Analytics / Back Office Discrepancy: and what if you were wrong? Analytics / Back Office Discrepancy: and what if you were wrong?

Analytics / Back Office Discrepancy: and what if you were wrong?

by ,
on 9 March 2018

0
0

E-merchants often observe a discrepancy between orders shown in analytics and those shown in the back office.

This discrepancy is often the submerged part of the iceberg because it’s one of the only means of detecting problems with analytics data collection and feedback. With traffic metrics, a discrepancy is harder to detect.

So how do you identify the discrepancy between the analytics figures and those in your back office?

For starters, step 1: extract the back office orders with the most information possible. The more information you have on orders in the back office, the greater the chance that you will highlight the problem, or problems, if they exist.

Here is a non-exhaustive list of data to extract from the back office to analyze discrepancies:

  • order number or ID & amount (requisite minimum)
  • order date and time
  • order status
  • payment method
  • number of articles in the shopping basket
  • content details with the name of the articles ordered
  • devices
  • etc.

Be careful that there are a significant number of orders in your file to conduct the analysis. If you have only 15 orders a month, it may be difficult to draw a conclusion.

Now on to step 2: extract the order IDs from analytics.

And finally, step 3: align the back office and analytics IDs to estimate the discrepancy gap between the two sources.

Below 10%, you can consider that the implementation is working properly.

However, if you have a gap that’s greater than 10%, here are a few things to look at:

  • problem or buyer behavior that does not trigger the payment confirmation page
  • the request size may be weighed down by the number of characters or the information requested (this is not uncommon on foreign sites, for instance, Russian or Chinese sites as a result of the Cyrillic or Hànzi alphabets which add weight to the hit).

– the way in which the analytics tag was implemented; the tag position influences data collection and feedback.

A little tip for reliability: be attentive with javascript re-execution on mobile terminals which resends the order hit (See: Tips & Tricks https://blog.altima-agency.com/en/analytics_en/analytics-conversion-tips-tricks-4-duplicated-google-analytics-transactions/)

In conclusion, limit and monitor the discrepancy between analytics and your back office as much as possible. Note that analytics is a tool that has a tendency, by nature, to give and show trends that are not 100% exact.

Where you have a large discrepancy, it’s always possible to find a solution with varying degrees of complexity!

tags:

, , , ,

, , , ,

to read next...

Leave a Reply

Your email address will not be published. Required fields are marked *

0
0

Analytics / Back Office Discrepancy: and what if you were wrong?