At the beginning of the month I attended a seminar hosted by User Experience expert, author, and speaker Gerry McGovern on measuring the customer experience with Top Task management for websites.
What is a Top Task?
The key to successful digital transformation is customer-centricity. Every website’s users have a small number of tasks they want to complete above everything else. These are regarded as Top Tasks.
Poorly-managed websites can become polluted with dozens (if not thousands) of worthless pages that are irrelevant to the user.
Websites are for answers, and identifying and improving the Top Tasks will help your customer success rate.
Top Task management is about streamlining your site by focusing on the content that matters most to your users. The European Commission has used the system to great success.
Benefits from Top Task management
Do you think you know what your website’s key content is? Well, you might be surprised.
One of the Norwegian Cancer Society’s main key performance indicators (KPIs) is gaining donations. Gerry McGovern’s Top Task management model highlighted that the most important content for their visitors was information about diagnosis and treatment.
They implemented a new site structure that reduced the prominence of content asking for donations and instead helped visitors get to the content that Top Tasks had shown mattered most to them.
This resulted in:
- 70% increase in one-time donations
- 88% increase in monthly donors registered.
- 164% increase in members registered.
- 348% increase in incoming links.
- 80% increase in visitors.
Looks like it worked out pretty well for them, so how do you go about identifying Top Tasks?
Identifying Top Tasks
Create a tasklist
Define scope for who your target demographic is and then create a tasklist. These tasks are similar to user stories. Here are a few tips on sourcing tasks:
- Analytical data from your website’s top 50 internal and external annual searches.
- Surveys and research on customers from the last three years.
- Customer feedback from social media, blogs, and community forums.
Shortlisting and survey design
After identifying an appropriate number of tasks, they should then be shortlisted to a maximum of one-hundred user tasks. Ideally, 80% of the tasks shortlisted should be customer focused with the remaining 20% focused on organisation.
Three to eight people from a broad group of representative stakeholders should be involved in the shortlisting process. Genuine experienced people who truly understand the customer’s needs. The same group should be involved in all shortlisting meetings throughout the process. A broad group from different departments is beneficial when presenting the data to management at a later stage.
When composing the final list of tasks, avoid framing the instruction from the point of view of the organisations. Instead compose the instructions from the open perspective of the user.
- “When you are choosing a university.”
- “Select the five tasks that are most important to you when interacting with the University of St Andrews.”
Participants and voting methods
The aim of the survey is to gather as many participants as possible for a thorough representation of data. Ideally, 400 or more participants, with an absolute minimum being 100.
In order to achieve a high participant count, it can be effective to be intrusive with links to the survey. Pop-ups are an effective method, or banners with a link at the top of a webpage. Sending out a newsletter to a list of customers is also effective.
The survey can run for two to four weeks in duration.
If your tasklist was thoroughly prepared the final results should display a logarithmic trend.
Commonly, 25% of the accumulative votes will highlight tasks by level of priority.
The top tasks have been identified, and the data / evidence can be presented to management as proof of what the customer truly needs to use the website for.
After identifying the Top Tasks, a thorough re-design process can then begin. But the process doesn’t end there. The customer success rate must be measured through a series of task performance indicator tests.
But, that’s for another blog post.