5 Things You Shouldn’t Do With Google Analytics

Chapters:
Article written by Claudiu Murariu
Product manager for PadiAct.
Google+ profile

Google Analytics is pretty much it, when it comes to data about your visitors.  However, Google Analytics is just that: a tool that provides statistical data about your visitors, based on which you can improve your website.

While it can do many other things, it pretty much sucks at them. So,  what shouldn’t you do with Google Analytics?

Google Analytics is not a financial tool

Yes, there is eCommerce tracking but it is only meant for adding revenue data to your traffic data and do correlations between them. I’ve seen hundred of eCommerce implementations and I am yet to see one that fits exactly the reality.

Why isn’t it 100% accurate? Chargebacks, plugins that block Google Analytics, pages that load slowly and are abandoned are just a few reasons.

However, financial reports need to be 100% accurate. Don’t base your financial reports on Google Analytics as it will only cause frustrations.

How big of a frustration? I once had a fake payment of 2 billion screw up all my eCommerce data. A user placed an order of $2.000.000.000  on a customer’s website.  Luckily  we were able to delete the transaction from Google Analytics, otherwise the eCommerce data would’ve been compromised.

Google Analytics is not a stalking tool

No, you can’t use your visitors name, email address or security number as custom fields or events in Google Analytics. First, it’s against their terms of use. Doing it might get you banned from their service. It’s also illegal in many countries.

Second, it won’t do you any good. It’s like judging a huge crowd of people by just analyzing one person, or 2 or 10.

Google Analytics is not for analyzing individual behavior.  It’s for analyizing what impact your website is having on segmented groups of traffic.

Google Analytics is not a tool for calculating CTR

Ok, maybe in some particular cases it can be used for this, but most of the time you’ll get into trouble.  There is a Google Analytics limit of 500 requests per visit and a limit of 10 requests per 5 seconds.

If you are going to track every impression of every banner on your website, the chances are you are going to hit that limit of 500 requests pretty soon, and you won’t get the rest of the valuable data of your visitors behavior.

As conversions usually happen at the end of a visit session, you might loose the most important part of a visit.

Google Analytics can’t calculate the real time spent by visitors on your website

The time spent on your website by visitors (as it appears in  Google  Analytics), is always smaller than how much time people actually spent on your website, even as an average.

If somebody visits a single page on your website,  Google Analytics won’t be able to calculate how much time she stayed and will report 0 seconds,  even if she spent 3 minutes on it.

The real bounce rate hack helps you get a better estimate of how much people spend on your website, but still, it won’t give you an absolutely accurate number.

But does it really matter if people on average spend 5 minutes or 5 minutes and 40 seconds? Would you take a different optimization decision if you would get the real number? My guess is no.

If you send a request to Google Analytics every second or even every 10 seconds of somebody’s visit, you’ll just hit the 500 requests limit so fast that Google Analytics will fail to log other actions done by a visitor on a website, and the time spent will be even more inaccurate.

Google Analytics is not a product database

Ok, this is something I actually tried to do a couple of times, but it ain’t a piece of cake. I’ve seen cases where people use it as a product inventory database. While they considered the implementation a success, and I have to agree I loved it, in the end I think it looks better as an experiment to try out your skills than a real solution.

If you are a geek and really don’t have another solution for it, ok, give it a try. Just do it in a separate profile as you might impact the rest of the data in an undesired way.

Can you think of any other Google Analytics misuses?

5 Things You Shouldn't Do With Google Analytics by

1.100.000+ email leads collected with

PadiAct helps you get more email leads. More and better.

FIND OUT MORE
  • http://techpad.co.uk Matt Clarke

    Thanks for the mention on number five, Claudiu. I completely agree with you. I am a geek and I had no other option. That said, it works quite nicely when coupled with the API. I definitely wouldn’t recommend it on a profile containing web data, though.

  • http://padiact.com Claudiu Murariu

    To be honest, I love your blog Matt and I’ll never hesitate to link to it. From geek to geek, keep up the good work!

  • http://www.analyticspros.com Caleb Whitmore

    Great post and good points!

    I would say that you can get relatively accurate Time on Site and Bounce Rates if you implement a tool like http://www.analyticsengine.net or http://www.skyglue.com to automate comprehensive Event Tracking. For example, we track scroll depth as an event, outbound clicks, etc… using Analytics Engine on our site and find that many people do in fact scroll down the page, reading content, click an outbound link, but never visit a second page on the site. Without automated Event tracking, getting comprehensive scroll, outbound, etc… tracking is very hard. But, with it, your data becomes illuminated by a much greater degree of data clarity and accuracy around true engagement.

    Best,

    -Caleb

    • http://padiact.com Claudiu Murariu

      I agree with you Caleb. With such implementation, some of which I use myself, you get more accurate data, but it’s never going to mirror exactly the reality. And it shouldn’t even mirror it, actually. The data we get, especially through implementations like you suggest, is more than enough for an accurate analysis.

  • https://twitter.com/#!/Babatoure Baba

    Hello,
    Thanks for the post. I am really doubting about this :

    “If somebody visits a single page on your website, Google Analytics won’t be able to calculate how much time she stayed and will report 0 seconds, even if she spent 3 minutes on it.”

    I have always considered that people who visited a single page of a website had their visit duration mentionned directly when you report that single traffic source.

    My questions are :
    1 – Did you think about those (strange example) websites that only have one (single) page
    2 – what do you think about details mentioned here ? ( http://support.google.com/googleanalytics/bin/answer.py?hl=en&answer=60127 )

    • http://padiact.com Claudiu Murariu

      Thanks for the comment Baba. Google Analytics calculates time by seeing the difference in timestamp between 2 different server requests.

      When a visitor just visited a single page, and no event data is triggered, Google Analytics only has one request to use for calculating the time, thus the time being 0. That’s why in Content reports you will see some pages (long tail especially) that have time spent on page as being 0.

      Here is the extract from their official documentation about this:
      “When a page is the last page in a session, there is no way to calculate the time spent on it because there is no subsequent pageview. For this reason, when Page A is the last page in the visitor’s session, its time calculation is not counted for that view. In addition, when that page is the only page viewed in the session, no time on page is calculated.”
      http://support.google.com/analytics/bin/answer.py?hl=en&answer=1006924

  • http://www.davidcrowther.com David Crowther

    Instead of hacking GA to do everything it should be doing in the first place, why not use an analytics service that does what you need?

    I’ve been using Clicky (http://getclicky.com) and haven’t looked back.

    ~ David

  • Pingback: Quora

  • Pingback: Digital Analytics Week In Review - 26 April 2013 - Data Runs Deep