Removing the Rashomon Effect from Market Analysis
The classic 1950 film Rashomon by Akira Kurosawa presents four people’s perspectives on the same event. Naturally, despite all observing the exact same thing, each person provides a different account. As each gets a chance to tell their story, it becomes more and more apparent how subjective each perspective is.
The movie’s influence is so great that it led to use of the phrase “the Rashomon effect” to characterize subjective, contradictory, and self-interested interpretations or descriptions of a situation. We are left wondering whom to believe, what their motivations are, and what really happened.
If you follow the news and analysis of the language services and technology market, this is a familiar feeling to you. Multiple people and their organizations look at the same industry. But while some of the information that each provides may be the same or similar, each tells a different story in terms of market drivers, size, growth, what happened, what is happening, and what will happen.
Wouldn’t it be great if we could eliminate the Rashomon effect from the industry? If somehow, we didn’t have to rely on guesses, estimations, or conjectures?
If you have read this blog before, you know that that’s exactly what CSA Research does. Our Global Market Study is an extremely complex and in-depth survey in which companies provide us their actual numbers and perspectives. That’s right, accurate and complete data provided directly to us from hundreds of companies. And not just the top few worldwide or in each region; hundreds of firms of all sizes and locations share with us their confidential data. It’s a significant ask of them to make their way through such a long survey and trust us with their data. We are extremely grateful that so many recognize our integrity and continue to do so year after year.
For those who want to be ranked globally, regionally, or (this year) for interpreting or for content types, we have them verify their most critical datapoints a second time after providing it in the main survey. We then run statistical tests and year-over-year comparisons on submissions. We contact the companies for which we identify any possible outliers, errors, or discrepancies in any of the data. In some cases, to provide the full picture of the industry, we also gather data from companies’ public annual reports and financial disclosures. We don’t deal in projections or estimates about companies, but facts.
For this reason, some companies from which we do not have first-hand data don’t appear in our rankings. There’s no way around it, unless every provider in the industry were public and were required to release data by law. Those omissions do not affect our market sizing or analysis because our survey gathers a representative sample of the market. Once we have that, we use our database of 18,000+ companies and our proprietary algorithm to provide consistently accurate analyses based on verified data.
We have explained our rigorous methodology in past blog posts and also publish an even more comprehensive explanation of it (20 pages!), available to anyone. If a methodology is not clearly and exhaustively explained, be wary of the data you are viewing. You should question the source and how the data is derived before trusting it to make business decisions.
The depth of the information we gather in our Global Market Study provides the insight for hundreds of pages of proprietary primary research every year – for example, staffing levels in different regions, volume of demand, how to target specific verticals, maturity levels, technology used and in what situations and volumes, standards, plans for mergers and acquisitions, words translated per year, and more. Rankings and lists of company names are the easy part, in comparison.
It is extremely difficult to do such large-scale, rigorous, primary research like this. It’s why no one else in this industry has attempted it. But it’s also why our rankings, lists, and data don’t change once published. We don’t need to revise any revenue numbers later in the year when 2019 financials have been finalized; companies provide us those numbers when they have them finalized. We expedite our study and analysis as much as is feasible, but we don’t rush to be first or even second. Not coincidentally, in Rashomon, the first account given is the least accurate of them all.
Why is this important? Investors, buyers, industry watchers, and LSPs themselves need data that is unchanging. Important decisions on strategy, investment, and everything else should not be made on data that is subject to change or on market analyses with a wobbly factual basis.
Rashomon still matters 70 years later because it brings up important questions about the nature of human experience and reckoning with our own subjective experience as we go through life. We want our data to have integrity and still matter not just months from now, but years from now. Though our primary research reports won’t ever reach the artistic heights of Rashomon, that’s OK. We aim to be the sun shining through the trees at the end of the movie instead of the overcast sky obscuring the truth that is seen through most of the movie.
We hope you continue to trust our market research perspective that comes from a solid foundation of immutable data and the unmatched analysis that this important industry deserves.
About the Author
Director of Data Products
Focuses on system integration and data validation, analysis and visualization
It is incredible to think that, less than eight years after the first publicly available neural mach...
Read More >
May is Mental Health Awareness Month. During this period, organizations seek to raise awareness of t...
Read More >
Companies all over the world are experimenting with breaking away from the traditional five-day work...
Read More >
Multimedia, transcribed audio, and AI-generated content in all the world’s digital languages join m...
Read More >
FOMO – the fear of missing out – relates to the anxiety people feel when they are worried about mi...
Read More >
It’s planning time once again, but this task is especially fraught this year as companies are facin...
Read More >