If, like myself, you started your online business 20 years ago, you remember that websites used to have “stats”. We then upgraded to “Analytics”, and here we are today, approaching the year 2020 and all the companies have whole departments dedicated to analyzing data and subsequently optimizing their website, based on these results. But let’s see, what is actually in our power to improve, without setting up a whole analytical department for our business.
Let’s start with a simple question: what is the role of Google Analytics in your business? Who are your customers? And, how can you anticipate their behavior? We sometimes forget these types of questions, due to all the technical advantages for measuring behavior. However, how can you really influence sales and predict users’ behavior, thanks to the right data from analytics? The technological skills are important, but it is only a tool. If you have been using Google Analytics but never really figured out how to properly analyze your data, optimize it, and ultimately improve your ROI, here are a few tips. You will be able to take your business to a whole new level, based on these 3 striking tips.
1. Bye-bye sessions: measure your events
Google Analytics measures everything based on sessions. They have been using this method since they took over Urchin, almost 15 years ago. Now, what used to be good 15 years ago is obsolete today, therefore this is going to change. Among other things, due to the increasing concern about privacy and ITP development aimed at Safari browsers. However, these changes also have to do with user behavior. You simply cannot ignore mobile and app behavior these days.
That is why Google Analytics has a new measurement method in which they link website and app use, based on “events”. Currently, this is still in beta. Former Google Evangelist Krista Seiden gave an extensive presentation at the Digital Analytics Congress about how this would look in practice. In terms of implementation, to me, it looked very technically challenging and advanced. So there is work to be done for your web builder or technical web analyst. If you have an IT guy on your team in your company, this will surely keep him busy for a while.
Measuring on an event basis
The possibilities are of course interesting, despite the fact it’s only going to e implemented by tech-savvy webmasters… You can analyze and optimize customer travel on an event basis. For example, Krista indicated that with this new measurement method you can now determine the ‘real’ commitment. By this, she means that you can measure the actual active time per user. With the current measurement method, this is only an indication.
What are the alternatives?
If you are not keen on diving deep into the new model of Google Analytics, you can give SolarWinds® Loggly® a try. This is a unique and modern analytics solution, which simplifies logs collecting and monitoring for those of us who operate in advanced IT environments. Unlike other log analytics solutions, Loggly has an easy setup and allows you to get started immediately. No need to install any agent, as you can upload logs using either TCP, UDP, TLS or HTTP/S. The main advantage is that you will have all your logs and analytics data in one place, where you can search the logs and see the events and their impact across all aspects of the website’s functionality. Visit their website for more information.
2. Determine the quality of your conversion optimization program
Many e-commerce based shops perform regular A / B tests to improve their conversion. While it did prove to be a potent indicator and, do you also check the data quality of the A / B test? I personally think that relatively few people do this. Here, I can give you a few practical tips on how to maintain the “quality” status of your A / B tests.
- Take a critical look at your sample distribution, i.e. the distribution of new versus returning users in your tests.
- Look critically at sample contamination. What percentage of users see both test variants?
- Be careful when generalizing your test results. You cannot compare a successful A / B test in a sales period one-on-one with a non-sales period.
- Analyze in moderation and correct wherever possible. Of course, you can get a positive result from every test. Therefore, analyze your test results based on your predefined test hypothesis.
- Provide a clear format. By this, I mean that you must use a template for your quality control. In other words: when can you assume that your A / B test provides reliable data?
3. Market research: how do you target the right customer?
What is the additional value of regular market research? With Google Analytics, for example, you sometimes forget this. Look further than your quantitative data. For example, what happens with an e-shop with incorrect e-commerce data in Google Analytics. How do we deal with it? Translating data into human behavior. Again, this is exactly the case where Loggy can come in handy. Google Analytics is a good tool, but it’s incomplete, you may find something such as heat maps and eye-tracking far more useful and easy to comprehend. After all, logs alone are useless, if you don’t know how to extract the needed information and how to manipulate it to your advantage.
Regular market research must be an inevitable part of every business, especially online. Talk, watch and listen to your customers. Use all the combined data analytics you can get, to solve all potential problems, optimize, optimize and once again, after a new round of testing – optimize once more. There are countless examples on the web, where website owners appeared to have everything in order, yet sales and traffic failed to increase or, even worse, declined for no apparent reason.
In moments of frustration, when you fail to see the picture as a whole, cold-pressed data extracted from your analytics, visually appealing and laid out for you in a colorful and easy to understand manner – can give you the right answer. It’s no wonder, that big companies have a whole department dedicated to data analyzing. It’s only when you understand the behavior of your visitors, combined with the behavior of your website, you can make small, tactical changes, which, in the long run, will lead to big results.