# Needfinding¶

Week 1 - September 27, 2016

## Lecture’s Learning Objective¶

to be able to discover user needs using participant observations and


## Definition¶

Finding potential user needs

## Why needfind?¶

Make something people want
- Y Combinator's Motto


## Example Question¶

### Attempt 1: “What do you need?”¶

• Positive: It’s an open ended question

### Attempt 2: “What problems do you have with [X]?”¶

• Primes the participant to think about [X]
• Reveals attitudes of incompetence

### Attempt 3: “What can I build for you all that would fulfill your needs?”¶

• Jumping to solutions (Negative)
• This is an AntiPattern

## Tactics¶

### Participant Observation¶

You can observe a lot just by watching
- Yogi Berra


Shadow the participant/client in their normal day of work to identify possible areas of improvement.

This approach is often more fruitful than asking the questions above.

#### What should we strive to learn by participant observation?¶

• Tacit (Unspoken) Knowledge
• e.g., truck drivers have big fingers
• “Deep Hanging Out” (Genevieve Bell, Intel)
• learning my immersing yourself in the client’s environment
##### A few questions to answer¶
1. What do people do now?
2. What values and goals do people have?
3. How are these particular activities embedded in a larger ecology
4. Similarities and difference across people
5. ...and other types of context, like time of day

#### Observation Technique: Be an Apprentice¶

• Set up a partnership with the people to be observed
• Play the role of a naive novice – don’t interject with your “expert” opinions.

### Interviews¶

Scheduling interviews facilitate depth

• Allows the participant to think about their needs and the topic of the interview

Planning is invaluable

• Don’t go in with a

#### Create a “Field Guide” (Steve Portugal’s insights)¶

• Introduction and participant background
• Main body
• Projection/dream questions
• “Blueskying”
• Wrap Up

#### Choosing Participants¶

• Representative of target users
• May be current users of a similar system
##### Approximate if necessary¶
• May not be ideal, but better than nothing
• e.g., getting computer science students when developers are unable to be participants

#### Conducting an interview¶

• Introduce yourself, explain your purpose
• The interview is about them, not you!
• Begin with open-ended questions

#### What are bad questions?¶

• “Is the daily update an important feature to you?”
• Is not an open-ended question - lend itself to yes/no
• “Would you like stores with less clutter?”
• Has an obvious answer
• Participant behavior might differ from their answer
• “What would you like in a tool for [X]?”
• Getting to a feature request
• Open-ended

#### Other types of questions to avoid¶

• What they would do/like/want in hypothetical scenarios
• How often they do things
• How much they like things on an absolute scale

#### Photos are powerful reminders¶

{Photos + notepad} can approximate fidelity of audio/video if done well.

#### What are the gems?¶

• You’ve uncovered a surprise or found what is missing
• You can explain why people do unusual things
• You want to tell others about what you have learned