Tweet

Knowing Is Better Than Guessing

Guest post by: John Cooney

Article Overview: Article Highlights * You don’t need to test more than 5 people at a time. * Test early and often. * Focus on specific flows.

Knowing Is Better Than Guessing

Usability experts have an annoying shared trait: they tend to answer a lot of questions by saying it depends. Each website’s unique content, audience, goals, and style ensure that there are no universal solutions to even the most basic questions. This means that there is only one way to know if your site really works for users: you have to test it. Fortunately, if conducted right, these tests shouldn’t require a lot of money or time.

When professionals begin designing a website, they typically follow a series of guidelines or design patterns that help ensure the result is easy to use. In the case of simple interactions, such as a log-in form, this method almost always results in a usable design.

However, when the interface becomes complicated or highly specific to an individual site, it’s unlikely that any existing guidelines or patterns will provide enough direction. At this point, the designer cannot—or, at least, should not—be sure that the interface will work: what they have produced is simply an educated guess at a usable design. Even worse, many inexperienced (and some experienced) designers will base their work on assumptions or personal preferences—either their own or their clients’. Now you can drop the educated and just call the design a guess.

Beyond the Guess Most major websites rely on user testing to determine if their features work for visitors. Although the specifics vary with each implementation, user-testing basics are generally the same. A representative sample of subjects are exposed to a website (or a portion of a website), data is collected based on the subjects’ interactions, and the results are analyzed to measure effectiveness.

Big sites like Amazon and Netflix collect data from huge numbers of users in a constant stream of specialized tests. If Amazon wants to test a feature with 5,000 new users, they only need to turn the feature on for 45 seconds and 5,000 new users will see it. Of course, big websites tend to have big budgets, and this type of ongoing testing isn’t cheap.

But while these tests generally produce a lot of great data, more data doesn’t always mean better results. Just recently Google tested its new social media tool Buzz with 20,000 people. Unfortunately, these 20,000 people were all Google employees—who have a specific set of shared experiences and web-usage patterns, to put it mildly—and this testing therefore failed to turn up a major usability flaw: Buzz automatically connects you to people you have emailed. The resulting and immediate target="_self">had Google publicly apologizing and rushing to make changes.

By any measure, Google’s user testing for Buzz was a massive failure. Think you could do better with five subjects and a couple days? You should.

Cheap, Fast and Good The idea that user testing can be done on the cheap Jakob Nielsen 1989 paper on discount user testing is often seen as the start of the low-cost usability movement, which also gets called guerilla usability.

The basic premise is simple: although larger studies are useful, discount testing produces valuable results and a far greater return on investment. (Predictably, Nielsen goes further and suggests that discount testing often produces better results than bigger studies because it encourages an iterative design process.) In fact, one of Nielsen’s more interesting findings was that while better studies produce better results, even poorly executed quick tests provide significant value.

At Orbit, we’ve distilled a variety of research and approaches into three main principles for guerilla usability.

1. You don’t need to test more than 5 people at a time. If your site (or feature) has a major flaw, the first subject will find it. So will the second. So will the 35th. You don’t need to watch people make the same mistakes over and over, and while five people won’t find every tiny problem, they will provide you with a lot of information about your site’s strengths and weaknesses. (Even testing a single person can be useful, especially since the obvious stuff is often the easiest to miss.) Similarly, while it’s best to recruit subjects who are representative of your actual audience, testing with friends, family, or anyone within reach will also produce worthwhile results.
2. Test early and often. If you wait until your site is complete to test it, any changes that need to be made will be expensive, time consuming, and complicated. But if you first test the flow of your main features by presenting a subject with hand-drawn wireframes (known as paper prototyping), changes require only a good eraser—or at worst, a new sheet of paper. There is no substitute for testing a live site, but running the major functions through a quick battery of paper prototyping will catch the biggest flaws before you start the expensive development process.
3. Focus on specific flows. Testing every feature on a large-scale site, even with a single subject, can take hours or days. (And that’s time for which these people will need to be compensated, once you find volunteers willing to endure such punishment.) After you finish a few of these extensive tests, you’ll be left with a mountain of data to sift through, and a potentially imposing array of changes to be made. But every feature on your site is not equal in importance. Start with the crucial functions—on an e-commerce site, this would probably be the checkout process—and save the less important features for later. If you do need to test a number of features, break up the work into several tests and conduct each just before the design or development of that specific function.

If conducted intelligently, discount usability can provide crucial data to improve your site’s design and greatly improve its effectiveness. Paper prototyping can produce results in a few hours of work, and tests on a live site can be completed and analyzed within a couple days.

These quick guerilla techniques won’t guarantee that your site is perfect, but they will ensure you catch the biggest problems. And that’s perhaps the most important principle of all: doing something is much better than doing nothing. Every user test you perform—however modest—takes you further beyond the guess and closer to the knowledge that your site works.

Related Articles
  Follow-up a Great Question with Dead Silence!
  LEADERS: DISCERN THEIR CLIENTS NEEDS
  “Market Research” Story
  Are Your Expectations Known to You?
  Deciding To Market Your Business Online
  RISKY BUSINESS
  Fear Free Selling
  Money Stress: Five Financial Mistakes to Avoid to Reduce Your Money Worries
  Five Financial Mistakes to Avoid to Reduce Your Money Worries
  So What
  Reading Between the Lines is for Dating
  QuickBooks Tip: Easily See the Accounting Entry Behind the Screen
  Play the Full Table
  Assessing Training Effectiveness
  Easy Keyword Forensics using Google Data
  Attractive People Inspire More Teamwork
  Lesson #2: Use Secrecy to Seduce Your Customers
  The Trouble With Taglines Why This Catchy Phrase Can Lead You Astray and What You Can Do About It
  The 9 Cs of Selling
  How to Write Better and Faster -- And Have More Fun Doing It!

Article Tags: chicago web design chicago website design, usability, website user testing
Referred by: http://www.orbitmedia.com



Related Forum Posts
Re: How do you maximize your time at business/networking events? - Hy Kevin. I think everyone is focused on those persons who can help business to grow but i consider it necessary to constantly meet new people and make new connections. Knowing more people, means more exposure for your business and you never know when you will need their services.
Affiilate Products - Knowing what you are trying to sell would be helpful. What sort of promotion have you tried -- in detail? I wouldn't jump into a big expensive ad. If the things you are doing isn't working on a smaller scale, they probably won't work on a big high dollar scale either. Why not give us your website address and we can take a look. There is plenty of money to be made in affiliate marketing, but it takes time, it takes trial and error and you have to evaluate what you're doing and make the needed changes to have an effective program. Shri
Re: Does it help to join business groups? - Different groups can have a very different feel to them, so it's good to try out several. I was in a lunch group for a while that was very formal, and a bit expensive as you had to purchase the event ticket and the lunch. It was a small group, and they knew each other well. They were nice people, but there was really no opportunity for us to help each other -- more of a social thing really. I've also done chamber of commerce, or women's entrepreneur events. I always have a good time, but I haven't found many business leads that way -- most of the newer entrepreneurs often don't have the financial resources to hire others (at first anyway). I did get one big job back when I was a freelance writer from one of these events, however. I've met more clients through regular socializing surprisingly! Knowing a person through friends can make others feel more comfortable it seems.
Re: Best Internet Marketing Strategies - [quote="christew":latsmlgi]Plus a niche is not a bunch of keywords, it is a demographic interested in a particular topic. You need to learn about the people in this demographic. Keyword research tells you their search habits and gives clues to competitiveness and potential profit, but you are looking at through a very narrow window of a much bigger picture.[/quote:latsmlgi] Good point. Knowing about both the group and the keywords they tend to use is of course the best combination, but if you thoroughly understand your niche market and your products, the content you write for your website and the articles you post elsewhere will almost certainly include natural "keyword phrases" even if you don't research the keywords - and if your niche is tightly defined, that will probably suffice. I am not saying you should ignore keyword research, but knowing your products and market is more important.
Starting a New Business - Congratulations on the first step of deciding to create your own business. First, you have to ascertain if you are going to be selling your products or services to people or to businesses. People purchase products and services for three basic reasons: •To satisfy basic needs •To solve problems •To make themselves feel good. You have to determine which of these three categories your product or service is the solution to. Businesses, unlike individuals, buy products and services for three reasons as well: •To increase revenues •To maintain the status quo •To decrease expenses. Knowing the reasons why people/businesses buy, you should have a picture emerging of who you think your "Target Market" (or ideal customer) is. Once you have zeroed in on your solution, you are ready to create your Marketing Strategy. If you look at any company that has grown rapidly over the past few years, (i.e. WestJet and Starbucks), it should demonstrate to you the importance of Marketing. In simple terms these companies have: •Clearly defined what makes them different; •Who their ideal customer is; •How to reach them; •Why customers should buy from them; •And most importantly why customers should come back again. It is strong Marketing Awareness that singles out the best companies. I can not stress enough how important it is to create a "living" Marketing Plan before you even start. I hope this helps a little in the creation of your plan. All the very best of luck. Remember: "A Goal without a Plan is just a Wish" Miles of Smiles, Marketing Maven


Share this article with your friends. Fund someone's dream.

Leave a comment below or share on the left and you'll help support entrepreneurs in Africa through our partnership with Kiva. Over $50,000 raised and counting - Please keep sharing! Learn more.

Newsletter

Get advice & tips from famous business
owners, new articles by entrepreneur
experts, my latest website updates, &
special sneak peaks at what's to come!

Name:
Email:

Popular Articles

Ten Reasons to take Notes during Sales Meetings

Attracting Passionate Employees

Using Social Media Marketing

Suggestions

Email us your ideas on how to make our
website more valuable! Thank you Sharon
from Toronto Salsa Lessons / Classes for
your suggestions to make the newsletter
look like the website and profile younger
entrepreneurs like Jennifer Lopez.