in columns to the right. This process allows you to filter easily and sort by category later. Sorting by a particular category makes it super easy to eyeball the data for that issue and understand what happened. Set up the spreadsheet to automatically tally up frequency counts as you analyze (see Figure 1.3). The resource page for experience sampling on the book’s companion website at leanresearch.co includes a template spreadsheet for you to use during the analysis step.
FIGURE 1.3 An experience sampling spreadsheet set for tallying up categories.
STEP 10: Generate bar charts.
As soon as you are finished classifying the answers, merge all of the classification data into one long spreadsheet. Create tables in which you calculate the number of times different category values happened and their percentages (see Table 1.3).
TABLE 1.3 SAMPLE EXPERIENCE SAMPLING TABULATED DATA
| Location Category | Count | Percentage |
| Home | 274 | 18% |
| Way to store | 12 | 1% |
| In car | 46 | 3% |
| At parking lot | 32 | 2% |
| At store | 1,058 | 70% |
| Way from store | 97 | 6% |
| Total | 1,519 | 100% |
You can further break down the data by “user type” in columns to the right. For example, you can use men vs. women, younger vs. older, or whatever other user types have been identified/recruited.
You can then easily produce bar charts to indicate what’s happening most frequently. These charts show how many times certain values occurred for any particular variable in any given category. For example, you can create a bar chart for the Location category (Figure 1.4). You can also create bar charts that look deeper into one variable across a certain category. For instance, what are the grocery shopping issues that occurred at people’s homes? (See Figure 1.5.) These bar charts will tell you the story of the data you collected in numbers.
FIGURE 1.4 A bar chart for the Location category.
FIGURE 1.5 A bar chart that crosses a variable and a category—grocery shopping issues at home.
STEP 11: Eyeball the data and identify themes.
Another way to get a good grasp of experience sampling data is eyeballing. Eyeballing means you simply read the answers to the experience sampling question and get a feel for what answers are like and what categories are out there. Without any analysis, you’ll be able to reach conclusions about what you found.
Glance over the sample list of answers in Figure 1.6. What can you say about what you see there? You can see several categories bubbling up here very quickly (Figure 1.7). Some people are taking notes for creating lists, some are writing down ideas, and others are sketching stuff. What you just did was eyeball the data.
FIGURE 1.6 Sample experience sampling raw data.
FIGURE 1.7 Categories bubble up quickly when eyeballing the data.
After you analyzed the data, it’s time to synthesize it into themes and an answer (or answers) to the question you started with, which is “What do people need?” Have a look at the information you gathered about frequency. What questions come to mind when you look at it? Is there a certain category you need to dig into more? Why? List to yourself the big, emerging themes that came out of the data. What insights do they provide about user needs? Are there any features you think might support things you discovered in the data? If you work in a team, it’s best to complete this step together. Different team members will reach different conclusions from one another. Have those conversations to understand better what the data tells you.
Once you have reached conclusions and answered your research question (“What do people need?”), consider developing a product concept and have potential users react to it. More on how to do that, in the next chapters.
A theme is an answer to the research question. Each theme has a title, a one- to two-paragraph description, and a design implication. Let’s look at the following example.
Men Are Lost in Aisles
Men have trouble finding items at grocery stores. They perceive the time they spend looking for items to be unacceptably long. They have difficulties identifying items they need to purchase and sometimes just wander, completely lost and helpless, in the aisles looking for what they need without asking for help.
Design Implication: Solve the problem either prior to getting to the store or at the store. Bypass the problem by repurposing past grocery lists or purchased items and provide personalized, in-store navigation guidance.
If a pattern starts to emerge during the research, you could follow up with participants to probe deeper. Without knowing the root cause, it could be easy to prescribe the wrong solution.
For example, if the issue is that they don’t know what the grocery items look like, then product navigation might also need to show images of items on the shopping list.
Other Methods to Answer the Question
While experience sampling is a fast, effective way for answering the “What do people need?” question, the following are three additional methods for answering it. Ideally,