Job Shadowing

One thing that most colleagues know me for, is that I advocate for Job Shadowing; I mean this is the one constant thing I insist on doing and teach everyone about.

It all started around the year 2000 when I came across a paper from Karen Holtzblatt about Contextual Inquiry, the official name of this technique. It blew me away! For the longest time I had been searching for the reason why software projects are late and over budget. And most people who study this have the same conclusion: get the requirements right, but most of all, get the right requirements.

Contextual inquiry, or the name that I find the easiest to explain to other people "Job Shadowing", is so simple, it's ridiculous. You can summarize it in two concepts:

• Real work, real people.

• Master-Apprentice relationship.

That is all you need to do to get the right requirements.

The first step in the Design Thinking process is Empathize and there are a few techniques to do this, but the most effective, by far, is Job Shadowing.

Real Work, Real People

You must observe real people doing real work. You need to go where work is being done by the people who actually do the work. This is in contrast to interviewing users in a meeting room or on a video call.

There is a ton of psychological and anthropological reasons for this. To start, we tend to summarize and filter out a lot of information when asked "how do you do your work" because we really cannot remember all the details and because we create a story that rationalizes why we do the things we do. So when asked how do you do your work? or how do you choose between two pairs of jeans we really don't know how we do it. Only when we observe people and ask questions on the spot then we have a chance to get to the truth. There are some ways to get around cognitive blind spots, like in the example of choosing between two jeans. You can ask "tell me about the last time you bought a pair of jeans"; and this is another rule of thumb "keep your subject focused on actual work or at least real examples of work; so as soon as people start generalizing you need to interject with "show me a previous document about what your are talking about", "tell me about the last time you did X".

When people are doing actual work, the contextual surroundings help them remember many details and exception that otherwise would go unoticed during an interview. And this is another source of information; their desk or work area. Ask about the postits on the monitor, the piece of paper with a scribbled note, the color pens they carry in ther pocket, what's in your backpack?, etc. Most people hack their way around their jobs. They create makeshift tools to help with their work. Also pay attention to their interactions with other people; you are looking for flows of information; what piece of information is being transacted at that moment?

Real People

For some weird reason when User Researchers ask company representatives for a person who can show us how a particular task is done, they choose a manager of the people who actually do the work. Don't get me wrong, managers know how to do the work, most of them at least, but they haven't been doing it for a while now. However I usually use this opportunity to understand the high level flow of a task, so this is a useful source of information. Right after that, I always insist on observing people who do the actual work.

Mass market products

For mass market products this is a bit easier because the customer is the same as the user, and you can go to places where a product is used or where a task is done, and ask questions on the spot.

A while back (2000 - 2012) I used to work for Nokia and Motorola, designing mobile applications mostly around pictures and videos, taking and sharing, music listening, and phone call handling. I also did some work for Nokia on social networks. My go to places where malls, parks and just the street aroud the office. I would wait for a person to take a picture and then I would introduce my self and ask them questions about the picture. What made them take that picture? What do you do with your pictures? How do you share them and to whom?

New Products or Markets

New products solve existing problems, or needs or desires. So you can go to places where people have these problems, or express these needs or desires, and observe them and ask questions.

Master - Apprentice Relashionship

How do you get people to tell you the truth? Put them in a position of power by establishing a Master-Apprentice Relashionship. You, the user researcher, are the apprentice, and your research subject is the master. The master is the person who knows how things are done, he or she is the source of truth. The apprentice is there to learn, ask questions, and be inquisitive but in the spirit of learning.

When you are with a person that you want to learn from, you spell your incantation "Could you show me how you do X?". Magic happens after this incantation. This simple question establishes that your research subject is the master, he or she knows how things are done. Additionally you are saying that you want to learn and understand.

When you spell this incantation in the place of work of your subject, they are in familiar settings with the colleagues they already know and trust. So there is nothing to fear. You, the researcher, are the only person at odds, and with practice you get used to it and even comfortable with it.

Another way of introducing this relashionship is by asking them "you can think of me as a trainee, and treat me like one". Sometimes you get to do part of their work and you learn so much; there is nothing like feeling the real pain of your subject.

Side Note - Usability Testing

Even during usability testing sessions when many users get a real fear of being observed and judged, you can get them to feel at ease by focusing on testing the design, and focus all your questions at the design. The user is always right, when they make a "mistake", you want to ask what part of the design confused them or lead them on the wrong path; the design is wrong and the user is right. Word.

Data, Information and Knowledge

So which data are you recording on your notes? The raw data that I pay attention to is "What do they say?" and "What do they do?"; they are not the same. Many people forget why they do certain things and they do them automatically. Also, they may say the do something but when you observe them they do something else. They are not lying, they just have a story in their minds about what they do and that's the story they tell you. I know, it's weird how the mind works.

The next level up that I ask questions about is "why?"; why are they doing X? What is their goal. Most people break down work or activities into blocks of work, and each block has a goal, so I ask about these goals. This collection of goals become my guiding light when designing a solution.

At the top of everything is the main goal, what are they trying to accomplish with all of this work? What is the desire or need?

I use all of this knowledge during ideation to think about solutions. In the Design Thinking process you are supposed to ask "How might we do X?", where X is one of these user goals you uncovered during user research. But one of the most intersting questions that I like asking is "what if you didn't need to do X?". I also like asking "what are they really trying to do?". I like the Jobs to be done technique but it kind of boxes you into a limited space of solutions; it pushes you into creating tools to do the same work people are already doing. So I prefer to think "is there a better way of doing this or not doing it at all?"

Pictures

Take lots of pictures. Picture help you remember what happened doing your sessions with your users. They are an excellent tool to transmit empathy to your team. There is nothing like showing your colleagues actual people doing actual work, using your software or doing the work you want to improve with your product.

Video is great too for very specific situations. When you want to record a very specific problem or case. But you can only do it after your subject is comfortable with you, otherwise they start performing for the camera and your source of truth is gone.

Continuously sharing stories from your user research with your team will create an "Informed Intuition". All your teammates are continuously making decisions about the product and they will have a better chance of making the right decisions if they know their users at a deep level.

How many persons?

The magic number is 5. You can get the job done with 3; but this is the dirt minimum. When you do user research you quickly see that after 3 subjects you start seeing a lot of repeated data; even from your second subject 60% of the information is the same. This is another weird thing in society. Most people feel uncomfortable saying that you only need 3-5 people; the closest I have seen in conferences is 8-10. I think people confuse answering questions about what people do and why, with saying that something is statistically significant.

Who do we observe?

What kind of people do you need to work with? In some markets you don't have much choice so I take what I can get. But in cases where there are plenty to choose from I look for extremes; the beginers and the experts, republicans and democrats, ocassional users and habitual users, the tall and the short, etc. My goal is to get as much variety of information as possible.

Continuous Discovery

User research is a continuous activity. You need to keep a steady stream of user research sessions; at least twice a week. And you need to keep a roster of users that you can ask them questions, feedback, etc.

Eventhough you need to get feedback from multiple types of people, there is only a small pool who are enthusiatic about helping you. You can easily identify them by looking for people who are already hacking a solution; their pain is so great that they already have something held up with nails and duct tape.

The innovators and visionaries will happily do the work for free. The innovators will even call you when they have new ideas. The pragmatists and conservatives you need to pay them for their time. The skeptics are just not interested.

Summary

• Real work, real people

• Master-apprentice relationship

• What do they do?

• What do they say?

• Why? What are their goals?

• Take pictures

• Share the stories with your team

• 3-5 persons

• Innovators are already hacking a solution

• Continuous discovery