UX Research: metrics and analytics — which one is better?

As we all know research is what drives and fuels a UX Designer to generate a user-friendly design for their clients. However, with a sea of jargon, Metrics and Analytics seem to stick out the most. Why is that? Are they the same thing? Is one better than the other? Are you getting tired of these questions? I know I am getting tired of those questions, but they bring up a good point.

Metrics and analytics are key tools that are used for both UX designers and the client as well. But what is the difference between the two? First off, they both refer to collecting data but in different contexts. Metrics are specific data chosen and measured by researchers and Analytics are chosen data measured by software. What do those mean though?

Metrics are specific data chosen and measured by researchers

Analytics are chosen data measured by software

Analytics, the pain is in the name…

Analytics are tricky statistics that are known to confuse people; however, this is not on purpose. This confusion is from trying to put meaning behind the data: Does this mean it’s good? Does this mean it’s bad? What do these numbers mean? etc. Most of the time, analytics do not tell us anything.

Some common measurements that are analytical data are Time on Page, Percent Exit, Bounce Rate, Page Views, Monthly/Weekly/Daily/ Average Users, and many others. These data categories seem like they will be valuable, right? Sadly, if you said yes, you are wrong. When I look at Time on Page and I see a high number, I have no clear idea if that is bad or good.

Is it ultimately missing the why, what, and how is this happening? The only thing that can be done with these, in UX terms, is to create assumptions from it, and we all know what happens when you assume. Therefore, to answer these questions, we need to use another form of data collection that is chosen to help answer those questions.

Friendly Neighborhood Metrics

Metrics are a UX design close friend in helping him solve for solutions. This type of data collection is geared to answer the what, why, and how a design is failing or succeeding. These ultimately give the meaning to the data to see a design or a brand is suffering.

Okay, let’s say I have a client that is a theater company that wants to make their educational programs highlighted to increase the enrollment rate. How can we successfully measure this? Well, we could measure the conversion rate on the education link in the nav and the other statistics for the education page…right?

Wrong. Those are more linked to analytic data because what are rates telling us? They are saying that people are clicking to the page and staying on it for a long or short amount of time. So, again this will not tell us if it is working or failing.

Right. A more accurate measurement is taking account of the percent of people contacting the school about the educational programs and the percent increase in enrollment. These two metrics are telling us that people are understanding, aware, interested, and ready to talk about enrollment. Therefore, it is measuring how successful highlighting the educational program to improve the enrollment rate of schools.

Well, which one is better?

If we are looking at just results, metrics win this fight. Metrics are designed to measure success because they are catered to each problem. A good metric measurement is focused on answering why, what, and how questions that will lead to a better solution down the road. With that said, however, analytics are still needed. Without analytics, metrics would not be as strong as they are.

Metrics, in a way, are analytics with a purpose. Think back to the theater company exercise, the percent increase we are talking about is a more focused conversion rate that is aimed to see how enrollment rates are being affected by the design. When you just look at a regular conversion rate, it is just saying that a percent number of people are interacting with that call to action.

Therefore, when you are talking to a client and they ask about the analytics, be sure to pitch that metrics are more useful than the analytics.

Figuring out why something is happening is better than assuming why it is happening.