The Insighter

April 2010

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

Top tasks and task-completion are central

There are a few key things every web manager needs to track about their visitors – like what did you come to the website to do and were you able to complete your task? Let's call them TASKS and COMPLETION. It's how web managers focus their organization's time and effort on supporting the right TASKS, and supporting tasks RIGHT. Applications are no different. Consequently we spend a lot of time providing evidence for tasks and completion. The following table shows how we do that in our various services.

How we help you support the right TASKS, and support tasks RIGHT

Our service
What we do
Support the
right TASKS
tasks RIGHT
Expert evaluation We identify common problems visitors are likely to have on your website, and recommend how to improve the user experience. We verify whether top tasks and search terms are supported, and prioritize usability issues and error conditions. We analyze how well top tasks would likely be completed in real-life situations, based on many years of observing web behaviour.
Customer Centric
We set priorities based on 13 key factors, positive or negative, that visitors say best describe their experience on your website.  We summarize priorities for action based on answers to the question what do you typically come to the website to do We compare CCI scores of users able to complete a task vs. users who were not, and identify where improvements will have optimal impact. 
Top Task Identification We gather empirical evidence of tasks most important to users, and of the words users use when scanning for links.  We prioritize the TASKS by gathering user votes on specific phrases that they have indicated are most important to them. The output focuses task execution on the “long neck” – those tasks which are most critical to a website’s success.
Usability testing We test a website or application for its ability to support users in effectively and efficiently completing their top tasks. We gather empirical evidence of usability issues and error conditions, and prioritize them as critical, major, or minor. We measure task-completion rate, time-on-task, and disasters – when people think they’ve completed a task yet have not. We analyze the cognitive factors that cause failures, and recommend how to improve task-completion and time-on-task.
Design  We make recommendations for improvements to the user experience. We focus navigational links and menus on top tasks and words people scan for, and ensure that users can find appropriate content. We optimize a design to help users complete tasks efficiently and effectively, reduce steps, minimize clutter, etc.
Task Performance Indicator (TPI) We empirically establish baseline performance metrics for top tasks.  We measure the time it takes to complete top tasks. We measure the time it takes to complete top tasks, and variance from optimal completion time.
Performance Indicator (SPI)
We empirically establish baseline performance metrics for top tasks in which users begin from search engines. We measure, for each top search phrase, how efficient and effective users can get to up-to-date accurate content. We measure how directly the most frequent search phrases get users to the information they need - visibility in search results, how and ability to complete the task.

Although the table is not a complete description of our services, it does show how top tasks and task-completion are central to much of our work. So if we seem to dwell on the importance of top tasks, time-on-task, and task completion, it is because our clients can't do business without knowing. Neither could we.

As you plan your projects make sure they focus your constrained resources on top tasks and improve task-completion. Make sure your team is focused on tasks rather than tools and technologies. We hope it helps you keep your eye on the main thing. Here is some further information from Gerry McGovern to stimulate your thinking:

Back to Top

When non-HTML documents become a problem

Although the web has long provided support for non-HTML documents (PDF, Word, etc.) that doesn’t mean it is a good thing to do. It can make it difficult for users to find what they need. Non-HTML documents are often not written to support a good web experience – they are created with the intent of being printed, presented, etc.

The government of Canada excels at making native-format documents available on the web – Google lists millions of PDFs on sites – and an even higher number of pages since PDFs are more than one page, and are sometimes hundreds of pages long. On one hand it is good that public information is made available. But the government should be concerned about the user experience – users generally expect to use web content like they use HTML pages. But PDFs and non-HTML documents sometimes need to be made available to clients or citizens to see what was printed or presented. Web teams strike a balance between findability and ease of publishing, so PDFs keep getting made available.

How does a web manager decide when to convert files on your website into HTML? When should web content creators limit the number of non-HTML versions? If you have lots of PDFs, where do you begin migrating to HTML? Here are some issues to be aware of when making your decisions – non-HTML content becomes a problem when:

Users need to get around within the non-HTML content

  • Non-HTML documents often have no global navigational links
  • If there is a table of contents it may not be hyperlinked, or may be several scrolls down
  • The first screen in a print-oriented document is a cover page and may have no links at all
  • Consequently, scrolling and the Find utility are the principle means of getting around
  • The most common way out of a non-HTML document is the Back button or closing the window altogether
  • TIP: Make the table of contents into good links, and add way-finding links to help users navigate around to key spots within the non-HTML document

Users need to arrive deep into non-HTMLcontent

  • Non-HTML documents do not use anchor links, so clicking a URL to a PDF takes a user to the first page – rather than deep-linking into a specific place within the PDF
  • Searchers scan for familiar words when they arrive, but are confused when the words are not the first things they see
  • Users can't point other users to content, or get back to the exact page they just visited
  • TIP: Put key links on the first screen, and other visible places in the non-HTML document

Users visit the non-HTML content frequently

  • Even when an HTML equivalent is provided, users arrive at the PDF instead, because non-HTML documents rank high in search results because they have more words and instances
  • But the HTML equivalent is often what searchers really need – the PDF is just the general overall document
  • TIP: Link to the HTML version or updates from deep in the non-HTML version

Back to Top

Top Neo Tweets

Follow us on Twitter

Quote of the month

"Tasks matter. Tasks are how people get to the answers, and answers turn into sales."

Kristin Zhivago, Revenue Journal

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

We invite you to subscribe to our monthly

  Home   About Us   Services   Case Studies   Training   Teamworks