I am at a company with a low UX maturity and have little access to actual user data (no interviews, feedback filtered through business analysts, no user tests). What we do have however is a large amount of support tickets in various stages. I would like to use these tickets as a basis for research but I am not sure how to best go about extracting relevant data / structuring it. Has anybody done something like that? I am thinking of going with qualitative coding to try and find patterns and major pain points. Appreciate your input!
Data preparation is a process of cleaning and transforming raw data prior to processing and analysis.
In the first step You can organize and collect data Later it's good to cleanse and validate data.
- Removing extraneous data and outliers.
- Conforming data to a standardized pattern.
- Masking private or sensitive data entries.
When the pattern and structure will be create for data it's easier to see corelation (for example if 30% questions concerned specific area of system - You have a strong research base for improving this scope.
It's difficult and time-consuming process. If you succeed with search automation tool for it, you save a lot of time.
In your case probably, every ticket have similar structure, if your system categorize user question into categories (for example problem with A, B, C (...) it will be easier to put appropriate data to appropriate bag.
Yes, this is another source of information from which we can gain insights into the overall user experience of an application (be it an actual product or service).
However, you need to understand what proportion of the end-users this information represents, and also the context in which these support tickets are generated and processed. For example, it is not always the case that a user will raise a support ticket due to a perceived problem with the application (they may simply abandon it and never come back). So you need to have some way of connecting this information to another point of reference to understand the data that you are looking at.
The first step in the analysis is in data preparation, but before that you need to plan your analysis around the type of questions that you want answered, and figure out whether the data that you plan to process will yield the results you need. For example, if you want to know about the type of issues that users experience but the information in the support ticket is not structured to provide clear categories for the type of issue raised then you will still have a lot of manual processing to do after the data preparation.
I think taking a first step in understanding the information available in the support ticket will help you work out the type of question you can answer, as well as additional information you will need to gather to help put the results in context.