What are conversion rates, and how do we measure them?
Conversion rate, in a nutshell, refers to the percentage of web users who move on to make a desired action, or a goal. The simplest example of conversion rate is the percentage of visitors to a website who go from browsing through the website, to buying something on it. The act of browsing to buying is where the ‘conversion’ happens.
Another example by usability experts the NN/g group shows this scenario:
100,000 users log on to an e-commerce site in April. 2000 of these 100,000 users made a purchase; therefore, the site’s conversion rate is 2 percent.
Of course, it’s not as simple as it looks, as there are plenty of questions to ask yourself when thinking of the conversion process. For instance, you may be wondering how to count baseline number of web users who visit your site. Should you only count unique visitors or should you mark down every time one user visits the site during the measurement period? Likewise, how do you count the ‘desired actions’ users make? Do you count a conversion event that’s unique to every person, or count every event regardless of how many times it occurs with a user?
Conversion Events Explained
While conversion rates are strongly correlated to e-commerce sales, it’s a concept that should matter to people concerned with other aspects of a website, even design. Conversion events are a good indicator of a site’s performance. These events can include anything from a purchase on a site, signing up as a registered user, signing up for a newsletter, downloading a mobile app, and repeatedly returning to the site, which means it might be better to count for unique users only.
Why Does this All Matter, Especially for UX?
Simple. Conversion rates are a key performance indicator (KPI) on the success of your user experience. But what about conversion counts, you might ask?
Think of it this way. If you opened a store and marketed it on all platforms (print, radio, TV, web, social media), there’s a good chance people will flock to your establishment. This is where conversion counts come in, thanks to your marketing efforts. What happens in the store, whether a visitor browses through products, signs up for coupons, and makes a purchase, can be attributed to your store’s design—in the context of websites, the web design, and ultimately, the user experience.
For design, conversion counts are unreliable because they count the number of people who visit the website; whereas conversion rates measure the impact of web design since they account for what people do once they’re on the site.
While it’s often better to track the ratio of users who proceed to make conversions, this is not always the case.
The ratios can be misleading, especially if there’s a sudden surge in variable traffic that’s of varying quality. For instance, if your e-commerce site were to upload content about a rumor of the next iPhone when the rumor mill is at its busiest, it’s likely to lead to a surge in visits from Apple fans but no actual purchases made. Just people getting hold of a link to that piece of information, probably leading to a share, and ultimately leaving.
A sudden spike in visits such as the above example but no conversion events will cause your conversion rates to plummet, but that shouldn’t be a cause for worry since you can connect the dots to whatever chances you made. In the example scenario, it’s the new content, not a design change.
This means that you should always pay attention to your traffic sources and find out whether new traffic is in line or different from your usual users.
Conversion Against Usability Metrics
A solid conversion rate usually hovers between 1–10%. However, when testing for usability, a good score will be at 80 percent and above. The reason for this disparity is that in a test situation like the ones NN/g regularly conducts, users are urged to perform a task, so the score is based on whether it’s possible for them to do so. It negates important factors like price, and looks purely on the success of pulling off the assigned ask. If a site scores 80 percent, it simply means 20 percent were unable to use the site, not that the 80 percent are good as paying customers.
Relating Conversion Rate to User Experience
We at Enform agree with NN/g’s recommendation of monitoring conversion rates in lieu of design changes to see whether a UX investment is worth it or not. While it’s true that other factors besides UX can impact conversion, design arguably has the biggest effect.