If you cannot measure it, you cannot change it. Many of you know this and try to measure as many aspects of your website as possible. These analytics provide you with an overwhelming amount of information. Yet, I caution you about what numbers you try to change. In some cases, there are unintended consequences to your metrics. Here are some tips on how to avoid or at least be prepared for them.
How to Get More Cobras
British colonists in India wanted to reduce the number of cobras. They offered a bounty for each cobra head as an incentive to reduce the population. Soon they discovered villagers were farm-raising these snakes to earn extra money. This solution became a problem when they removed the bounty. Those farmed snakes were released into the wild increasing the total number of cobras. This is just one example of an incentive program gone awry.
Examples of Unintended Metrics
What metrics will change because you are trying to change? Here are a few examples of what I have seen during my tenure in private corporations. You will see some are direct relationships, while others are inverse.
- An increase in social media awareness may mean a decrease in engagement.
- An increase in site traffic may result in a higher bounce rate.
- A newly optimized page template may decrease the time spent on each page.
- Adding often-accessed information to the homepage may result in less time on site.
- Increases in time on site may be the result of customer frustration.
- Putting an email subscription modal may increase your email list, but can result in fewer email open rates.
- Paid advertising can increase traffic but drive down the percentage of meaningful interactions.
Don’t Ask What without Why
Remember that analytics only tell you what happened. There is no insight into why it happened. Unless you are interviewing users on a regular basis, any attempt to discern analytics is conjecture at best. What is worse is that some people see a number change and rush in to make a change. Start by looking at your numbers and ask visitors to your site “why”. If you conduct surveys, encourage subjective data. This means long-form answers to questions about why they performed certain actions.
Do Not Attach Job Performance to Analytics
Your and/or your team is likely doing all they can to create a good digital ministry. When you attach someone’s performance to a metric, they might place that number above quality. Also, you may be driving down meaningful interactions on your digital platforms.
Review what analytics you are tracking as well as the key metrics you associate with good performance. Think about any unintended consequences and how you can either account for them in your metrics, or mitigate those effects. Brief your leadership team about what these numbers do and do not say about behaviors on your properties.
Photo by RODNAE Productions.