Research is for decisions
The Research module starts here. In the Foundations module you wrote questions in your measurement brief. Now you’ll answer them.
The most common research mistake is starting with a method: “Let’s do some interviews.” The better start is a decision: “Next month we decide whether to redesign checkout or fix product pages first.” Research exists to make that decision better. In WXM, this is the heart of the Wonder phase: ask better questions before you touch anything (Wavesy).
Step 1: From decision to questions
Write the decision, then the two or three questions that would change it.
Decision: Fix checkout or product pages first? Q1: Where in the journey do we lose the most valuable visitors? Q2: What stops mobile shoppers from finishing checkout? Q3: Do product pages answer the questions shoppers ask before buying?
If an answer wouldn’t change the decision, drop the question.
Step 2: Match each question to a method
| Question type | Good methods | Gives you |
|---|---|---|
| Where and how many? | Funnel analysis, segments | Size of the problem |
| What do people do? | Session recordings, heatmaps | Behaviour, hesitation, errors |
| Why do they do it? | Exit or on-page surveys, interviews | Reasons in their words |
| Can people use it? | Usability testing (5 or so people per round) | Specific usability problems |
| What do customers already tell us? | Reviews, support tickets, sales calls | Objections and language |
Use at least one “what” method and one “why” method for each important question. Numbers without reasons lead to guesses; reasons without numbers lead to fixing rare problems.
Step 3: Keep it finishable
- Two weeks, not two months. A small study that finishes beats a big one that doesn’t.
- Set numbers up front: “30 recordings of the shipping step, 100 survey answers, 5 usability sessions.”
- Decide how you’ll write it up: a short list of findings, each with its evidence and source.
- Keep surveys short. Long surveys and intrusive pop-ups annoy visitors and get fewer, rushed answers. One or two questions is often enough.
Recap
Start from a decision, write questions that would change it, pick a “what” and a “why” method for each, and keep the plan small enough to finish in two weeks. The next lessons in this module show each method in depth, starting with user interviews.
For the full picture of methods, read CRO research.
Practical exercise
Write the one decision your team has to make next month about your website. Turn it into two research questions, pick a method for each from the table, and write how many participants or responses you'll aim for.