Watch and learn

Editor's note: Leanne W.S. Loijens is a behavioral research consultant at Noldus Information Technology BV, Wageningen, The Netherlands.

Customer experience research is growing in popularity. For companies and service providers it is crucial to listen to customers and find out what their needs are. Customers have become more demanding and new product development is increasingly characterized by close interactions between producers and consumers. Even co-development is common practice nowadays. 

How do you know how your customers experience your product, service, shop, museum, etc.? You can ask them, of course. This is done in traditional market research with techniques like questionnaires, surveys and focus groups. Asking questions seems simple but the answers you get are not always very useful. Why not? “Experiencing” is largely a subconscious process. When you ask people how they experienced your service, they may not know or cannot articulate their thoughts and feelings. People may also want to please you. They may not like your product very much but they may not want to offend you and will say “Yes, I will probably recommend it to other people.” 

What is also important is that many market research studies are done in a lab. Companies invite a number of test participants to come to their facilities and try a number of prototypes of, for instance, a new type of yoghurt. It is often not so difficult for the test participants to choose the prototype they like best but if you ask them why they chose Option A over Option B, the odds are that they have no idea (but they may make up something really good!). Choosing from three prototypes in a lab is, of course, very different from a real-life situation in a supermarket with many brands and packages that may influence choice as well.

Additional techniques are needed 

It is generally accepted that questionnaires, interviews and focus groups have their drawbacks and that additional techniques are needed. In addition to asking people questions, companies move more and more towards observing people in real-life situations. By measuring quantitative data with techniques like eye-tracking, video recording/coding, indoor tracking, facial expression analysis and physiological data measurement, you can get a good, objective impression of what a customer experienced. Based on these data, you can ask questions about the subjective experience of the customer.

Video recording/coding. Observing people in a supermarket or at home can give you a wealth of data. You can use, for instance, camcorders and position them in such a way that you can cover the whole shop or kitchen. The data storage capacity of present-day camcorders is in the gigabyte range which means that you can easily record 8-10 hours. Coding of the videos needs to be done manually. You can use dedicated coding software/services including analysis of the data. You can observe whether people stand still at a shelf and look at your products, take products from the shelf to have a closer look and put them in their shopping basket. In addition to this detailed observation you can also code how people walk through the shop, what route they take and what the hotspots are.

Our firm used videorecording in one of our research projects. The aim of the study was to assess how elderly people in a nursing home experience their dinner and whether their food intake increases when the dinner conditions are optimized. The control group was offered standard meals. For the people in the experimental group, the dining room was altered (tables with table cloths and candles instead of bare tables), there was more interaction between the staff and the elderly during the meal and the meals themselves were of a higher quality. People were observed two days per week for a period of 12 weeks, starting with a four-week control period. During the study the staff asked people about their experience in a number of interviews. During the control period meal satisfaction (taste, appearance, ambience) was rated as good in both groups. This remained unchanged in the control group. There was a significant increase in meal satisfaction in the experimental group. This was reflected in a longer meal duration in this group. Food intake increased in the experimental group. The participants consumed more vegetables (+32 percent) and more potatoes and pasta (+29 percent). The body weight of the experimental group increased (+0.6 kg) while in the control group there was a slight decrease (-0.4 kg). 

Indoor tracking. You can also use indoor-tracking systems to get data on shopper routes, dwell times and hotpots. You can use real-time location systems like ultra-wideband or video-based systems. With ultra-wideband each shopper wears a tag (or the tag is integrated in the shopping basket); with video-based systems a tag is not necessary. In the future, tracking may be combined with experience sampling with questions popping up on the customer’s smartphone when he/she is in a certain area.

Eye-tracking. To get insight in what your customer sees, you can use eye trackers, either eye-tracker glasses or eye trackers built into your shop. Noldus IT carried out an experimental eye-tracker test together with the principal Dutch railway operating company. The aim was to assess how passengers experience their train journey. Two travelers were followed, one traveler (the Must traveler) used the train to travel from home to an appointment for his work, the other one (the Lust traveler) had a day off and was going on an outing. Both travelers started at Amsterdam Central station. The Must traveler had a deadline to meet the test leader outside the train station in Utrecht before noon. The Lust traveler could take any train to Utrecht and did not have a deadline. Before catching the train both passengers had to buy a train ticket, a sandwich/snack (in Shop 1), coffee/tea/soft drink (in Shop 2) and a present (in Shop 3). Both passengers were equipped with eye-tracker glasses and a smartwatch that measured GPS and heart rate. 

After the test, the gaze videos were manually coded. The behaviors that were scored included: searching (in the station hall, on the platform, in the train, in the shops) and gathering information (from paper, from notice board, from mobile phone, from map, from route planner sign). The travelers were asked to fill in a questionnaire when they had reached their destination to get an impression of their subjective train journey experience. 

Of course, it is difficult to draw conclusions based on the data of only two passengers but the test made it clear that by combining quantitative and qualitative techniques, one gets a good impression of what the travelers experience during their train journey and how they experience it.

Facial expression analysis. Reading people’s facial expressions is what you do in everyday life to see whether they like something or not. Even subtle changes in facial expressions can give people away. There is professional software available that can automatically read people’s facial expressions and can tell you whether people look, for instance, neutral, happy, surprised or disgusted and how intense the emotion is. Even the activity of individual muscles or muscle groups (such as for instance, nose wrinkling) can be detected.

Physiological data measurements. Physiological data are important in experience research because consumers’ emotional responses generally co-occur with physiological responses. It has, for instance, been shown that heart rate increases in response to emotions such as anger and fear and decreases in response to disgust. Skin conductance is a method of measuring the electrical conductance of the skin which is used as an indication of psychological or physiological arousal. In addition to heart rate (ECG) and skin conductance you can also measure parameters like muscle tension (EMG) or neuronal activity (EEG).

Virtual supermarket. The above-mentioned techniques can be combined in a virtual supermarket, which may be a good alternative to tests in a real supermarket. The virtual supermarket may consist of a number of computer screens and a PC on which a software package runs that simulates the supermarket. As a researcher you can change the product range, shelf layout, prices etc. Eye-tracking can be used to assess what the shopper looks at and physiological and brain responses can be measured. The virtual supermarket can be used to investigate why consumers like or dislike certain products and how the choice and purchase process works.

Advanced data analysis. You do not always need complex statistical analyses to get an impression of how your customers experience your products. For instance, with edible products you can measure how fast people eat/drink. The speed of eating/drinking and the bite size can give a good impression of how much your customers like the product.

The Austrian Post carried out a study in its post offices before and after they were renovated. Prior to the renovation people had to wait in line for their turn. The post offices underwent a complete makeover. They were equipped with a coffee corner, a play area for small children and a table with magazines. In addition, a ticket system was introduced. Interestingly, people had the feeling that the waiting time was shorter after the renovation while objective measurements made it clear that the actual waiting time was longer. 

Sometimes advanced statistical analysis is a must to get the insights you need. When tracking people in a shop, for instance, it is good to know how often and how long they visited each zone in the shop (and make heatmaps) but it is also interesting to calculate the most-used route through the shop in a sequential analysis. Pattern analysis may also be of interest. Human behavior is often highly structured but people may vary in their routines and it can very well be that people with different routines experience products differently.

Bringing the pieces together

As we have seen, there are several techniques available to get a good indication of consumer experience. Combining methods is crucial to gain insight into something as complex as experience. Each method (including the traditional ones like interviews and questionnaires) can give valuable information. The combination is like bringing the pieces of the jigsaw puzzle together.