Analyzing Customers to Identify Competitors

Marketing strategy is usually cast as finding an effective competitive position in a market. How does your value proposition compare to competing value propositions and customer needs? An effective position should lead to competitive advantage. The trick, of course, is to know who the competition is.

Ultimately, customers decide who your competition is: they choose to spend their money with you or someone else. That means the best competitor identification is customer-focused competitor identification. Here are some ideas under the two fundamental ways to do any research on customers: you can observe them or you can ask them.

Observing Customers

“Watch what they do, not what they say.” Even sincere customers do not always do what they say they will do. Psychologists describe this as the gap between attitude and behavior.

The most powerful observation is that of purchase. Customers are rarely 100% loyal to a product. They usually switch back and forth across a smaller “consideration set” of satisfactory products for reasons as simple as distribution availability or variety-seeking. In B2B, customers often prefer to support second suppliers to improve negotiating power and minimize risk.

Companies or third-parties (e.g., research firms) sometimes track customer panels that allow them to observe this switching behavior over time. If you consistently share customers with another product, you are in competition with that product.

Here is a disguised version of an actual brand switching matrix for a frequently consumed food category. The rows indicate the number of customers purchasing a brand in time 1, and the columns show what brand the same customers purchased in time 2.

From this table, you can learn what products substitute for one another. Brands A and B are very popular and also share a fair number of customers from time 1 to time 2. This is typical of many brand switching matrices. Notice how the switching numbers decline within a row; switching often occurs in direct proportion to market share, mostly to the top brand, then #2, then #3. Byron Sharp (2010) calls this the “duplication of purchase” law. Only Brand E is a bit odd here, as customers of Brand E are more likely to switch to Brand D than the higher share Brand C, and more likely to receive switchers from Brand A than Brand D.

A variation on this approach is to look at what happens when you engage in tactical experiments. What happens when you cut price or increase advertising? Do you draw disproportionately from certain brands? A switching matrix with percentages gained or lost due to a change in price is called a cross-elasticity matrix. Competing brands that are strongly affected by your actions are probably in a customer’s consideration set with yours. Similarly, which competitors affect your results?

Beyond purchase, you can also track behaviors that indicate consideration. Digitally, you may be able to track behaviors or speech indicating two products are in the same consideration set. For example, if someone spending a lot of time on product 1’s webpage also spends a lot of time on product 2’s page, that suggests the two are in competition. Similarly, the extent to which two products are searched together also can uncover competitors. Finally, text mining of online content can identify the extent to which two products are discussed together (“product a vs. product b”), which suggests they are in competition.

Asking Customers

Complementing observing with asking is a very useful tactic for three reasons.

• First, you may not have enough observational data or data on the right shoppers.
• Second, many observational techniques assume you know what to observe (e.g., switching between Brand A and Brand B), meaning you’re less likely to find the potential or indirect customers you don’t know (Brand Q).
• Third, observations are very good at telling us what is happening. They’re much less good at telling us why something is happening.

The simplest approach is to talk to your customers. If you have a sales or service force, they can be a great tool for uncovering competition. Going to your customers’ trade shows (as opposed to yours) can help you learn how your customers view the world. If you use intermediaries such as distributors and retailers, they may also have insights.

All that is going to be useful but informal. It’s also likely to be highly unrepresentative of your customers as a whole, since your sample will consist of the people who are willing to talk with you. But it can be great for generating ideas that you explore or confirm through other means. You may want to go on to more formal market research techniques. Here are some good tools to consider.

Survey questions. Formal questions that get at a customer’s consideration set are good to try. Here are three good ones:

• Name all the products you can think of in [context].
• What products would/will you consider the next time you purchase [context]?
• What is your favorite product [in context]? If that product weren’t available, what would you buy instead? (You can keep asking this question until they run out of products.)

The first question simply identifies what is called “unaided awareness.” If customers don’t think of your brand, you’re in way more trouble than you thought! Note the order of answers given. Early brands in the list are literally more “top of mind” than later answers.

The second question identifies consideration sets. In most cases the answer will be relatively small, typically between two and six products. These are likely direct competitors depending on the context.

The third question is a different way of identifying consideration. By deleting products one at a time, you get a hierarchy of competitors within the consideration set.

The [context] part of these questions is very important. The temptation is to ask about a product category. “What is your favorite car? What cars will you consider the next time you purchase a car?” This may be helpful in identifying and ranking direct competitors, but it’s often more interesting to ask about a customer problem or usage situation. “What is your preferred way of getting to work?” is likely to identify a lot of competitors to owning a car (e.g., Lyft). Ask about a handful of common usage situations or problems for which your customers use your product to reveal competitors from outside your industry.

Substitution in Use. Substitution in use formalizes the usage context question. It works in two phases. Take a sample of customers and ask them the situations in which they would use your product, e.g., when would you take an ibuprofen? Collect these usage situations and identify the popular and the surprising.

Popular uses are probably your core product-markets. Surprising uses may be an opportunity for innovation. A surprise might be that people take ibuprofen for a sore throat. Now take a second sample of customers and ask them to identify all the products they might use for that situation. For a sore throat, ibuprofen’s competitor may be throat lozenges or tea! That could be an opportunity for a product variation: for example, Vicks extended its NyQuil brand to create a ZzzQuil sleep aid.

Similarity judgments. Products that customers see as similar to one another are likely to be competitors. There are a variety of ways to solicit similarity judgments, but the result can be visualized in either perceptual maps or tree diagrams that can be very insightful.

Conjoint Analysis. Conjoint analysis is a research technique in which customers are presented with sets of alternatives from which they make a choice in each set. Over a series of choices, customers reveal what they care about. If you incorporate existing products and brand names into this effort, you can get a sense of what other products are particularly competitive with your offering.

The one caution I want to note here is sampling. The easiest thing to do is to talk only with your current customers. That’s also problematic because your current customers are not likely to be representative of the market as a whole. Be sure you talk with people beyond your own customer base, e.g., competitor customers, non-users of the category, lost customers.

In summary, try to find out which other products customers consider in competition with yours, remembering that some of the most interesting competitors may come from outside your industry. Do this by observing customer behavior and asking about customer attitudes.

I hope this helps you in your practice. This article is an updated excerpt from a longer piece on competitor identification that appears on Medium. Below are some additional references. 

 References and Further Reading

Clark, Bruce H., and David B. Montgomery (1999), “Managerial Identification of Competitors.” Journal of Marketing 63 (3): 67–83.

Lehmann, Donald R. and Russell S. Winer (2008), Analysis for Marketing Planning, 7th edition. New York: McGraw-Hill Irwin.

Sharp, Byron (2010), How Brands Grow. Oxford, UK: Oxford University Press.

Shugan, Steven M. (2014), “Market Structure Research,” in The History of Marketing Science, R.S. Winer and S.A. Neslin, eds., Singapore: World Scientific Publishing, 129-164. For those of you who want the details, this is the best recent review of academic research in competitor identification.