Usability for Novel vs. Routine Tasks

by Jakob Nielsen on May 13, 2013

Summary: Repetitive actions on websites often work well, but when users try something new, they frequently fail.


You might feel a mismatch between your personal user experience and some of the usability findings I’ve reported recently:

  • On intranets, employees had an average success rate of 74%, meaning that 1/4 of the measured tasks ended in failure.
  • Teenagers—those supposed masters of technology—had an average success rate of 71%, and thus failed even more often than corporate users.
  • E-commerce users found what they wanted 64% of the time in their first query and only 28% of the time in more difficult searches.
  • Generalizing across many recent studies, I concluded that "users are incredibly bad at finding and researching things on the web."

A bleak picture (though it's better now than in the past).

But do you fail at least a quarter of the time when using the web? Probably not. Speaking personally, the web does beat me occasionally, but not as often as the research data would imply.

Why You Don't Fail As Much

There's a simple explanation as to why you and I don't experience the low usability levels seen in usability studies. In conducting user research, rule #1 is to recruit users who represent the target audience. So that’s what we do.

Thus, unless we run a specialized study—say, on how professional journalists use the PR area of corporate websites—our findings report on the performance of average users.

Dear Reader, you are not average. If you were, you wouldn't be reading this website.

  • You most likely have above-average IQ and education.
  • You certainly have above-average reading skills. (I employ more complex sentence structures and vocabulary than usually recommended, because I don't write for a broad consumer audience.)
  • You most definitely have superior knowledge of web design patterns and computer concepts. Even if you're a visual designer or marketer, if you work on Internet projects, you’re a veritable geek compared to normal people.

Given all this, you simply cannot predict how well the average person will use something based on your own personal experience. You're too good.

Why Users Don't Fail As Much

Even average users don't fail as much on an everyday basis as they do when we test them in usability studies.

The reason? Users spend an inordinate amount of time on websites they already know quite well. Furthermore, most of the tasks people do are fairly simple things they’ve done many times before on the same site, such as checking news headlines, scanning Facebook updates, or seeing whether their paycheck has been deposited into their bank account.

This is why users hate change: they've finally learned how to perform common tasks on the sites they visit the most. They therefore have a high success rate on these tasks—assuming the site design stays the same.

(This is not a reason to avoid all redesigns. Just because users are capable of doing something doesn't mean that they can do it well; still, you should be conservative, and you shouldn’t change the UI for the sake of change itself.)

I don't have an exact number, but based on observing users' free-form behavior in field studies, repeating well-known tasks probably accounts for about 90% of users' time online. So, users are trying something new only about 10% of the time.

For simplicity's sake, let's go with these percentages, and furthermore say that users achieve 100% success on their well-known tasks and 80% success on the new tasks. Under these assumptions, the everyday user experience will have a 98% success rate averaged across all tasks—new and old alike, weighted by how often they're performed.

So why worry about usability if the everyday user experience is so much better than what we measure in the usability lab?

One reason: People won’t become experienced users of a design if they fail at being first-time users of that design.

For sure, this is true for websites: if a site is too difficult or frustrating to use, people leave and try the next site down on the SERP.

If the task or application is very important to users, they might try harder and make it past a few failures. Still, the point remains: something with poor early-use usability will not become a favorite destination that people return to again and again.


Share this article: Twitter | LinkedIn | Google+ | Email