I wrote about this on my blog a few weeks ago but thought it was worth sharing with our DigitLab clients. We often have clients asking why the traffic stats they get from google are so different to the traffic stats they get from their banner advertising service providers.
It is important to understand how Google Analytics works.
Google Analytics only measures the click from users who can accept a cookie on there computer system. GA also only measures the click once the entire page is loaded because the code is placed in the footer which last section loaded.
This means that clicks from referral (advertising) websites may be measured differently. IE: The advertiser will measure the clicks on their website of the advert. However they cannot measure how many of those people leave the advertised website before your website page has fully loaded.
This means that there will be a variation between the stats that your advertiser sends you and what we see on GA. This difference will be the people who don’t make it to the website for a variety of reasons:
- They clicked the advert by accident
- They thought the advert was for something else
- They did not realise the banner would take them to another website and then chose to close the window or click the back button
- They saw enough information on your website to decide that they did not need the information – before the page had completely loaded
- They got tired of waiting for the page to load due to a slow internet connection or un-optimised website code.
- The clicked on the advert from a mobile phone and decided to leave the site once they saw it was not designed for mobile. (leaving before the page loaded fully)
What this means to your advertising partners
I do think that you advertising partners are giving you the right stats. Most likely they are not lying to you or trying to rip you off. The problem is not the supplier analytics or the GA stats but rather the consumer. The user of the internet pulled out of the lead and this happens often on the internet for a wide range of reasons (some of which I explained above.)
You can cut down the level of variance by finding advertising partners that engage audiences that are similar to your audience. This is difficult to do because we mainly choose to segment by age rather than personality and each personality treats the internet differently. Before I get carried away with this you can read more about the digital lifestyle framework on my blog.