The return of investment (ROI) of UX for internal applications

Published in

UX Collective

9 min readOct 28, 2021

The field of design has been struggling for years to prove the value of good UX design. The costs can clearly be shown, but the return of investment is more difficult to prove or can be easily expressed in percentages or money earned.

Fortunately, the results are becoming more and more visible with contemporary examples from practice throughout the years of improving UX. So, what I want to discuss in this blog is: how do you measure and prove the value of interface design (UI) and User Experience (UX) for internal business applications and why it is so important.

It is difficult to gather evidence directly (otherwise I wouldn’t need to write this blog), so I am going to try and explain it step by step and give some examples why the investment in good UX easily earns itself back.

Firstly, we start with measuring the effects of UX. If you cannot measure what you are doing, you cannot prove anything. So here we go!

The measuring points (KPIs)

The term KPI is probably a well-known term for many who read it. For those who aren’t familiar with the term, KPIs allow you to measure how your department is performing within the company and whether your results improve. The KPIs for a UX department are probably different than for other departments within a company.

The KPIs for a UX department can handle can be subdivided into two categories: ‘behavior’ and ‘feeling’:

‘Behavior’ KPIs:

· Abandonment rate
· Pageviews / time on site / number of clicks
· Problems and frustrations
· Task successfulness
· Task time

‘Feeling’ KPIs:

· Loyalty
· Usability
· Credibility
· Look and feel

Kids looking happy while on their phone with homework on the table

‘Behavior’ measurement

How to measure and improve

The Behavior KPIs are measured by tasks in your application. You should run and measure these once per month/quarter and compare the results with the previous score to see if your product is improving. For example, you can also choose to compare it with the competition.

Abandonment rate

This is the ratio of actions initiated to completed actions. For a web shop for example, this is how many users started a ‘checkout’ and actually finished it. For a website, this could be how many times the homepage has been requested and how many of them actually clicked on articles for example.

For internal applications this is a bit more difficult to measure and you will have to put in some extra work. Like a user research to determine the satisfaction of your users. Because the users have to use the application to do their work, you won’t be able to automatically measure the abandonment rate.

Pageviews / time on the site / number of clicks

The number of pages that are viewed, the time spent on the website or the number of clicks that someone has to make to reach the goal, are the measuring points for this KPI. You can tinker with this in a targeted matter. For internal applications, optimizing the flow or reducing clicks is an easy way of saving time and decrease bad experiences.

Problems and frustrations

The unique problems and frustrations and/or the number of users experiencing these problems and frustrations. You can measure this by doing a user survey: let users think out loud and let them tell you what they do and/or expect when they click through your applications or software, write down the problems and/or frustrations and then quantify and group them so that you can see how often and how many participants encounter certain problems and frustrations.

Task Succeeded and Task Time

How many of the tasks started are actually completed without problems and how long does it take users to complete the task.

‘Feeling’ measurement

To get a score for the ‘feeling’ part of an application, you need to measure loyalty, usability, credibility and visual appearance. The most popular ways to do this are CSAT, NPS, SUS, TPI and SUPR-Q.

The KPIs under ‘feeling’ are measured at different points and different time periods than the KPIs covered under ‘behavior’. Also with these KPIs it is important to repeat them in order to register any improvement.

CSAT (Customer Satisfaction Score)

This measures the satisfaction of your user (customer) by creating a questionnaire with as little as one question to a complete satisfaction survey and let the users fill this in. More about CSAT

NPS (Net Promoter Score)

This test consists of one question and you do it immediately after a user test or after someone has successfully accomplished their task in your application. Example of an NPS question is: On a scale from 0 to 10, how likely is it that you recommend this product/service/app to someone else? Scores of 0–6 are dissatisfied customers/users who do not want to use your software again which are called detractors. The ones that give you a 9 or 10 are your promoters. NPS = Promoters — Detractors. (read more)

System Usability Scale (SUS)

You can let the users fill in a score using the Likert scale. After a task or goal within your application, the user receives a number of questions and can answers these with ‘strongly agree’ to ‘strongly disagree’.

TPI (Task Performace Indicator)

With TPI, users are asked a number of questions right after the task they just completed. These are all kind of different questions in different categories. You can read more about them on A List Apart.

SUPR-Q (Standardized User Experience Percentile Rank Questionnaire)

This questionnaire is linked to a certificate and consists of 8 items that measure the quality of the User Experience. This review uses user-friendliness, reliability, loyalty and appearance as units of measurement. More about this test can be found here: www.suprq.com

How do all these efforts give me any ROI?

When your employees, the ones working with the application you developed, have fewer clicks or less scrolling, it will give you a big time saver and therefore a huge productivity boost. Let me give you an example:

Your business application has a search function to filter all records. An employee needs to use this function 10 times a day. To start the search, the employee needs to click the ‘search’ button. When the search is folded open, there are so many search fields that the employee needs to search where the correct input field is. To execute the search, the employee needs to scroll down because the button to execute the search is a bit out of the screen.

This whole exercise will cost an experienced user around 2 seconds extra. That’s 20 seconds wasted per employee per day.

When you have 20 employees, that’s a total of 400 seconds wasted every day, which is almost 7 minutes a day. Doesn’t seem like much, but lets calculate it for a whole year…
Let’s say there are 200 working days in a year, making a total of 1400 minutes. That’s about 24 hours (a small work week) wasted every year for something that can be fixed in less than 2 hours.

I hear you think: that doesn’t seem like much. But most likely you have more than one of these situations in your business applications. Plus, some of these not-so-perfect-flows will cost your employees more than just 2 extra seconds.

And maybe even more important, your employees will be frustrated by the current design. It will reduce their willingness to work hard and efficiently and it will reduce their error/bug tolerance for the application.

With a good application and good user-flows, you will get employees who are more productive, supportive and more willing to give their best for the company.

Adding it all up

Better sales

If the sales team has to sell a product which is difficult to demonstrate or to explain how it works, they (and the product itself) will seem less reliable. Also, a prospect may decide to look further and come across a product that initially looks much better and is presented better. This means that if your product is easier to use, it sells much easier. Whether the product is technically well assembled, often comes second.

Support costs

When you haven’t thought about the UX of your product, support desk costs will be higher. Your support desk will be busy answering questions about how the application works or where a user can find something. In addition to higher support desk costs, the costs of learning new employees are also higher. The more complex the application, the longer it will take new employees to learn it.

A great example of very high and unnecessary support desk fees is that of a bank. Its call center spent a lot of time changing passwords and resetting phone calls. By adjusting the online environment, customers were able to change the password themselves and the process of requesting a new password was also greatly improved.

Because of this change, said bank has literally “saved millions a year” in call-center hours.

Productivity

Already shortly explained, but if software has (too)many steps for someone to do their job, it can slow down the user’s productivity. Through good research, wireframes, prototypes and user research the fastest route can be determined for someone to carry out their work and thereby significantly increase productivity.

Rebuilding

If you start building your own software, without proper user research, it may occur that the end user needs a different interface than the one that is being build. As a result, your users may not improve their way of working with the brand-new business application, or simply may not even be able to do their job with the new application. In both cases this will lead to the fact that they just do not support the app. As a result, (a part of) the application needs to be rebuilt. These are, of course, costs that you don’t need and can easily be prevented.

Build too much

Without proper research, the software can have many features that users will not use or don’t know how to use. These costs are easily to avoid, especially when you are building your own application. An example of too much functionality comes from BMW and their concept to build an app to unlock your car. Seems like a good idea: you have your phone with you anyway, you already use your phone for everything and people are used to installing apps. But in practice there appeared to be a problem…

In the worst-case scenario: 13 steps were needed* and in the best scenario: 9 steps. You will understand that this part of the application was hardly used and that it was not, in any way possible, an improvement on a traditional key at all.

*
1) Walk up to my car, 2) Pull out my smartphone, 3) Wake up my phone, 4) Unlock my phone, 5) Exit my last opened app, 6) Exit my last opened group, 7) Swipe through a sea of icons, searching for the app, 8) Tap the app icon, 9) Wait for the app to load and try to find the unlock action, 10) Make a guess with the menu and tap “Control”, 11) Tap the Unlock button, 12) Slide the slider to unlock, 13) Physically open the car door (my goal)

So, what is the ROI of UX?

The cost savings mentioned in this blog can be traced back to actual costs. But this will be different for every company and every software application and every user. However, gross estimates can quickly add up.

If you want to reduce the cost of a department or improve the productivity of your employees, please think about UX next time. You are probably surprised that a designer can help. But you should be more surprised of the amount of unnecessary costs incurred by poor UX.

When you use these steps and start calculating for yourself, how big of an improvement can be made within your organization, it is possible that user experience could become, an even more important asset within your company.