Julie Søgaard

Designing for expert users

On most design projects the UX designer will tell the team not to overestimate the users.

I usually say something like this: “Please remember that the user might not be as experienced as you. Think of him or her as a complete novice who knows nothing about the application at all.” This is said to make sure that the team does not set the bar too high for the user, that the application does not become too complex, and that the team have a realistic view of the user in mind.

But this is not the case with work tools or business software.

Trained users

Users of work tools often become experts over time as they use the application on a daily basis. Sure, they were beginners when they started using the application for the very first time, and in some cases they are going to learn to use new features as well – thus becomming novices in these features. But because work tasks are rather specialized for most people, business applications are facing ‘expert users’ for most of the time.

In my previous job I worked at Microsoft designing ERP solutions, and we rarely considered the users beginners – they were domain experts as very often also expert users of the application. And this lead to a very different view of the user than the ‘novice user’.

Experts vs. novices

Let’s take a look at how the expert user differentiates him or herself from the novice user:

  • Focus on efficiency and speed: For the expert user efficiency is much more important than learnability. This is because the user has already invested the time needed to learn to use it the tool, and this is a one-time investment. What is now relevant is efficiency and productivity. In other words, the tasks must be completed as fast as possible without errors. Therefor efficiency will often be a more prominent criteria for the expert user than the basic ease of use.
  • Keyboard: Expert users are reluctant to use the mouse for repetitive and frequent tasks. It is simply too slow (and not very good for the hands). They will rather learn a set of shortcuts and keyboard commands – even though these can be complex and hard to memorize.
  • Overview: Expert users look for oppportunities to oversee large amounts of data in relation to each other or in context of other factors.
  • Control: The expert user wants to feel in complete control over his or her work. This is an important success criteria that also effects the pleasurability of the application. This means that the more feeling in control, the more they like the experience.

Challenges

Expert users are much harder to design for than regular users. It takes more time, is more complex, and requires the designer to make much harder design decisions.

Why is that?

First of all, expert users have a deep and thorough understanding of the domain which can in some cases be very very complex. And the UX designer needs to know both about the domain AND how the expert user sees and experiences it. This requires lots of user research, such as interviews and observational studies, as well as actual domain knowledge. Before designing the user experience, the UX designer needs to understand the typical tasks and workflows in the domain, user competencies and goals, existing pain points, business objectives and much more.

Secondly, it is hard to design for expert use, as you cannot run a tradiotional usability study with expert user scenarios. The expert user scenario is not a walk-up-and-use case, but rather a situation where the users has already learned to use the system through a previous learning period.So forget about a typical ‘think-aloud’ usability test – this will simply not cover the case of a user having become an expert.

A study of an actual expert user scenario would require that you study the entire learning period and identify the problems herein and how fast it goes. It also requires you to test effectivity, speed, error percent, success rate and other relevant measures. These types of studies are harder and more time demanding than a regular usability study.

Thirdly, the design needs to solve more complex problems than for novices. Expert users deal with complex problems and the user interface needs to facilitate these problems and processes. The complexity rises even further because expert users and companies differ widely from company to company and from industry to industry. This often means loads of complecity to deal with for the designer – quite overwhelming when all you really want to do is make everything seem easy as a pie in the UI.

What expert users say they want

Well, why don’t we just ask the expert user?

Unfortunately expert users can be rather difficult to ask about user experiences. When you ask the expert user what they want, they will often have requests that are not necessarily things that will benefit them if actually implemented.

Expert users will sometimes  ask for thing like this:

  • Not to change the current user interface at all. (Since they have spent so long time to learn it by now and changes will make them do more work than not.)
  • Extremely crowded interfaces, so they can see every bit of information in the system at once. Expert users often have the false conception that they need to see all details at once and that this gives them the best type of overview and control.
  • Designs that put a focus on edge cases, exceptions and details, that are used in very rare cases.

All of these asks are things that the UX designer will need to take into account but often not design for, as they will not make the users happy.

Design guidelines

The dilemma between designing for novices versus experts still comes to me as one of the most challenging ones in interaction design.

Here are some of the design guidelines I have picked up along the way:

  • It is not always wise to just give the user what they want (ie. crowded UI’s) because even though the user is an expert in the domain, he/she is not a UX designer. Listen to the user’s requests, and meet their needs in the best possible ways – but not necessarily howthey think it should be done. And be sure to test early prototypes to see if alternative solutions will work.
  • Design for the learning experience by specifically adressing scenarios in which the user discovers and learns about the application for the first time. Bear in mind that the user will return after this first experience, so they can build upon this first experience and their knowledge about how the application works.
  • Count the number of mouse clicks and keyboard presses and optimize this parameter to make the application as efficient as possible. Additionally you can measure the speed of actions, number of screens involved in a task and so on.
  • Design for the realitistic scenarios. This includes interruptions (which will often occur in real work life), the specific work environment, as well as colleague/customer interaction during interaction with the application.
  • Include use of mobile devices when this is adds value to the expert user. For example, users will often want to check on statuses and events on their mobile phone before and after work hours.
  • When there are both experts and beginners using the app, the UX designer must address both groups. This might mean creating 2 different UI’s or using progressive disclosure to reveal more information bit by bit.

What about simplicity?

With all this complexity will there still be room to the design principle of simplicity?

Yes, simplicity is still the main design principle in my book – also when designing for expert users.

Every user should have a great first experience that will make them learn to understand and use the software, followed by a pleasant learning period that leads into expert usage without problems or breakdowns on the way. And none of this can be achieved without applying the principle of simplicity to the user experience wherever this is possible.

This means that complexity in the expert user’s domain needs to be translated into simple and intuitive UI’s as much as you can. But nevertheless, the user interface will often end up showing more data and actions than a non-expert application and this is OK. I find that it is all about finding the right balance between the power to perform (mulitple) actions, having a good overview, showing enough complexity to make it work and still keeping it as simple as possible.

The fun side

Even though UX designers often fear expert users, designing for experts can actually be fun. To me it is always exciting to learn about how an expert operates in a complex domain, and I enjoy to make complex worlds appear simple and easily accessible in the UI of the app. Finally, it is also rewarding because improving the UX for people at work can make a rather large impact on their daily life.

Keep on reading

Calculating Return on investment for UX

When I started working in the field of UX and usability, the...

Why work tools should be designed differently

I must admit that some of my favorite projects are those where...

6 challenges when designing for work

I recently gave a talk at about "designing for work". This made...