Loading

Understanding customers in depth The insights

Andy Raskin is a man who knows a lot about expressing ideas in a narrated way, and he uses this talent to help new businessmen -most of them start-ups- define what they do. He helps them to create a short phrase (sales pitch) to make understand not only what they do and sell, but also to create the will to buy or invest.

On his own website he could say something like “I write texts for start-ups to help them achieve success”. This sentence would clearly indicate what he does and who he is addressing it to. Instead, his sales pitch is:

“Strategic messaging for aspiring unicorns”

Raskin detected a very common insight among those who create a startup: they seek that their company becomes a “unicorn”, a word that in the financial jargon is applied to companies worth more than 1000 millions USD.

Stating such a common ambition of his priority customers gives his sales pitch a really superior sales capacity, without losing an apex of the clarity of what he does.

This chapter aims to be a brief guide to capturing insights, including an explanation of the most relevant techniques to detect them.

What we mean by insights

Nowadays it’s very common to hear about insights quite often. Most of the marketing departments have them in their regular vocabulary, and they affirm basing their strategy on those insights. But many times they are just data or observations which, even though they might be interesting, are not actual usable insights.

Even the expression is on everyone’s lips, a totally convincing definition is not easy to find.

So first we should be clear about what we mean by insights. Below, I suggest some definitions I find interesting, given by Conroy and Frost (Ramalingam & Ghosh, 2009). They are complementary and show the main characteristics of the insights.

Conroy defines the insight as:

“A statement based on a deep understanding of your target consumer’s attitudes and beliefs, which connect at an emotional level with your consumer, provoking a clear response, which, when leveraged, has the power to change consumer behaviour”

For Frost,

"It’s a truth that has not previously been articulated […] and has actionable implications"

which means it can be used in a practical way to achieve a deep connection with potential customers.

Data are not insights

One of the most frequent problems is the confusion between data and insights.

Data, whether a statistic, a conversation, a piece of news we hear on the radio, etc., informs us of a certain thing, intervention or situation. It allows us to know something that already exists, and that can also be put in context.

Insights, on the contrary, allow us to understand why certain situations or behaviours take place. It’s about understanding what makes our potential customers behave in a certain way.

In other words, an insight provides a deep and correct understanding of the customer, which can not be provided by statistical data. That’s why it is so relevant.

If it is widely known, it’s not an insight

Another characteristic of the insights is that they are “hidden”. Those things that belong to the general knowledge of the sector are not insights; they are data. For this reason, detecting insights entails the discovery of something relevant and unknown. At the same time, paradoxically, when it’s discovered, most of the times it seems so obvious that it makes one wonder how he/she couldn’t realize it before.

For instance, some years ago an American company, leader in producing sliced bread, detected that the purchases made in this product category by the Spanish customers during the weekend varied depending on the weather.

They sold more ....

... during those weekends that were cloudy ...

... but without rain.

The reason is that if customers think there will be bad weather, they stay at their city home, but if after all it doesn’t rain, they end up going shopping.

They are not normally based on something conscious

The brain is a “machine” that never stops deciding. Neuroscience assures that we mostly do it in a non-conscious way. In the previous example about what to do during the weekend, I don’t know anyone who would take the decision based on an Excel file or deep mathematic calculations.

Therefore, it’s easy to deduce that the insights -as the lever of a customer decision or behaviour- have a con-conscious nature. This is why most of times they have their roots at the limbic system of the brain.

If it is not actionable, it’s not an insight

Detecting an insight is an “Aha moment” in which one feels that deep understanding of the customer can be used to transform the business through a marketing or commercial action. Then we say that the insight is “actionable”. In the opposite case, it’s a finding that, although may interesting to know, is not useful to improve the business and it’s not monetised. In the sliced bread example, the company started to use the weather forecast strongly to make the production and distribution plan to replenish the stores. The economic results improved significantly.

A single (strong) insight is much better

It’s tempting to find as many as possible. However, that isn't the way. Although it’s not impossible, it is really difficult to find 5 authentic and relevant insights per project. But even if they were so, as the insight has to be like a thread running the strategy, it’s pretty clear that it is more efficient if the insight is based on a single lever and not on many. If not, the strategy could lack of focus.

"Strategy is a denial exercise,"

as I heard from Prof. Paco Vilahur many times.

The sequence to capture insights

The following chart shows the process to detect insights and to create the solution or marketing and commercial strategy.

How can we capture insights?

These are some of the techniques that allow us to capture insights, providing information or clues that will have to be processed and verified later. These are not the only ones, but all of them are efficient if well-executed, and can complement each other.

1. Observation

A way to detect insights is through observing a process done by the consumers or business customers: open the eyes and listen.

But to go further, this observation should have a previous intention, that is, it should be focused on aspects previously formulated as hypothesis. An observer is more than a voyeur. Same way, listening is much more than hearing.

Another important aspect of this method is to detect the context of what is being observed. It’s not the same seeing someone reading the product characteristics standing in front of the supermarket shelves as sitting on the sofa with a tablet.

2. In-depth interviews

An in-depth interview is like a conversation between the interviewer and the interviewee, characterised by a semi-structured script, with a sequence that can vary depending on the relational dynamic. This characteristic even allows to explore additional clues to those planned.

The goal of the interviewer is to achieve a global understanding of the interviewee point of view (Berry, 1999), as well as of his/her inner perceptions, attitudes and feelings (Zhang & Wildemuth, 2009).

We should not confuse in-depth interviews with the typical survey. In the in-depth interviews the questions are open, and normally not really direct, aiming not to condition the answers that might be obtained.

The interviewer has to be able to generate new questions based on the answers and comments made by the interviewee, following the thread until the last answer is not fruitful any more.

In the interview not only the verbalised things matter. The gestures, body language, tone of voice, doubts, facial expression, etc., matter too. The interviewer has to pay attention to all these aspects, because they are very relevant for insights detection.

The interview also needs to be relaxed, with a conversational tone. If the interviewee feels comfortable and relaxed, the results will be better. So in-depth interviews have to take place in a quiet and relaxed place, that doesn’t “smell like a lab”.

3. Evocation studies

Evocation studies aim to know what is the first thing that comes to the minds of those answering when they are told a word, a concept or a short phrase.

You can ask about the own brand, the product category, some key competitors, etc.

It’s recommended to formulate the question approximately this way:

“If I tell you (phrase or word), could you, please, tell me the first thing that comes to your mind?”

It doesn’t take more than some seconds to answer this kind of question. Unlike with in-depth interviews, in the evocation studies the answers obtained are short.

This type of studies don’t need to be conducted face to face, but they can also be done remotely, for instance, through computer assisted phone interviews (CATI).

The interviewer gathers the exact answer given by the other person, and writes it down. As we ask only the first idea that comes to mind, spontaneously, we make sure the answer has been articulated non-consciously, which is what really matters.

If we let them answer more evocations or let longer time to answer, it’s more than probable that the answer will be conscious. In this case, as it would come from the cortex, it wouldn’t be much useful or it would even contaminate the results.

Once we have the raw results (verbatim answers), a very careful clustering process must be done for the later analysis, based on which we will be able to look for the insights, as the same time as to understand the nature of the situation of the research topic.

An alternative field method would be an online panel. The inconvenient of this method is that, if the software doesn’t restrict the answering time intervals, cortex answers can go through.

4. Focus groups

Focus groups are, together with in-depth interviews, the most common technique in qualitative research.

They consist of a group interaction about a topic determined by the interviewer, with the objective of gathering information (Morgan, 1999). In this type of interactions, people are encouraged to talk to each other instead of with the interviewer, so that he/she can collect information about many people at the same time (Kitzinger, 1995).

Even though focus groups offer advantages such as the ones mentioned above, they also involve some disadvantages that need to be considered. Focus groups are an artificial environment, that can condition the answers given by the participants and the information obtained from it. In addition, they don’t enable to go as much in depth as the interviews do, because the interaction is (mostly) among the participants and not with the interviewer. The skills of the focus group leader are crucial for the effectiveness of such technique.

As it doesn’t have statistical validity, it’s essential that all the people invited to a group belong to the same customer or consumer segment.

Nowadays, there’s a trend of managers who underestimate the power of the focus groups. My opinion is that it is due to the fact that many reports have been done just by “drawing up the minutes” of what has been said. If the person leading and interpreting the focus group is an expert professional, the method is really very useful.

5. Projective techniques

Will, Eadie and MacAskill (1996) use different definitions of projective techniques, considering them as those that “indirectly encourage respondents to reveal feelings and non-conscious attitudes, without being aware they are doing so”, by asking them to interprete other people’s behaviour.

When they are asked “to interprete the behaviour of others, they are actually projecting their own way of seeing things”, without feeling approached directly by the interviewer.

If we ask interviewees about their own motivations, they might answer from the cortex what they consider to be socially accepted or stereotyped. On the contrary, with the projective techniques they are more likely to answer from the limbic system.

An example of a question with projective technique would be:

“You see your neighbour buying a gossip magazine in the kiosk. Why do you think she bought it?”

This technique can be used in any of the previous cases: in-depth interviews, evocation studies and focus groups.

6. Big data and artificial intelligence

A relatively new method has recently been added to the previous ones and it looks promising. We are talking about artificial intelligence applied to huge amount of customers-related data, previously gathered by the company.

For instance, a phone company wanted to detect the profiles of people who stopped being loyal. Through these fuzzy logic techniques, the company detected this profile:

Women around 55-60 years old, who had been customers for many years, with very basic use of their phones, without much voice consumption, with no notice of serious complaints or problems, and always paid the bills on time. All of a sudden, they leave the phone company. Up to here, this is the profile given by the method.

The company went in depth with some of these cases and understood why. Those women had just become grandmothers and their sons or daughters offered them the possibility to receive pictures of their grandchildren via Whatsapp. But with their conventional cell phone that was not possible. They went to the phone company looking for a smartphone, and the company asked for quite a big budget, as they had few points in their customer loyalty schemes. Then they went to the competitors, who provided them with a preferencial low budget as new customers.

The new status of “grandmas” was the insight that made their behaviour change.

If done correctly, this quantitative technique has the same validity as the previous ones, but it can’t be used if the company is rather new or if it hasn’t being tracking the customers commercial activity before.

BIBLIOGRAPHY

Berry, R. (1999) Collecting data by in-depth interviewing. British Educational Research Association Annual Conference

Kitzinger, J. (1995) Introducing focus groups. Glasgow University Media Group.

Morgan, D. (1996) “Focus groups” en Annual Review of Sociology, Vol. 22, pp 129-152.

Ramalingam, S. & Ghosh, A. (2009) “The “insight” story”, en Consumer Insights. ESOMAR

Zhang, Y. & Wildemuth, B. (2009) “Unstructured interviews” en Applications of Social Research Methods to Questions in Information and Library Science. Wesport: Libraries Unlimited.

Will, V., Eadie, D. & MacAskill, S. (1996) “Projective and enabling techniques explored” en Marketing Intelligence & Planning. Vol. 14, pp 38-43.

______________________________________________________________

Text by Lluís Martínez-Ribes, BCN, 2017

________________________________________________________________

Now that you've read the chapter, we invite you to take the quiz

If you want to give better answers (and get a better grade), think if you should read the note again before going to the quiz.

Credits:

Created with images by Sharon Mollerus - "Statue of Phillis Wheatley" • MorganK - "question mark question why" • Ramdlon - "sky light cloudy" • Unsplash - "beach seascape ocean" • MrRockker7 - "reed" • Skitterphoto - "binoculars dusk sunset" • Matt From London - "Lego interview" • daniel.stark - "evocation of springtime" • robynhobson - "imagine motivation belief" • NASA Goddard Photo and Video - "From "Pursuit of Light" - Perpetual Ocean"

Report Abuse

If you feel that this video content violates the Adobe Terms of Use, you may report this content by filling out this quick form.

To report a Copyright Violation, please follow Section 17 in the Terms of Use.