The Insighter

June 2011

View all articles on our new site

Neo Insight's e-newsletter on Customer Experience topics and techniques.

We invite you to subscribe to our monthly

In this issue

The Top Task Law: Where to Allocate Scarce Web Resources

Managers of large websites often struggle with how to allocate scarce resources (people, time, money) in the most cost-effective way, to maximize improvement of the user experience and also meet business goals. The Top Task Law is one simple, strong principle to help do this:

Total website quality improvement is proportional to
the frequency and importance of the improved task to users.

This Law has implications for managers, in particular how to allocate resources most efficently. But to understand why, we need to take a quick look at how this Law was derived.

Amdahl’s Law – A Bit of History

Back in the 1960s, the SuperComputer Wars were fought between Cray, CDC, Burroughs, Amdahl and others, for the honour (not always riches) of having the fastest computer on earth. At that time, Gene Amdahl came up with what became known as ‘Amdahl’s Law’, identifying limits on the amount of improvement that can be achieved in any system.

Gene Amdahl


Amdahl's Law

Performance improvement is proportional to how much a program uses the thing you improved.


What does this mean? For illustration, take a look at the image below. Imagine you're the manager trying to maximize this circuit's overall performance, wondering whether to spend resources to improve the components in area A or in area B. Maybe area A is easier to improve. A or B? That is the question.

However, supposing the computer uses area A only 5% of the time, but uses area B 90% of the time?

Amdahl's Law means that, if the component you're trying to improve contributes only 5% to your total processing time, then the maximum performance improvement you'll be able to make is only 5%.

So, if you want to maximize performance of this circuit, you should probably spend your resources on area B, because that’s where you can realize maximum return.

Amdahl’s Law applies to more than computers and software, it’s a management law. It helps managers decide where to focus their budgets and technical effort: where the biggest return on investment would be—at least in terms of maximising processing speed.

Website Performance and Top Tasks

So is there is an equivalent of Amdahl's Law that could be applied to help managers decide where to allocate resources on websites?

From dozens of projects and surveys, and from observation of thousands of users, we know that roughly 5% of a website typically provides 25% or more of its value. This is because, on any website, it turns out that there are a small set of tasks which are way more critical in terms of frequency and importance to end users than all the others.The Long Neck of users' top tasks

We call this the ‘long neck’ (Lou Rosenfeld has recently started discussing a similar concept, calling it the ‘short head’).

Similar to the situation Amdahl was addressing, then, some parts of a website are more critical for improvement than others—often  massively so.

So, let's propose a Top Task Law, and see if we can simply characterize maximum website improvement, from the users’ perspective.

The Top Task Law—First Draft:

Overall website performance improvement is proportional to how much
your users use the thing you improved.

This is pretty good. If your users carry out a particular task frequently on your website, then that should be a prime candidate for improvement. Just small improvements to these tasks could make a big difference. For instance, we analyzed a task on a higher education site recently where we calculated that shaving just 5 seconds off the task time by removing one unnecessary mouse click would save potential students visiting the site a total of over 322 hours in one year.

And on the other hand, if you have content supporting tasks which users just don't want to do, then improving those should be a very low priority (this is 'long tail' content which should usually be removed from the website).

But What About Poorly Supported Top Tasks?

This draft Top Task Law doesn’t quite capture everything. We’ve seen many usability tests where users want to carry out a task, but the site doesn’t support that task well so the users fail to complete it, or spend many long minutes before they finish.

So, in addition to improving tasks that users carry out frequently on a website, it's also important to improve those tasks that are a high priority to users, but that are poorly supported. Just looking for the high-frequency tasks on a site would optimize the current site's content, but would miss these important opportunities for improvement. We need to make sure the Top Task Law captures this factor.

And What About the Quality of the User Experience?

Traditional web analytics like hits and visits say very little about the quality of the user experience, or even about users’ basic successes or failures – did people find what they were looking for or give up or call a helpline etc.? These analytics also don’t tell us what tasks the users are trying to do. In fact, sometimes a high score on hits and visits tells us something is wrong – for example, frequent use of an A-Z Index, of FAQ pages, or of the on-site search is often due to sheer user desperation!

So measuring quantity (how many visits to a page, how long people stayed, etc.) to gauge users' priorities is to measure the wrong things. It can also lead managers down the garden paths of misguided design. For example, they can end up trying to make a site ‘sticky’, rather than helping users complete tasks quickly, efficiently and correctly.

So, the draft law needs revision. We need to augment its focus on frequency of use with consideration of the users’ priority tasks, and the quality of the user experience.

The Top Task Law—Revised:

Total website quality improvement is proportional to the frequency and
importance of the improved task to users.

ThFire alarm - an infrequent but important task.e Top Task Law is not as simple as Amdahl’s Law, because it needs to take into account not just frequency of use, but also user priorities; for example there may be infrequent but important tasks (Fire Alarms are physical-world examples!).

But similar to Amdahl’s Law, the point of the Top Task Law is to help guide managers when allocating budgets and resources for improving websites.

When put into practice, the Top Task Law prevents resources being wasted on ‘the long tail’ of content and tasks that are unimportant to users.

The implication of the Top Task Law for managers is:

To get the biggest ROI on improvements to your website, concentrate on
improving performance of the most frequent and important user tasks.

See also:

Learn more about Top Task Management. Email us or call us at (613) 271-3001 if you would like a free consultation about our Top Task Management services.

Back to Top

#uxinsults - Useful for those special occasions!

Have you ever used a website that is so awful you wish you could vent and communicate your extreme frustration to the web designer, developers or managers (yes, it could even be your own website or intranet!)? Well, we're here to help you.

Just for fun, we've put together a few concise aphorisms you might consider using when you feel brave.

  • The search engine is so useless, it can't even find itself!
  • Home page? Homeless page, more like it!
  • Redesigning the top levels of this site would be like rearranging the deck-chairs on the Titanic!
  • It's got dry ROT, wet ROT, in fact it's all ROT (Redundant, Out of Date, or Trivial content)
  • The navigation? “You are in a twisty maze of passages, all the same...”
  • 'Welcome' is not how I feel when I see this 'Welcome!' text for the tenth time!
  • For your ‘hero shot’, I recommend a huge picture of the devil.
  • Even 3 layers of lipstick couldn’t hide this pig!
  • This site is so sticky, it’s like taking a walk in a tarpit.
  • You know that revolving, flashy promotion that takes up half the page? It just needs a picture of a kitten to be perfect.

Tweet your favourite #uxinsult from our list, or tweet your own - use this button and edit the text. Make sure to use the hashtag #uxinsults: we'll keep track of them and retweet the best!

Back to Top

Quote of the month

“Instead of chasing after the search engines, chase after the user experience because the search engines are chasing after the user experience".

Matt Cutts, Google

If you have any comments on The Insighter, or ideas on usability topics you'd like to send us an email.

We invite you to subscribe to our monthly

  Home   About Us   Services   Case Studies   Training   Teamworks