Joni

About the author : Joni holds a PhD in marketing. He is currently working as a postdoctoral researcher at Qatar Computing Research Institute and Turku School of Economics. Contact: joolsa (at) utu.fi

Customers as a source of information: 4 risks

english

Introduction

This post is based on Dr. Elina Jaakkola’s presentation “What is co-creation?” on 19th August, 2014. I will elaborate on some of the points she made in that presentation.

Customer research, a sub-form of market research, serves the purpose of acquiring customer insight. Often, when pursuing information from consumers
companies use surveys. Surveys, and usage of customers as a source of information, have some particular problems discussed in the following.

1. Hidden needs

Customers have latent or hidden needs that they do not express, perhaps due to social reasons (awkwardness) or due to the fact of them not knowing what is technically possible (unawareness). If one is not specifically asking about a need, it is easily left unmentioned, even if it has great importance for the customer. This problem is not easily solved, since even the company may not be aware of all the possibilities in the technological space. However, if the purpose is to learn about the customers, a capability of immersion and sympathy is needed.

2. Reporting bias

What customers report they would do is not equivalent to their true behavior. They might say one thing, and do something entirely different. In research, this is commonly known as reporting bias. It is a major problem when carrying out surveys. The general solution is to ask about past, not future behavior, although even this approach is subject to recall bias.

3. Interpretation problem

Consumers answers to surveys can misinterpret the questions, and analysts can also misinterpret their answers. It is difficult to vividly present choices of hypothetical products and scenarios to consumers, and therefore the answers one receives may not be accurate. A general solution is to avoid ambiguity in the framing of questions, so that everything is commonly known and clear to both the respondent and the analyst (shared meanings).

4. Loud minority

This is a case where a minority, for being more vocal, creates a false impression of needs of the whole population. For example, in social media this effect may easily take place. A general rule of thumb is that only 1% of members of a community actively participates in a given discussion while other 99% merely observe. It is easy to see consumers who are the loudest get their opinions out, but this may not represent the needs of the silent majority. The solution would be stratification, where one distinguishes different groups from one another so as to form a more balanced view of the population. This works when there is an adequate participation among strata. Another alternatively would be actively seek out non-vocal customers.

Conclusion

Generally, the mentioned problems relate to stated preferences. When we are using customers as a source of information, all kinds of biases emerge. That is why behavioral data, not dependent on what customers say, is a more reliable source of information. Thankfully, in digital environments it is possible to obtain behavioral data with much more ease than in analogue environments. The problems of it emerge from representativeness and on the other hand fitting it to other forms of data so as to gain a more granular understanding of the customer base.

Related Posts