Hack82.Measure the Shopping Cart

Hack 82. Measure the Shopping Cart

For online retailers, there is perhaps nothing more important than measuring visitor flow and abandonment through the shopping cart. Your use of metrics and measurement in these pages can make or break your business.

In retail web analytics, measuring the shopping cart is analogous to gazing into a crystal ball. Each visitor interaction with the shopping cart is a measurable activity that generates data useful for sales and demand forecasting, promotion planning, and shopper behavior analysis. This hack introduces measurement approaches for key shopping cart interactions and also discusses factors that can degrade the accuracy of shopping cart measurement.

6.3.1. Measuring Products Being Added to the Cart

Think of this interaction in terms of distinct products (line items or SKUs), item quantities, and visits or visitors. More importantly, think of each of these metrics in relative terms to arrive at meaningful conversion rates for the add-to-cart action, often termed shop action and expressed as a percentage:

• Item-Based Shop Action Conversion = Items Added to Cart / Total Item Views (for a specific product)

• Visit-Based Shop Action Conversion = Visit Adding Items to Cart / Total Visits

• Visitor-Based Shop Action Conversion = Visitors Adding Items to Cart / Total Visitors

As an online retailer, one of your primary objectives is to drive shop action conversion through effective product selection, presentation, pricing, and messaging on availability, shipping costs, and other handling and delivery options. Shop action conversion rates express the performance of your site at achieving these objectives, so it is essential to actively monitor them.

6.3.2. Measuring Products Being Abandoned in the Cart

Removing items from the shopping cart results in changes to item quantities or complete abandonment of the cartneither outcome is welcome! But when measuring abandonment, it is important to distinguish between two related abandonment metrics: item and cart.

• Item abandonment measures the number of times a specific product or SKU is left in an abandoned cart and not purchased.

• Cart abandonment is the overall measurement of abandonment for all carts during a time period.

Misusing these metrics can make the abandonment issue seem to be a much bigger or much smaller problem than it really is. Table 6-1 illustrates the importance of both abandonment metrics.

Table 6-1. Illustration of how item and cart abandonment can affect your aggregate conversion rates

	Cart A	Cart B	Aggregate
Items added to cart	6	4	10
Items purchased from cart	3	0	3
Item abandonment	50%	100%	70%
Cart abandonment	0%	100%	50%
Cart conversion	100%	0%	50%

Cart A does convert but not at a true 100 percent. Over half the items placed in Cart A were not purchased, so a significant sales opportunity was lost. Cart B failed to convert, as a total of four items were added but none were purchased. Item abandonment and cart abandonment for Cart B were 100 percent, resulting in both a lost sales and customer acquisition opportunity.

In general, when we talk about "shopping cart abandonment," we're talking about cart abandonment and when we're exploring look-to-book ratios [Hack #81], we're exploring the nuances of item abandonment. The interesting thing in the previous example is that only 50 percent of the carts completed and in the one cart that did convert, three products had been removed from the cart prior to purchase, lowering their individual look-to-book ratios.

6.3.3. Measuring Maximum Shopping Cart Value

Given the misleading signals that a single shop action or abandonment metric can send, it is important to consider all of the metrics that may be used to describe the state and performance of your site's shopping cart.

Maximum shopping cart value (the maximum dollar value of items added to the cart during a specific time period) is a useful metric to trend, as it may expose spikes and lulls in online demand as a result of the launch or decay of a recent marketing campaign, such as an email blast or major keyword buy.

Trending maximum shopping cart value against online sales (Figure 6-2) enables you to instantly gauge the size of the abandonment problem in dollars, the most useful measure of all for a web retailer. The area between the two trended lines represents an enormous sales and customer acquisition opportunity.

Figure 6-2. Online sales (bottom line) versus total maximum cart value (top line) report

6.3.4. Measuring What Matters in Shopping Carts

To summarize the hack so far, the list of measurements you need to make to accurately understand how shoppers are using your shopping cart includes:

• Total number of items added to the cart

• Item-by-item list of products and SKUs added to the shopping cart

• Dollar value of all items added to the shopping cart, even if they're abandoned or removed

You then want to be able to follow each of these measurements through each step in the checkout process [Hack #83]. Again, you're gathering this information to calculate the following:

• Item-Based Shop Action Conversion = Items Added to Cart / Total Item Views (for a specific product)

• Visit-Based Shop Action Conversion = Visit Adding Items to Cart / Total Visits (similar to order conversion)

• Visitor-Based Shop Action Conversion = Visitors Adding Items to Cart / Total Visitors (similar to buyer conversion)

• Shopping Cart Potential = Total Dollar Value of All Items Added to the Cart

Because of the complexity involved, you're much better off if your measurement or commerce application will make these calculations for you, but getting a good handle on these metrics now will save you time and trouble when you get deep into diagnosing problems with your merchandising and your checkout process.

6.3.5. When Is a Shopping Cart Not a Shopping Cart?

Sometimes a shopping cart is not used as you intend. For example, visitors might use a shopping cart as a wish list, or the shopping cart may simply persist over time, yet never complete a sale. These two factors, which can obfuscate your analysis of the shopping cart, are worth understanding.

6.3.5.1 The cart as a shopping list.

Sometimes visitors use the shopping cart as a shopping list, such as a personal wish list or gift list for an upcoming birthday or holiday. This activity may significantly skew same-session shop action rates by artificially inflating the number of items added to the cart (because there was no real intention to purchase the items in the "wishing" session).

It is difficult to know when this activity is happening without analyzing shopper cross-session behavior, but you should be aware of it, particularly if you sell products that potential buyers prefer to research extensively before purchasing. High-ticket items that may require a spouse's or parent's approval before being purchased are also often repeatedly added to the shopping cart and revisited over multiple sessions prior to purchase, which may take place offline.

6.3.5.2 Persistent shopping carts.

Many online retailers implement persistent shopping carts, a move that often exacerbates the skewing effects of using the cart as a shopping list. However, persistent carts do present an opportunity to extend the definition of shopping cart conversion (and abandonment) to include activities that occur across multiple sessions. In-depth analysis of cross-session activity often yields valuable insights into the online research and buying patterns of your customers.

If your site uses a persistent cart, be aware of its impact on your web analytics solution's measurement of abandonment. Most solutions define shopping cart conversion and abandonment in same-session terms, which may overstate the size of the abandonment problem for retailers using persistent carts.

Brett Hurt and Eric T. Peterson