Guest column: Top Tasks

I am a big fan of the Top Tasks methodology. As far as I know, I am the only person to ever have written about its use in libraries.

Column originally published in The Informed Librarian Online, September 2025.

Top Tasks: Finding Out What People Really Want Most from your Website

This article presents a survey methodology designed to find out what library-related tasks really matter most to your community of users (“Top Tasks”) so you can prioritize testing and changes to your library website on a firm foundation of direct user data. 

It is a fact universally acknowledged (well, by web librarians anyway) that library employees use the library website way more than any individual library users ever will, and also way differently. 

Hours and hours of library website use in the course of one’s library job have been known to create strongly held and felt positions on what the library website ought to do, or be, or look like. Are these strongly held and felt positions necessarily incorrect? No, not at all. But as a user experience librarian, I need to work with direct feedback from library users to drive action regarding the public website. Spoiler: it’s hard to get this. People are busy, they don’t care, they care too much, they get too many surveys, it’s hard to write a really good survey, and so on. Lots of libraries have limited or contradictory data about their websites. Website analytics can be murky. Even when we have data, different people interpret it differently. 

Despite a truly appalling number of surveys, usability studies and the like over the last twenty-five years, I found myself still sitting in meetings, struggling to find common ground on how to approach this very important question: what were the most important tasks for the people visiting our website? 

I’ve run across what I think is a better way to ask this question so we can really find out. It’s called Top Tasks, and I’m writing about it today because I have yet to see anything about this methodology appear in the library literature. It was very helpful to me and my colleagues, and I hope will be very helpful to you. 

An Overview of Top Tasks Methodology

The overall idea: once you have identified the Top Tasks for the people visiting your website, you can prioritize testing and improving the user experience associated with accomplishing them: make them quicker to find, easier, more efficient, maybe even a source of delight. Fantastic!

I first heard of Gerry McGovern and the Top Tasks methodology somewhere around 2015 in an article published on A List Apart: “What Really Matters: Focusing on Top Tasks.” Put simply, he said there was too much clutter on the Web because it’s so easy to publish digitally (still true, by the way). It becomes hard for people to sift through all this clutter to do what they are trying to do. 

He gave a very inspiring example of working with a company that started by generating a list of over 600 possible tasks for which people might visit their site (!!), winnowed those down to a “short list” of 67 tasks, and then asked website visitors to choose their five most important tasks from a single randomized listing of the short list tasks.

Okay, great. But how many responses do I need to feel that this survey “worked”? “Ideally, you want 400-plus responses to get statistically reliable results, but definitely need a minimum of 100,” says McGovern.

Results are presented in a league table where the ranking criteria for each task is the total vote, the percentage of the overall vote for that task, and the cumulative vote. The tasks that get more than 50% of the cumulative vote are the Top Tasks. Let’s pretend that I did a survey on Top Fruits, which I realize are nothing like tasks, but it will give us an example table. 

Table 1: Top Fruits League Table Example

In my example (for which I used the same numbers as McGovern) the Top Tasks are italicized. Tasks one through nine accounted for half of the cumulative vote. To be clear, that means slightly more than half the people picked one of those nine tasks as one of their top five most important. Those nine tasks represent about 13% of the entire list of 67 tasks presented to survey participants. Are the other 58 tasks irrelevant? No, but they are not the highest priorities for attention for library visitors – therefore, nor should they be for the library web team. 

Shortly after learning about this in 2015, my team and I tried it out. I also replicated the Top Tasks methodology and survey a second time at a different institution in 2023. Meanwhile, McGovern and his team continued their work on the methodology and have since shared a few more suggestions about process. He wrote a book (Top Tasks: A how-to guide, first chapter freely available online) in 2018 and another article in 2022, “Top Tasks: To Focus On What Matters You Must De-Focus on What Doesn’t.” 

Caveat Bibliothecarius (Let the Librarian Beware)

The first caveat in this section: it’s going to get rather nitty gritty after this, so I can imagine many of you might most enjoy considering this the end of the article. Thus far I hope it’s been thought-provoking. Perhaps you might pass the idea along to your colleagues in web or UX work. Thanks for reading this far!

For the rest, if you decide to read on – don’t say I didn’t warn you! I’ll conclude with some very specific, practical notes drawn from my experiences of doing this twice specifically within libraries and McGovern’s additional directions (do make sure you read at least the articles and freely available first chapter noted in my final section below, they include essential information you’ll need to know).

This project might sound like it’s only about the website. It’s not. It will work most completely in an environment where there’s an openness to considering all aspects of process and workflow in the pursuit of improving user experience. But, it can also have some effectiveness for you even if you really just need to find a way to have more productive conversations about what does or does not belong on the home page. 

You’re going to need a team for this, and you’re going to need them to commit to several weeks of work (intermittent work, yes, but still work). Aim for a group that is broadly representative – this is important for ensuring that the list is accurate and comprehensive and also for practical reasons, like buy-in and such. (See a few more Secrets to Success from McGovern). 

I advise to give each team member their own spreadsheet in which they can be as comprehensive as they’d like, then when it’s time to create the short list, you can merge all the entries (with initials, in case of questions) into one long list to begin the process of eliminating duplicates, finalizing wording, etc. (Here’s a simple template you can review.)

Going from the comprehensive list to the final short list is the most difficult part of the entire process. Libraries have particular challenges here. Some of the words we use all the time are extremely vague, some of the words we use are quite technical and jargony, and one way or another people might often be confused. To give just two examples: Resources (absolutely no, no matter what, simply too indefinite); Journal (it’s a noun, it’s a verb, how’s it different from an article, whee!).

• Definitely read the shortlisting notes from McGovern, which includes his full list of specific direction about short list entries.
  • Avoid verbs, statements, questions. Use nouns, don’t bother with ‘and’ – just use commas between (ie, “Fines, fees”).
  • Keep it short, each entry 65 characters or less.
  • No: 
    • special brands (I know this is especially hard and I confess we were not able to steel ourselves to dis-include our branded discovery tool, but if I do this again I will try)
    • departments
    • subjects
    • audiences (e.g., “information for current students”)

Do your best to be as clear as you can. You can review our final short list from 2023 if it helps to see an example, though it’s important to complete this step internally with your own team. Our list is certainly no exemplar of perfection. If I had it to do again, I would have taken more to heart the advice to avoid listing formats, which I think would have made our final results more impactful. Specific to formats, consider these questions: 

• Is it really critical to know what type of materials individuals want most via this specific survey instrument? Or is it more important to know that materials are what they want, full stop?
• If, for example, you find out that books are the most important thing, followed by articles, then journals will that really be something you can take specific action on within your website? 
  • If no, are there groupings of materials that might allow people to answer more broadly but also give you some insight beyond just ‘materials’ 
    • Scholarly materials (books, articles, journals); Media (video, streaming, audio); Fiction; etc.; etc. 

On the other hand, you may have a case for including select formats. Do you have specific format categories where you need hard evidence about their relative priority to your constituents? Perhaps a sizeable collection of globes is creating contention in discussions, for example.

When it’s time to do the survey: yes, it has to be one complete and continuous list on one page of the web survey. If not, don’t bother doing this. Limit demographic questions as much as possible, because the mental load of choosing five items from the continuous list is considerable. Working in an academic environment, I ask about affiliation (faculty, staff, undergraduate, graduate, other). I also usually ask if the respondent is a library employee, for two reasons:

1.     I exempt library employees from being eligible for any incentive program (e.g., “chance to win”);

2.     I want to be able to group (or filter) the library employee responses. 

Be gentle with your colleagues when you get the results. Just because a task was the very last on the league board does not mean it is without value. It might be someone’s specific job to do that task. That task just might not require ongoing attention as far as the website is concerned. However, let me be clear. If you find content you can prune (and you will!), do so vigorously, thoroughly, and promptly. Gentleness is for people, not for webpages. Delete. 

In our hearts, I believe we all want to focus on what matters most to the people that our libraries serve. This method is one means of collecting data that will set you on that path. Furthermore, if communicated clearly, consistently and diplomatically, completing Top Tasks can produce a lot of interest and enthusiasm amongst colleagues which you can later reap in the form of more robust future collaboration across your organization. 

References, resources

McGovern, Gerry. 2015. “What Really Matters: Focusing on Top Tasks.” A List Apart, April 21. https://alistapart.com/article/what-really-matters-focusing-on-top-tasks/.

McGovern, Gerry. 2018. Top Tasks: A How-to Guide: A How-to Guide. First Edition. Silver Beach. First chapter freely available at: https://gerrymcgovern.com/books/top-tasks-a-how-to-guide/read-the-first-chapter/ 

McGovern, Gerry. 2022. “Top Tasks: To Focus On What Matters You Must De-Focus On What Doesn’t.” Smashing Magazine, May 13. https://www.smashingmagazine.com/2022/05/top-tasks-focus-what-matters-must-defocus-what-doesnt/.

Developed by me, freely available via Google Sheets: Example Task Worksheet; Final Short List 2023

Copyright 2025 by Courtney McDonald.