6-tips for retailers on using web analytics for marketing decisions

Marketers have very specific expectations from web analytics. But the need of the hours is a high-quality single purpose tool – much like the Mora knife pictured above. A swiss army knife approach works well as a general tool but is a poor choice for specialized tasks. For retailers marketing their products on the Internet, a good web analytics service must deliver much more than tracking and reporting website traffic. The primary metric for retailers is sales, and a key tool for driving sales is online marketing. So it’s not just important to drive traffic but have a clear visibility on what’s driving sales for the business. With that in mind, here are 6-tips for retailers when using web analytics for boosting their marketing performance.

#1: Don’t measure vendors or campaigns in isolation

No campaign or vendor sits in isolation. Channels overlap in their influence on your shoppers, as do campaigns. Retailers must look at marketing performance at an aggregate level and apportion sales impact across all influencing channels. The example below from Polytab shows how the contribution across all channels adds up to the total number of orders. This is distinct from the contributions reported by the media companies (Bing, Google) and the email service provider individually – who often double dip on the sales.

Note also the ROAS (Return on Ad Spend) for Google Ads in comparing the sales contribution (in the blue line) to the ad spend (in the red line). Dashboard-Capture

#2: Don’t measure marketing effectiveness by the first/last click.

One way some tools measure effectiveness is by first or last click. First click means the channel that first brings a visitor to your website. The last click is the last channel that brings the visitor to the website at the time of purchase. Neither is a good choice. Every  relationship is a journey. To condense it to a single interaction is doing disservice to the complete relationship. Here’s an apt argument : first click attribution is akin to giving my first girlfriend 100% of the credit for me marrying my wife.

#3: Make sure your tracking pixel is uploaded asynchronously

If a pixel loads synchronously, it means that it is loaded at the same time as your site. Asynchronous loading means that your website gets priority over the tracking pixel. Back in 2009, Google Analytics took down 14% of all sites who used that tool because the pixel was set to load synchronously. So if the pixel was not loading due to a glitch, it kept the websites from loading as well. In online retail, having your website out of action is lost business. So make sure the pixel loads asynchronously. The downside is that if a visitor jumps to another page too quickly before your asynchronous pixel triggers, then you do not log the visit. That’s totally OK. Chances are if someone’s just doing a fly-by, he/she was not a genuine visitor.

How to check when your pixel loads and how long it takes (on Chrome)
1. Open the Developer Tools from the menu (Menu > Tools > developer tools), or by pressing Ctrl+Shift+C on your keyboard.
2. Click on the Network tab

#4: Track cross-device behavior

Ensure the service is able to track multi-device shopping journeys. So if the customer researched the retail website on her handheld but completed the purchase on the desktop, the journey can be linked across all platforms.

#5: Measure the influence of website visits to sales (without measuring sales)

Not every website visit must convert to a sale, but the sale down the road may be attributed (in part) to the preceding site visits. A customer relationship spans acquisition, conversation and purchase triggers. Most legacy web analytics services focus on the purchase triggers – the discount offers, the coupons, the emails etc. But the relationship is built over the conversation phase. Make sure you have a means to monitor and measure the impact of channel conversations. A next generation web analytics service should be able to attribute impact of

  • Social shares, likes, comments,
  • Display retargeting impressions,
  • Video plays and likes

#6: Separate out the sales baseline

If a retailer were to shut off all digital marketing, it would still continue to get sales – through organic engagement, repeat purchases, seasonal behavior. The visitors will find their way to the website or to the brick-and-mortar stores regardless of the marketing push. Make sure that the web analytics service can separate out that baseline. Look for a service that uses statistical modeling or econometrics to parse out the macro-trends.

A specialized, single purpose web analytics toolkit for marketers

With a view to all the key requirements listed above, and knowing the shortcomings of the freemium offerings in the space we built the Polytab utility. For a first hand understanding on how Polytab works for retailers, sign in as a guest to get the users view.





Photo by erik forsberg Attribution-ShareAlike 2.0 Generic (CC BY-SA 2.0)

Leave a Reply

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