In a world where everything is digital (or, if not, it’s definitely it’s moving that way), we are completely reliant on website analytics data to help us understand which of our marketing efforts is working and which isn’t. The problem is, we are dependent on the data we analyse being accurate. All too often this isn’t the case; reporting doesn’t match up, data is missing altogether, and in some cases we are let down by the dreaded (not provided) or (not set). In order to try and help in these highly frustrating hours of need, here are some top tips to troubleshoot your Google Analytics data discrepancies.
Check for User Errors Often the bad workman blames his tools, and in some instances this can be the case with analytics reporting. It’s not always our fault, but things like Google’s fiddly date selector can mislead us by defaulting back to last 30 days when we least expect it. Always check your date range. Leaving advanced segments on is another hurdle. It can sometimes be all too easy to carry on with your report without realising you have left your SEO filter on when trying to analyse PPC data! Double check these things; as we all do it, I have mini palpitations on a regular basis from seeing data I did not expect.
Is the Code on Every Page? It is something I have seen time and time again, from some of the biggest and some of the smallest businesses. When the bounce rate is extremely high, or an exit page seems to be a major drop out of traffic, it can often be down to one page missing the tracking code. This also has the bigger knock-on effect of self-referring data; it looks like the traffic from the page with no code is new traffic to the site, whereas actually (you guessed it) it comes straight from your domain. Always check your code is on every page of the website.
UA Number If your data is not importing at all, it could be the case that you have the wrong UA number in your code, and the code is not matching the analytics account. This simply means google is capturing data from your website and populating it to an account with the relevant UA – not good. To check this, the simplest method is to visit your page and view the source of the page, then search for the following in Google analytics:
ga(‘create’, ‘UA-1234567-8, ‘auto’);
It is also fundamentally important to keep the code as it is when you take it from the Google interface; any additional characters or white pace could end up having a knock-on effect.Profile Blocking Traffic Finally, one of the biggest reason for mismatches between Analytics data and other platforms is that you have a filter set up which is blocking and filtering out activity. An example of this would be blocking your own company visits to the website so as not to skew your numbers; often, you would block the business IP address. Another cause of irreconcilable data is blocking a particular location of traffic that is irrelevant to your business. For example, if you work mainly in the US, you may want to block any external-country traffic to the site. In this case, if you are correlating your data with your web stats, the numbers might be quite different. It is also worth taking into account that all tracking and analytics systems are likely to vary, and will have discrepancies between them based on when, what and how they track. Some tools track the first interaction, some the last, and others a blend of the two as well as all points in between. It is therefore important that you know what you are tracking and how; from there, you can use the points above to hopefully identify any data discrepancies you are seeing. These are some of the key points I would suggest checking out first, but there are others. Feel free to share any of the challenges you have come across, we would love to hear your tips.