Do not rely on the memory of your analysts

In the last part in the your analytics data should resemble usage series, I want to point out how manually tagging forces your analysts to compensate.

When you manually tag, you decrease predictability.

Having your analyst search to find how where a certain event happens is expensive.

When you have analytics that doesn’t match with your site, you’re relying on what your analysts remember.

They have to remember…

  • What action produced an event
  • Where the action is on the site
  • What steps there are before that action is taken

For example, let’s say you have the conversion event “Buy”.

Is “Buy” a button being clicked? Or a real purchase being logged? If it’s a purchase, when are we logging it? Is it on the order confirmation page?

This ends up with your analysts creating as few events as possible, so they can remember them over time.

And when an analyst leaves, you have created an information gap in the analytics team.

A handover may be enough, but you can’t expect everything to be contained in the handover.

Therefore, having analytics resemble usage is the easiest way to remove unwanted risks in your analytics.

– Samy

Component Analytics daily

Every weekday, I publish about Component Analytics and analytics collection strategies.