Dan's Take

A Tale of Two Cloud Assessments

'Market Leader' is a highly subjective phrase.

• By Dan Kusnetzky
• 08/11/2016

An article I was reading in the Wall Street Journal, "IBM's Cloud Position: Two Different Assessments," pointed out that the assessments of IBM's Cloud computing business published by research firms -- Gartner and Synergy Research Group -- differed dramatically. So what else is new? Having lived in the industry reasearch world for quite some time, the different approaches and different conclusions didn't surprise me at all.

Each research firm segments the market in different ways, using different research methodologies, aimed at providing useful research, insight and opinion to different people. What does that mean? It means that the results are going to be different. Get used to it.

The key question, which was the focus of the article, was whether IBM was gaining or losing ground to rivals such as Amazon and Microsoft. The two research firms offered different views of the market.

Suppliers and Consumers
The first thing to consider when evaluating the research published by research firms is the target audience being served. Different audiences need different information. Suppliers hope to learn more about market opportunities: that is, suppliers want to know more about their customers' needs. Consumers of technology want to learn more about who is using what technology and products, and why.

Suppliers, working with limited development and support budgets, wish to find the silver bullet, i.e., the smallest number of products to serve the needs of the largest audience. Often, they're faced with the fact that, at best, they can provide products that meet 75 percent - 80 percent of the requirements of the worldwide market for a specific product or service.

Share of What?
Another point to consider is market share. Although the article did its best to shine a light on the different views offered by Gartner and Synergy, what wasn't clear was whether the research focused on revenues, number of subscribers, the total value of the computing being done by subscribers, or some other metric. As any market researcher could explain, market share by revenues is different from market share by number of subscribers; and that is different from market share by value of the computing being done by subscribers. For example, if a cloud service provider sold its services to every member of the Global 500 and had only 500 clients, it might have significantly more revenue than one that served all of the next 1,000 enterprise customers.

Would market share by revenue highlight that difference? No, it would only show that the first supplier had greater market share by revenues than the second. The second, on the other hand, might win if the metric was number of subscribers.

I'm reminded of something Albert Einstein is supposed to have said: "Everything should be made as simple as possible, but no simpler." Industry research is often a complex topic. Each firm has a different take on how the market works, how technology can be applied to resolve business requirements, and how suppliers should go to market to sell their approach.

Dan's Take: We're No. 1! All Four of Us!
I'm reminded of a report I wrote on the relational database market quite a number of years ago. One supplier had the largest market share when revenues were considered; another had the largest share when the number of licenses shipped was the metric; still another was the leading supplier when the number of customers doing technical computing was the metric; and a fourth was the leading player when the number of commercial users was the metric.

I found it amusing when all four suppliers published press releases saying that I said they were No. 1.

Remember, it's very important to know what's being counted, how it was counted, what was included in the count and what was excluded, if one really wants to understand what a particular variant of "market share" means.