Usability costs and benefits

Development projects should spend 10% of their budget on usability. Following a usability redesign, websites increase desired metrics by 135% on average; intranets improve slightly less.

Ease of use doesn’t come from wishful thinking. It comes from conducting systematic usability engineering activities throughout the project lifecycle. This is real work and costs real money, though not as much as some people fear.

You can conduct simple forms of user testing in a few days and gain extensive insights into both user behavior and recommended design improvements. Still, before most people will commit to a lifecycle approach to usability, they want to know what it will cost and what they will gain. We set out to find the answers.

Usability Costs

To assess the total cost of usability (as opposed to the price of a single test), we collected data from 863 design projects that included usability activities. Depending on how we estimated it, usability costs were between 8% and 13% of the projects’ budget.

Based on this finding and findings from other surveys, we conclude that current best practices call for devoting about 10% of a project’s budget to usability.

Our full survey data reveals a slightly complicated mathematical model that relates project size to recommended usability spending. In essence, the cost of usability doesn’t increase linearly with project size, since many usability activities cost about the same, regardless of how big the project is. A project that’s ten times bigger, for example, typically requires only four times more usability spending.

Usability Benefits

We analyzed data from 42 cases where usability metrics were available for website redesigns. For the purpose of our analysis, we needed projects that collected the same metric both before and after the redesign so that we could accurately compare them and estimate the percentage improvement in usability. Such projects are hard to find, however, because qualitative studies are more common (and indeed the cheapest and usually the most practical approach to improving usability).

Averaged across the 42 pairs of measures we found, usability increased by 135% when we excluded five outliers with exceptionally big usability improvements. (If we include these outliers, the average improvement jumps to 202%.)

The improvement in usability metrics differed depending on the metric, as the following table shows:

Metric Average Improvement
Across Web Projects
Sales / conversion rate 100%
Traffic / visitor count 150%
User performance / productivity 161%
Use of specific (target) features 202%

Our current study does not include intranet redesign projects. However, based on our studies of intranet design potiential, I estimate that the average improvement in intranet usability metrics is a bit below 100%.

That intranet usability appears to increase slightly less than website usability is easily explained. Website usability is dominated by users’ ability to avoid errors in navigation and interpret new information. Data that we collected from computer design projects in the pre-Web era, for example, showed that usability can improve error avoidance substantially more than skilled user performance. Where Web usability is closely related to error avoidance, intranet usability is a mix of error avoidance and skilled user performance: In a well-managed intranet, users deal only with a single design, and thus eventually achieve a degree of skilled performance.

Estimating ROI

In terms of gross averages, I estimate that spending about 10% of a project’s budget on usability activities doubles usability. Unfortunately, such estimates do not produce an ROI number in the classic sense, because the two parameters are measured in different units: project cost is measured in money, and usability is measured in increased use, more efficient use, or higher user satisfaction.

Converting usability improvements to dollars is easy for e-commerce, where doubled sales have an immediate value. For intranets, productivity gains are also fairly easy to convert into monetary estimates: simply multiply the time saved by the hourly cost of your employees.

Other types of design projects are harder to convert into an exact ROI. What is the value of increased customer satisfaction? Of more traffic or increased use of your website’s target features? Those estimates vary between companies, and thus the monetary value of doubled usability also varies. But it will be substantial in most cases.

Typically, the more people use a design, the bigger the usability ROI since the benefits come from the added value that ease of use brings to each user. Doubling sales on a large e-commerce site obviously results in bigger numbers than on a small one.

Similarly, the estimated productivity gains from redesigning an intranet to improve usability are 8 times bigger than the costs for a company with 1,000 employees, 20 times bigger for a company with 10,000 employees, and 50 times bigger for a company with 100,000 employees.

Because usability gains far exceed the costs, I believe that the budget share allocated to usability will increase in the future, at least in big companies. Currently, I recommend spending 10% of a project’s budget on usability, but within a few years optimal ROI will probably require spending 20% or more. Obviously, there is a point at which the value from extra spending on usability will be less than the value of extra spending on other project components. However, I don’t know where that point of diminishing returns will be, since we’re nowhere near it in current practice.


This entry was posted in Introduction to HCI and Usability and tagged , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s