Traditionally, usability tests have been moderated.

A moderated test just means that there's a moderator present asking

questions in order to deeply understand each choice the participant makes.

They may be sitting in your office and using your prototype, or you may go out

and observe someone test your product whatever you intend for it to be useful.

Alternatively, moderated testing can be done remotely.

There are many remote testing services that allow you to stay in the office

while observing and holding discussions

with people using your service from the comfort of their own home.

So what are the advantages of a moderated study?

Probe deeper.

When you're with a participant at each step,

you can ask what they're thinking and why they've made a particular choice.

For example, why did you click here, or did you see this instructional text?

Capture participant reactions.

Did you notice that sigh or look of confusion?

You can add that to your notes as additional data.

Without someone there, nonverbal reactions can easily be missed.

[SOUND] Protect sensitive information.

Say you're working on a product that really shouldn't

be revealed to the public until much, much later.

By moderating, particularly in person.

You've limited any exposure to the highly vetted group of study participants,

who've all ready signed an non-disclosure agreement, or NDA.

[SOUND] Evaluate low fidelity concepts.

If testing early and possibly even with just a paper prototype, you will need

to be present to guide the participant on what features are functional.

[SOUND] Keep on track.

If you notice the participant is focusing somewhere irrelevant to

the study's objectives, you could politely guide them back to the task.

Test complex products.

When testing products such as business software, you will need to make sure

that your participant has the background to complete the required test.

And in addition,

you will likely need to provide clarifying information throughout the study.

Let's say that your study is the perfect candidate to be moderated.

For example, you have a click through prototype for

your complex business software.

One next question is, will you be running this study at your office,

in the wild, or remotely?

A moderated study needs to be in real time but

you do not need to be in the same physical space as your participants.

If you do choose to bring the participants into your office, you will need to make

sure your office is set up in a way that will make people comfortable and

limit distractions.

Ideally, this would include a dedicated space in the form of usability lab.

Let's take a look at the key attributes of a usability lab.

First, you need a room with a closed door to minimize distractions during the study.

Inside the room, you'll need a computer with a monitor.

An external monitor rather than just a laptop is preferred, so

that it's easier for you and the participant to see the screen.

Attach a camera to the monitor if it doesn't have one already,

to record the study.

And confirm that the microphone in your computer is working.

Make sure there's a keyboard and standard mouse available as well.

There should be at least two chairs in the room.

Place the chair for the participant directly in front of the monitor.

And place your own chair slightly behind, and next to the participant's chair.

You can also include an additional chair for a note taker towards the back.

The key component of usability study is also being able to invite outside

observers.

You can do this in a couple of ways.

One is to have a couch or row of seats in the back.

However, I believe this can be intimidating for the participant and

prefer to put the observers in a totally separate conference room.

And stream the study into that room.

The observers would then be able to hear the audio, see the product being used,

and if possible also see the participants facial expressions.

Make sure to set up a messaging system between the usability lab and

the observation room, so that the observers can text, email, or

chat their questions throughout.

You may need specialized software to record, observe and

analyze the final videos.

You can find the list of options in the teacher's notes.

Remember to compensate your participants generously.

They're taking time out of their day to commute to you and

to provide you with the valuable feedback that you need.

The exact costs will vary by location, but a rough guideline,

if you're in a metro area and asking people to come in for about an hour,

anticipate providing $100 per person.

Testing a product in person is great for keeping things confidential, and for

testing physical hardware when you space design.

However, you will be limited to the people living close to your office.

Now, imagine designing a product meant to be used by everyone in the world,

as I did there in my How representative do you think our research

would have been if we tested with just people in the US, or California?

Or those in Silicon Valley willing to commute to our office in Mountain View?

We'd be left with a group of early adopters, tech enthusiasts, and

frequent usability testers.

Not exactly diverse representative group.

Now, let's go back to the usability lab setup we just described, and

adjust it for a remote testing study.

Remove an extra chair and take a seat in front of the monitor yourself.

You can use remote testing software to allow the participant to take control of

your screen, and use the prototype available in much the same way as before.

Alternatively, you can take an available conference room and

run the session straight from there.

If you include outside observers, they can then join you in the same room or

from their own desks if the session's being streamed.

Remote testing is rarely seeded for

hardware products, paper prototypes or highly confidential material.

Mobile devices can also be tricky,

depending on the kind of hand gestures are important, or if an install is required.

However, remote testing is a fantastic solution for

the online products you're most likely to be working on.

Remote testing is convenient.

You don't need to leave the office, and it provides access to a much,

much wider range of participants.

While working on advertising software at Google,

we relied heavily on moderated remote testing.

Because of this,

we were able to test frequently as we developed new features and products.

However, it was also important to occasionally visit our users in their

workspaces to understand the greater context in which our tools are being used.

In one site visit, we're able to discover that clients would use

an additional monitor that displayed other software,

such as spreadsheets to make up the features that our product was missing.

A very insightful observation that would have been impossible to discover remotely,

or in a lab.