations with people who observe the flow of goods every day can help reconstruct the proportions between brands and verify conclusions drawn from document analysis. The same mechanism applies in expert research<\/strong>, in which independent specialists estimate the market structure, while discrepancies in their answers help assess uncertainty and identify assumptions that require verification.<\/p>\n Market share should rarely be derived from a single source – it is usually produced by reconciling several independent estimates. A figure that appears to be a single number is, in fact, the result of comparing the top-down approach, bottom-up approach, and channel data. If these three approaches produce results that differ by several dozen percent, this does not in itself mean that one of them is wrong, but it does require the category definition, assumptions, and data quality to be reviewed.<\/p>\n A technique for structuring this reconciliation is triangulation with explicit weights, where the method of aggregating results allows for it. For each estimate, the source, assumptions, and reliability assessment are recorded, and the final result can be presented as a range or a central estimate together with an uncertainty interval. A report stating a share of “approximately 12-15%” and describing the assumptions made is more useful than one reporting 13.4% without explaining where the precision to tenths of a percentage point comes from.<\/p>\n Most unsuccessful attempts to answer the question of how to calculate market share fail not because of a lack of data, but because of recurring methodological errors:<\/p>\n It is also worth comparing an estimation-based approach with alternatives. Retail panels and scanner data provide hard measurement in the categories and channels they cover, often in FMCG, but their scope depends on the data provider – beyond that scope (for example, in parts of specialist sales, B2B, services, or the installation channel), they may not cover the entire market. Transaction data from payment cards or receipts covers selected consumer categories and requires careful conversion to the brand level. Public registers, such as public procurement data, are precise but describe only part of the market. Triangulation-based estimation is therefore a useful method where no complete, continuous market measurement exists.<\/p>\n A market share estimate that is to be used by several departments and updated in subsequent years requires assumptions to be documented. The following list serves as a minimum checklist for model auditability:<\/p>\n Market share models without such documentation may lose their usefulness when the person responsible for the analysis changes – the figures remain, the assumptions disappear, and comparison with the following year becomes impossible.<\/p>\n Market share is a company’s sales divided by total market size, expressed in the same unit and at the same level of the value chain. Consistency between the numerator and denominator is crucial: if own sales are calculated at net manufacturer prices, the market must also be valued at net manufacturer prices. Value share and volume share should be determined separately, because in categories with a wide price range they lead to different conclusions about a brand’s position.<\/p>\n Competitor data is assembled from several layers: financial statements available, among other sources, in the KRS Repozytorium Dokument\u00f3w Finansowych and foreign registers, foreign trade statistics, public procurement registers, recruitment advertisements and information on production capacity, as well as interviews with distributors, installers, and industry experts. None of these sources is sufficient on its own – financial statements cover the entire company rather than the category, while interviews primarily provide proportions rather than absolute values. Only combining these layers can make it possible to reconstruct volumes at the level of individual players.<\/p>\n The basic tests involve checking whether the shares of all players plus the “other” category add up to one hundred percent without forcing the result, and whether the company’s own share calculated from the model corresponds to the best-documented company sales under the adopted category definition. It is also worth comparing the result with an independent demand-side estimate (a buyer study) and a supply balance (production plus imports minus exports). Material discrepancies between approaches indicate a need to verify the definition, data sources, and model assumptions.<\/p>\nChannel-side approach<\/h3>\n
What errors most often undermine market share estimates?<\/h2>\n
\n
What should be included in the estimation documentation?<\/h2>\n
\n
Frequently asked questions<\/h2>\n
How do you calculate market share?<\/h3>\n
Where can you obtain data on competitor sales?<\/h3>\n
How can market share estimates be verified?<\/h3>\n
Ask about estimating market shares in your category<\/h2>\n