Who to Interview
How many customers should I test with? (2, pg 144)
- Focus groups lead to participants not being fully honest
- Interview between 5 and 8 is a good balance
How to recruit in target market? (2, pg 148)
- Refer to personas created to help with writing screener questions
- Ensure to cover all of the segmentations you used forming your target market
- Use conferences/ networking events
Starbucks User Testing (2, pg 151)
- Low cost and immediate
- Harder to screen for the customers that you actually want (demographic)
- “Hi sir, do you have 10 minutes to share your feed back on a new website in exchange for a £20 Starbucks card?
- More success found in shopping centres than Starbucks as people tend to have more time to kill
How to Interview
In-person, remote, and unmoderated user testing (2, pg 145)
- Moderated tests have the customer perform the test on their own, but recorded for the moderator to view afterwards
- In-person allows richer data collection (body language)
- Remote – be aware of technology issues of screen sharing
- Unmoderated can product a quick turn around as providers already have users lined up ready
Question Guidance
Open vs Closed Questions (2, pg 158)
- Open usually begin with why, how and what
- Closed questions limit the customer’s possible responses (e.g. to yes or no)
- Tip to write the questions down in advance to check they are open
Non-leading Questions (2, pg 158 – 160)
- Be careful not to tell the user how you would like them to respond, e.g. ‘how would you prefer it to be sorted? By date?’
- Be ok with long pauses and don’t help them as it will invalidate the test. Real users won’t have you to help them.
- Don’t predict what the customer will say out loud and try to recede into the background to get the most honest feedback
Measuring Importance and Satisfaction (2, pg 59)
- A bipolar scale goes from negative to positive (i.e. satisfaction)
- A unipolar scale is a matter of degree (i.e. importance)
- More than 10 choices is overwhelming
- Fewer that 5 will lack granularity
Unipolar Example
- Not at all important
- Slightly important
- Moderately important
- Very important
- Extremely important
Bipolar example
- Completely dissatisfied
- Mostly dissatisfied
- Somewhat dissatisfied
- Neither satisfied nor dissatisfied
- Somewhat satisfied
- Mostly satisfied
- Completely satisfied
Product Market Fit Question (Sean Ellis) (2, pg 231)
- Do not invest in trying to grow a business until after you have achieved product-market fit
- This questions helps you assess if you have achieved product-market fit
- Survey users and ask:
- “How would you feel if you could no longer use product X?
- Very disappointed
- Somewhat disappointed
- Not disappointed (it isn’t really that useful)
- N/A – I no longer user (product X)”
- “How would you feel if you could no longer use product X?
Follow-up Questions (2, pg 157)
- ‘Echoing back’ is a useful technique – e.g. ‘I see you clicked that button. Could you tell me why?’
- If user asks question, ask another question rather than replying with yes or no
The Data
Make sense of the data (1, pg 103)
- Look for patterns
- Place outliers in a parking lot (outliers could form their own pattern with more data collection so don’t bin them)
- Verify with other sources (do people inside the office agree/ customer support staff seeing the same trend) If not your sources could have been skewed
Feedback Table (2, pg 162)
- Group into feature set, UX design, and messaging
- Ask valuable and easy to use questions
- Get overall scores to compare to next wave when improvements have been made
Regular Cadence
TIP – randomly schedule users on a routine basis so that you always have some ready to test with, rather than panicking (2, pg 150)
Continuous Learning in the Lab: Three, Twelve, One (1, pg 99)
- Three users, by Twelve noon, once a week
- The regular cadence ensures that you ‘test what you’ve got’ so that there is no deadline pressure
| MONDAY | TUESDAY | WEDNESDAY | THURSDAY | FRIDAY |
| Start with the recruiting process Decide what will be tested | Refine what will be tested | Refine what will be tested Write the test script Finalise recruiting | Testing day Review findings with the entire team | Plan the next steps based on findings |
References
- Lean UX by Jeff Gothelf
- Lean Product Playbook by Dan Olsen
One thought on “User Interviews”