The Fifth week

Here's what we're going today

  1. Great news!
  2. Midterms
  3. Theory
    1. How to write a usability report
    2. A game :)
    3. Evangelism
    4. Workflow
  4. Praxis
    1. OCAS
    2. Your weekly reading

Great news!

Timbits
We're getting paid (in food) to do that test we talked about last week.

Midterm stuff people struggled with

  1. On any project, there are client requirements, and there are team standards. In the case of the midterm, the client requirements were my instructions. The team standards are Humber's rules against plagiarism (among other things).
  2. The entire first paragraph of the assignment.
  3. Scope: don't just factor in all the cool stuff that <em>might</em> be useful - take a look at the resources you have available and scope for what you can deliver. There's nothing wrong with a narrow scope that provides really crisp data.
  4. Some people will probably want to change their scope. Or maybe you're just not happy with your grade. You've got a week to resubmit.

Also, please have all your first-term reading assignments in by this time next week. That's the cut-off.

How to write a usability report

  1. Compile
  2. Analyze
  3. Make it readable

Compile

So, you have your data. The first step is to compile it so you know what you're working with.

Most of you have chosen to collect multiple datasets in parallel - you'll want to compile that separately (even though it was collected during the same test).

If you do this step diligently, you'll save time in the long run. Make it easy to read and understand.

Analyze

If you're doing this step with the help of some fellow experts (recommended), it's time to look at what your data actually tells you.

If you have a good sample size, and tight controls, it will be pretty apparent.

On the other hand, if your sample size is too small, or your variables too... variable, it will be hard to draw conclusions with confidence.

Even a large sample size can create the illusion of fact

Strong corellation between divorce rate in Maine and per capita margarine consumption

Listen, you're not doing hard science. But understand that people will ask you hard questions about why you're coming to your conclusions. And if your conclusions are wrong, you'll have to answer for that.

People aren't looking to you for maybes - maybe it was the browser, maybe it was the device, maybe that user was really tired - they want answers. Specific answers. 'If you change the order of your menu items, conversions will go up 10-12%' answers. And they want those answers to turn out to be right.

Be useful

  1. Be specific
  2. It's not the user's fault
  3. Don't recommend making small changes to a big problem
  4. Don't redesign for them
  5. Align with the client's priorities
  6. Help the client prioritize

Make it readable

A usability report is, at it's best, a plan for action. Make the case for your recommendations.

Make it readable

  1. What we did
  2. Why we did it
  3. How we did it
  4. What we learned
  5. What should be done
  6. All the boring data that backs this up

Another possible format:

Via https://uxdesigntemplates.com/resource/user-testing-report-template

  1. Cover sheet
  2. Test outline
  3. Test participant details
  4. Executive summary
  5. Bugs list
  6. Sheets for key findings (including usability problems and positives)
  7. Post-test question responses
  8. Full list of recordings

Being informal doesn't mean being lazy. It's actually more difficult - you need to explain thoroughly, but in simple terms, and in a way that's easy to synthesize.

You're telling people how best to spend their money. Think about what they care about - what works, quick wins, and big wins.

That being said, if you want a jumping-off point for a more formal documentation format, here's a couple:

A game :)

  1. In an even number of groups of 3-4...
  2. ...you are paired with another group.
  3. Create a scorecard for your group with 4 columns - month, price, profit, and total profit.
  4. Subdivide the latter 3 columns into subcolumns, each labelled alternately 'us' and 'them'

You are competing businesses. Each month you set the price on your product. Your costs are absolutely fixed. Bankruptcy is impossible. Your profits are driven entirely by market competition.

Each month, you must write down 6 figures: what price you chose, what price your competitor chose, what profit you made, what profit they made, what total profit you made, and what total profit they made.

The goal of the game is to maximize your company's profit.

Price/Profit

Your priceTheir priceYour profitTheir profit
$30$30110110
$30$2020180
$30$1020150
$20$3018020
$20$208080
$20$1030150
$10$3015020
$10$2015030
$10$105050

  1. Who won?
  2. What was your strategy?
  3. Assumptions did you make?
  4. What would you do differently?

Evangelism

You know a lot already. It won't be long until you're working with someone who knows less than you. They will probably be your boss.

You'll learn that when you're in a room full of developers, you'll argue differently.

You will know how to solve problems. You will know the cutting edge of technology. You will feel the inefficiencies start to weigh on you.

What will you do when you're the only person who knows how to make a website accessible?

What will you do when you're the only person who cares about responsive design? Or load times? Or security?

You need to make it so your team understands

You need to make it so your boss doesn't need to understand.

When your team can improve

If there are improvements you can make within your team without additional resources, present it as an opportunity to make your lives easier, and/or grow your skillset.

But remember, anything that starts with 'If everyone would just...' isn't a solution.

Peer review, linting, automation. You need to get your team to agree to be held accountable.

When you need resources

  1. Get team buy-in
  2. Explain the cost of the issue
  3. Explain the cost of 2 options, both of which you are ready and willing to implement

OCAS

Time to start! ... or is it?

Praxis: Listening to your reading

Your reading this week is https://webaim.org/projects/screenreadersurvey7/

If you have headphones with you, please read it with either NVDA or >VoiceOver. Or choose another!