Reasons Why Shopping Cart Metrics Are Wrong or Not Accurate

I am writing this article because I think a lot of the shopping carts are very unique for every sites, and based on my experience in dealing with web analytics tagging for shopping carts, there are always issues.

What I need to make clear is that this article is not intended to discuss how to make your web analytics tagging perfect. It could be nearly accurate to address your data needs in areas of high priority, but everything can not be tracked perfectly accurate in 100% manner. You have to accept that truth. Also web analytics helps you find actionable insights and trends, and the main objective is not to use it as a report for reporting accurate numbers.

Shopping cart is one complicated beast. Some eCommerce sites may have their shopping cart handled by third party applications or vendors, complicating web analytics tagging delivery. Also there are many if and else conditions so web analytics implementation may not handle every little single data capturing instances.

Let me list out some possible shopping cart tagging/tracking/data challenges, and see if your web analytics handle such instances.

• Item addition and removal. You may take cart add into consideration, but not car removal.
• Login session considerations. Some carts require login after an item is added to cart or some time before the confirmation. This dependencies may throw off your tagging exepectations.
• Page refresh on order confirmation page (thank you page). If your analytics software counts conversions based on pageviews or don’t dedupe redundant sessions on confirmation page.
• Campaign variables not being pass along to cart, or conversions not attributing properly to campaign tagging.
• Search Engines indexing shopping carts with items in it, which could inflate unnecessary cart add. Not really a tagging issue, but felt like mentioning it here.
• Any optional activities and trackings required to address those usage. Example, promotional code, shipping calculations, billing info update, edit profile, etc. Anything in such nature may cause page refresh, and could be a potential cause of data inflaction.
• Different combos of quantities, products, special offers, etc. Depending on how these instances are handled, item, quantity, and revenue may not sync up right. Example, a user buys two pens with special promo of buy one get one free: You could tag it so it reflects two orders of pens applying average price on each, or two orders of pens with one full price attached to only one of the order, or handles it as one pen order and special offers tracking in separate bucket. The key is to have your data expectations aligned to what is really getting tracked.

The most important thing is to have your analytics planning well defined in advanced, and QA properly. Learn the trade offs, and be flexible about it. Do not lose focus on addressing the desired outcomes!!

Connecting the dots between site objective and desired outcomes could be done in multiple ways, and do not expect to answer every possible scenarios including stuff that is okay to know. Focus on NEED to know and ACTIONABLE KPIs.