On May 7, I was fortunate to attend a workshop led by Daniel Eizans on mental modeling. I learned several useful techniques. These are my notes, and you can see Daniel’s slides and downloads on his site too.
Part 1: Contextual Inquiry
Contextual inquiry: a structured form of interviewing that allows the research team to:
- Gather interview data – interview them doing the tasks you want insight into. You are asking people to be an expert when they may not be.
- Understand ethnographic insights
- Capture physical behaviors, feedback, and work patterns – participant must build trust in interview over time
The data gathered helps inform everything from personas, to product design to workflow/workplace modification to content strategy.
This technique is influenced by designer, user, and strategist. You observe user behavior and have them explain as they do it.
Conducting interviews – needs
Keys to a successful interview:
- An interviewer for formal questioning
- An observer for recording physical behaviors, workflow models and note taking
- Trust between interviewer and subject
The interview is about building a relationship. Even if we feel we have the right answers our effort should focus on developing a shared interpretation of an activity or work.
Users may take a different process depending what they think they should be saying. You need to make them feel like an expert but also probe them along the way.
Conducting interviews – Goals
- Watching and probing observers as they work and inquire to understand motivations and strategies
- Learn from the user perspective even if you feel you know the answers
- Look for gaps in what a user is leaving out and ask questions to fill them
- Gather both concrete and abstract data to have them focus on actual artifacts and actions
10-15 interviews (one hour minimum, less than two hours per interview) for 3-4 personas. At least 2 hours of analysis for each interview.
The interviewer role
In contextual inquiry, interviewers must:
- Facilitate conversation and behavior
- Make the participant feel like an expert in whatever they’re doing
- Ensure that the interview environment is contextually relevant to the research or tasks to be addressed
- Ensure the participant articulates the processes and methods for work as he/she is doing it
- Be dumb whenever you can.
The observer role
- Document physical behaviors
- Understand how the environment/context affects the participant’s ability to complete a task or perform work
- Note patterns that emerge from 1 interview to the next
- Take photos or video
The participant role
- Doing and answering
- Showing the observer and interviewer complete tasks
- Avoiding assuming that the interviewers will understand why something is being done
- Talk to the researchers’ level of understanding (avoid jargon)
Conducting the inquiry
- Don’t use conference rooms, Skype, or usability lab. Do it on their territory.
- Send an email to the participant.
- Note ambient factors (lighting, temp, time of day, layouts, culture, artifacts present)
- Start with a short questionnaire
- Note that you’ll be doing something different and set them to work
- Ask questions during the task, keep the interview free-flowing
- Ask specific questions relating to their thought process, task, and ensure they feel like they’re the expert
- Work up to an hour, be ready to move locations if necessary (work station change, environmental stimulus requirement)
- Ask, ask again
- Spend 2 hours on analysis for every hour of inquiry. Talk to the design team, others on the project. Collaborate.
Prepare for interpretation
- Avoid open-ended questions – be specific to the task
- Balance between probing for a pattern and allowing participant to work freely
- Cycle between observing, probing, and interpreting
- Be flexible but consistent to develop trust
Interpreting Contextual Inquiry
Interpretation is crucial for persona development and content planning.
Fact > Hypothesis > Implication > Idea
Just as crucial as the interview itself. Do not jump from data into how to use it.
Bring a cross-functional team together to hear the whole story of an interview and capture the insights and learning relevant to the problem.
- With 48 hours of the interview
- Moderated by an interviewer who discusses each interview for the team
- Assign roles: moderator (interviewer), work modeler, recorder, task master
Valuable for the transfer of knowledge that interviewers and observers learned.
- Listen and participate in conversation
- Ask questions
- Draw work models
- Record issues, interpretations, and ideas based on the interviews
Models for observing work/tasks
- Show actors and their interactions during a process
- Map all events that take place during work, process, or task
- Not responsible for showing the sequence of events during work process
- Document steps taken to perform a task
- Build from data captured in a flow model or produced exclusively
- Can include multiple actors if necessary
- Provide a view of all artifacts, tools used in process
- Typically forms of information, screenshots of how people use digital info
- Crucial for placemaking, in that they provide understanding where the digital and physical worlds should intersect
- Document how environments and actors in a given place influence the work being done
- Impacted by things like socioeconomic status, open vs traditional office environment
- Vary significantly depending on participant background
- Show work environment in which work or tasks occur
- Highlight physical structures, layouts and materials that take part in task completion
- Vary greatly with participant and ambient factors
Consolidation of the work
Materials: sticky notes/cards, marking pens, large work surface
- Record each idea or pattern with a marking pen onto a separate note/card. Randomly spread notes on the work surface so everyone can see them.
- With minimal or no talking, identify relationships and group notes accordingly. You can move something someone has moved. If something belongs in 2 groups, make another note.
- Discuss the patterns as a group. Why did things move or not? Create categories or headings for each group. Identify notes that capture the meaning of the groups.
- Combine groups into larger categories or supergroups if appropriate.
- Map interview data to the affinity diagrams if you can. Establish patterns among similar participants and begin to note where personas are emerging.
Representation of the thought and actionable processes used to achieve a set of goals in a narrowly defined scope.
One of the most important parts of a communication-focused mental model is understanding the engagement that you want to influence with content.
This gives your content a reason for being and something to measure against.
Dig out inventories and plug them into your model.
From inquiry: mapping intentions
Intentions come directly from your personas, which are based on your inquiry.
Intentions relate directly to the engagement.
Tasks: new content builds upon intentions and serves as the towers for our mental models.
Frame tasks as questions to use the persona’s frame of reference to talk about a content need.
When we focus on tasks we are probably making content that is usable for humans and search engines.