How does your website stack up against your closest competitors? Is the competition seeing the same seasonal trends? What about % percentage of mobile traffic and average time spent on site? Until recently it was hard to get a handle on this type of data without spending a lot of money on 3rd party analytics. The good news is that in 2014 Google made benchmarking a standard feature for analytics. All you have to do in order to start viewing this data is to agree to share your own data in the same process. This might sounds a little worrisome but keep in mind that your competitors can not view data individually and just like you can only see the aggregated data for a particular industry. To get started just log into your analytics account and click on Audience in the left menu and then click on Benchmarking.
The above figure gives you some idea of the immediate value of benchmarking in that you can quickly see how you stack up in terms of sessions vs. other sites in the same vertical. If your website had a one day drop in traffic and the entire industry did as well it might not be as much as of a concern as if it was just your website. Google allows you to look at a number of factors when it comes to Benchmarking that include acquisition channels, device channels (mobile vs. desktop) and geographical channels.
Here are four key pointers that I think are important when it comes to drawing conclusions from any benchmarking analysis.
1) Find the Proper Industry Comparable.
It’s important to make sure that you are comparing yourself to the correct group of competitors and with so many choices this can be a bit of a challenge. Google lets us choose from choose from 1600 different industries in 1250 geographical markets and then allows us to divide them into 7 size buckets. If you look at your weekly session data over the period of a year and you have selected the right industry you should recognize the same broad-scale trends. If you don’t you might want to try a few other industry possibilities. This process of backwards fitting the data is one way to make sure you have picked the right benchmark group.
2) Question What You Are Actually Being Benchmarked Against.
If you don’t consider geography you have over 11,200 options to benchmark against. Some of the options contain a large number of sample websites, but some only contain a small handful. If you see that the benchmark data only consists of 10 or 20 websites I might question its value compared to a larger sample size of websites.
3) Understand the Relationship Between Date Range and Sample Size.
When you start looking at the little green and red boxes that show how your website compares to your benchmark set you need to be a little careful. If you are only looking at a few days of data the results might be much different than if you look at an entire months worth. Make sure your date-wise sample size is large enough (30 days minimum) to give a meaningful result.
4) Use Larger Date Ranges to Get the Big Picture.
Looking at benchmark data and trying to make comparisons using a daily timeframe can be confusing due to the noise introduced by day-to-day variability. I suggest looking at weekly or monthly data which smoothes out the short term variability and allows you to see the bigger trends that benchmarking is actually most useful for.
Suggestions for further reading;
Benchmarking is an important tool and when used correctly can help us understand the performance of a website within a larger context. If you have any questions or comments about Benchmarking please comment and I’ll get right back to you.