Usability key tasks and scenarios

When planning a Usability test you need concentrate on the key tasks users will need to complete when visiting your site.  In order, to explain these tasks you need to expand them into Scenario’s.

What are the most critical tasks on your site?  These are the tasks that users must be able to complete or your site will be a failure.  If you sell tickets to concerts, users must be able to search and find the tickets they want and be able to buy them.

A usability test usually lasts 50 minutes of which only 35 mins or so will be used for testing.  In this time you might cover one long test or may be even as many as ten.

What you should try and do is make sure you have extra tasks at hand if you find that you have time to spare.  For example, get them to try the same test on a competitor site.

Deciding what to test can be tricky but the best thing to do is write a list of all the key tasks that are critical to your site and rank them in order of importance.  You can also talk to client facing staff like customer support to see if there are any issues being raised with them.

Finally, take your key tasks and write scenarios.  For scenarios try and give the tsak context “you are…”, “you need to…” and keep them short and sweet.  For example:

Key Task

Apply for a car insurance quotes.


You’ve received your renewal notice from your current car insurance provider.  They’ve offered a good renewal price but you want to double check there isn’t a better deal for you.

Apply for a car insurance quote.





Usability test participants

When selecting users for usability studies you should refer to them as participants not subjects. This is because we are not testing them we are testing the website.

Recruit your target audience?
The natural assumption to make when selecting test subjects is that they need to match a profile of your sites target audience. But this isn’t as important as some may say. There are some circumstances when users may need to have “Domain Knowledge” because some tasks may require specialist knowledge to complete. However, the majority of usability issues relate to navigation, page layout and visual hierarchy. These are issues most people would encounter.

Do try to recruit close to your target audience but don’t get to hung up on it. It’s better to test than not. But use your judgement – if you need your audience you need your audience.

How many participants do you need?
The general rule of thumb is that 5 participants can reveal 85% of issues. This infers that there are diminishing returns on more subjects. But for DIY tests 3 participants would be sufficient. The key point to make though is that you should be aiming to carry out multiple tests through out the project. This way you can fix the original issues found and then identify new issues to fix as the project progresses.

How do you recruit participants?
If you’ve got he money you can hire someone to do this for you. You can approach companies that have focus group facilities to do this. They will usually be experienced in recruiting or will be able to recommend someone to you.

However, if you don’t have the budget to pay someone, you could:

  • look in the places that the type people you are looking for might congregate (user groups, places with an affinity to your audience)
  • ifnyou need people who actually use your site, put a link on your present website or introduce a pop-up.
  • if you need anyone, recruit friends, family and neighbours
  • if you require a specific type of user consider remote testing as this automitcally increases your geographic search area.

Once you’ve found them you need to advertise to them. You need tell them:

  • what you’re doing
  • when you’ll be doing it
  • where you’ll be doing it
  • who your looking for
  • contact email and name

now you just need to place the advertisement where you think your audience will be:

  • put it on bulletin boards in community centres/li>
  • post it on message boards
  • use your social networks – both professional and personal
  • put a pop on your website or ad on your homepage

when people reply you just need to screen them.

  • are they available on the day
  • do they match you target audience criteria
  • tell them what to expect
  • explain how they will be compensated for their time
  • make an appointment

Finally, confirm everything to them. Where and when. Directions, maps, parking instructions. Contact number. No disclosure agreement.

There is a chance that users won’t turn up on the day. The thing to do is to try and have a back-up plan. Either someone on stand-by to come in or test remotely. Failing that, grab a stranger off the street.

Usability What and When to Test?

The key to usbaility testing is to test as much as possible.  It doesn’t have to be expensive or time consuming.  Just a little bit of effort will reveal something usefull.

You can start testing as soon as you have the brief.  You dont need a protoype or a finished site.

Learn From Experience

Test your existing site. If you’re redesigning an existing site this is an obvious place to start.  You can identify what currently doesn’t work and know to address this in the main site.  Analytics from your existing site will be usefull to.  They can tell you about important pages and user behaviouir (although not exactly what causes their behaviour, but it’s a good place to start).

Test Other Peoples Sites. There are sites already out there serving a similar audience or content or with similiar interactions.  You can use these as prototypes and see how well they work at achieving similar goals and tasks.


Sketch out a wireframe. A simple drawing on a sheet of paper can be used as a starting point for research. You can ask anyone what they think your sketch is meant to be?  Listen to their answer.  Was it close, did they get it was a homepage about s pecific subject.  What did they think to the heading?  You can get a lot out of sketch.

Wireframes. Create a wireframe with clear headings and ask people how they would achieve certain tasks.  How would they search for something?  Where would they click to get to a piece of information?  Wireframes are concerned with establishing interactions.

Page Designs. After wireframes you’ll need to create some page designs of the key pages – pages templates that will appear through out the site.  These visual treatments are commonly reffered to as “comps”. Comps focus on testing visual design.



DIY Usability Testing Kit

In order to carry out a Qualitative test you’ll need the following kit:

1. Microphone

a. You’ll want to record the Test so you can review what was said at a later date.  This means you dont have to write notes as you carry out the test and can give it your full attention.

b. So others who are observing the test in another room can hear what’s being said.

2. Speakers

So the observers can hear whats being said.

3. Screen Recording Software

You’ll want to review where the user participant clicks on the screen in response to your tasks so can review at a later date.  Again this allows you to concentrate on administering the test.

Notable Screen Recording Software includes:

4. Screen Sharing Software

In addition to hearing whats happening in the test lab, your observers will want to see whats happening onscreen.  Screen Sharing software will allow this to happen.  Options include:

5. Video Camera (Optional)

Sometimes people want to record the User’s reactions but this isn’t necessary (unless you have a boss that needs convincing).

6. Incentives

You’ll probably need to incentivise your participants to take part.  Gifts or Money usually sorts this out.



Usability Testing Options

One of the most important aspects in web design and development is Usability.

Anyone working in web design and development will tell you that Usability testing is invaluable in helping to produce web sites that work.

Usability Testing falls in to two categories:

Quantative Tests

Tests where you want to prove something.  For example, is the new search engine better than the last? We can look at the accuracy of the results from old and new search engines to see which is the best.

This is a more scientific approach.  They have to be well planned and carried out in lab environments to ensure the results are sound.

Qualitative Tests

Qualitative tests are more about observing users and how they interact with your site and asking them to talk aloud.  In this way you can understand what they are thinking and gain insights that enable you to improve what you’re bulding.