How can we help you today?

Research planning

This section covers the milestones of CX research and introduce you to important decisions you will make over the course of your research project.

Aim and research question

The starting point of successful research is a clear research question. Define clearly the aim of your research. Do you want to find out more about a specific service offering, a determined step in your service offering, or a distinct customer segment? A typical research question asks “Why…?” or “How…?”, and stay away from questions that can be answered by yes or no. You could for example ask, “How do my customers experience the booking process?” or “Why do my customers rate the restaurant’s service so negatively?”

It’s helpful to write the research question down so you can always look back at it and align your research with your aim. It might sometimes also be necessary to change the research question over time. That is okay - a qualitative research process evolves. Often, you need to dig deeper or shift once you find an interesting user need or problem.


Based on your research question, you can start to think about how to conduct your research. When you make decisions on the following aspects, always align them to your research question. One well known research approach, triangulation, can help you gain deep insights on the customer experience and helps cross check your data. You can  triangulate with methods (e.g. include various methods like interviews, surveys, and observations), data types (e.g. text, pictures, and video), participants (e.g. customers, employees, and management) and researchers (e.g. customer service, marketing, and developers).


Deciding on a time frame is necessary in order to get valuable data. The time frame of your research will depend on your research question, and the time frame of the experience you want to examine. 

If you are interested in a specific part of the service offering that usually takes 15 minutes (e.g. a customer gets in contact with your customer service in order to solve a problem with her flight booking), you’ll have a different time frame than with a service offering which takes 15 days (e.g. the period from the booking until the flight). 

Thus, you should be clear if you want to research the customer journey during a specific point or if you want to zoom out and look at a your offering from a higher level. 


Who are the people relevant for your research? Is it customers? Employees? Other stakeholders? Do you want to get information about interactions between these groups? This decision will make sure that you only get relevant data out of your time and financial resources. Also, think about how you can reach those people.


Once you know who the relevant people are, you can decide how to approach them. There is a variety of research methods to collect CX data, including surveys, interviews, observations, auto-ethnography, cultural probes, and mobile ethnography. All of them have their pros and cons, thus it depends on you which method you think is most promising in collecting useful and actionable data. As each method comes with certain biases and yields specific types of data, you should never use only one method for CX research. Instead, use method triangulation to level out potential biases by applying two or more methods at the same time:






Cultural probes

Mobile ethnography

Data collection

Customers are provided with the same questionnaire 

Customers are asked to talk about specific issues or experiences

Customers are observed during the situation of interest 

Customers observe themselves, reflect about themselves

Customers collect diverse material in the situation of interest

Customers use their mobile phone to report experiences in real-time


  • paper-based or digital
  • structured, semistructured, or unstructured
  • contextual or non-contextual
  • participatory, non-participatory, or somewhat in between. 
  • covert vs. overt
  • diary studies, photos, videos, audio, artefacts, ...
  • diary studies, photos, videos, audio, artefacts, ...
  • open vs. structured approach
  • mobile device, SMS, paper-based
  • service safaris


  • makes data and respondents perfectly comparable
  • depending on the grade of structure, respondents can actively influence the treated subjects and express what is important to them
  • more objective view on behavior
  • insights into the person’s inner thoughts

  • abstract descriptions become more comprehensible. 
  • recall of information is supported
  • recall bias minimized through reports in real-time
  • consideration of geographical movement


  • static
  • respondents can only answer those questions that are asked
  • interviewer effect: the interviewer influences the situation and consequently might impact the answers
  • observer effect: people might behave in a way they think it is expected
  • time- and cost-consuming
  • bias caused by researcher’s prior knowledge and experiences
  • collection might be effortful
  • effort for participants

Researcher’s challenge

  • asking the right questions
  • asking the questions right
  • guiding the interview’s path
  • remaining objective 
  • perceiving important information
  • controlling the influence he has on the situation
  • researcher: briefing the auto-ethnographer correctly
  • auto-ethnographer: conscious reflection and report of situations 
  • collection/report of cultural probes
  • researcher: briefing the participant correctly
  • participant: conscious reflection and report of situations 

Depending on the method chosen you can collect diverse data types: text, numbers, photos, videos, audio recordings, or artifacts. Each media offers different insights and the combination extends the view enormously.


Depending on your research method, you can analyse your data differently. Since most of CX research methods have their roots in qualitative research, analysis is often done by codifying data. Codifying data is the process of assigning specific keywords to all pieces of data you collected - might it be a quote from a customer, a photo, or an artifact - in order to count the number of keywords afterwards. This allows researchers to get an overview of the most relevant keywords. 

When analysing data, keep in mind that any researcher is biased by his or her past experiences and knowledge. Thus, including more than one researcher into analysis helps to increase objectivity. Researcher triangulation is also helpful to gain diverse views onto the same set of data. This allows you to discuss critical points and by doing so gain an even deeper understanding.

Once you have done your first analysis and you’ve found first critical moments in your customer's’ journey, you can start to improve your service offering. You do not have to find every error in the first attempt. The research is not about exact percentages of how critical a specific issue is, but to get a ranking of user needs, problems, and desires. This is similar to a usability test in software development. You start by fixing the biggest problems first and then start another research loop to identify the next greatest issues. It is best to start by fixing some of the issues, then repeat the research process and do another loop. 

Also, check to make sure you really got the data you needed. If this is not the case,you might need to choose another sample, or simply brief your participants more clearly. 

Did you find it helpful? Yes No

Send feedback
Sorry we couldn't be helpful. Help us improve this article with your feedback.