Tracking most valuable customers based on behavior

If you would get an opportunity to “bid” on the ingredients for you lunch meal, would you choose the cheapest possible? Probably not, as you are aware of the consequences and care for maintaining your body appropriately.
Good marketers that place emphasis on sales growth, are looking for the big fish and are willing to bid very high for second engagements, because these customers are the most valuable for their company.
The Pareto principle says that 20% of the customers contribute 80% of the revenue. Once a customer has placed a few orders it’s easy to rank him relatively with our other customers. We can then upload that data into GA (Google Analytics), then analyze and treat him appropriately as presented in the article Remarketing based on imported offline data.

What about our new potential customers? How do we differentiate between the big fish, small fish and those that are only “observing”.
One way of tracking these potential big fish is by glancing at their cart. The bigger their cart is, the more valuable they are. We can measure either the amount of items in their cart or the total revenue. We can also sum the amount of items that are on sale in the cart or the number of coupons.
After we have all of these figures we can target the biggest fish and be willing to place a very high bid for them to reengage.

So how do we do this?

I want to introduce the Custom Metric (see Figure A). This feature (only available in Universal Analytics) gives us the ability to segment users using operators such as “less than”, “greater than” and others.

Setting up a Custom Metric

Figure A: Setting up a Custom Metric

Every time a user adds to his cart we can fire the quantity into a CM (Custom Metric) and the revenue generated from the product, into another CM.
Every time a user views his cart we can fire the following into different CM:
Total amount of items
Amount of items on sale
Total revenue
Percentage of discount
Total of coupons

After we set all of our CM wisely, we can begin with building our audience’s lists for remarketing. Unlike the segments we built before, using user level and session level, in this segment we want to track a single “hit” that reached the sufficient amount for being included in the audience list, as CM is an aggregated metric.
For example, a customer viewed his cart while he had there 2 items:
Hit level = 2, Session level = 2.
During the same session he added another item to his cart and viewed it again:
Hit level = 3, Session level = 5 (=2+3).
He viewed a different page, then returned to his cart:
Hit level = 3, Session level = 8 (=2+3+3).
Obviously he does not have 8 items in his cart, and that’s why we present the rule using the hit level.

There is also a “product” level, which is only available in Enhanced Ecommerce implementation, this will be presented in another article.

Implementing the example of “Total revenue” in cart, means we want to target customers that have in their cart a total value of more than $200.
This will be done as seen on Figure B.

Segmentation of customers that have more than $200 cart value

Figure B: Including only customers that have more than $200 cart value

Back to the example of the customer that has 3 items in his cart, he suddenly decides to remove two items from his cart:
Hit level = 1, Session level = 9 (=2+3+3+1).
Is he still eligible to be served high bids of remarketing ads as he only has one item in his cart right now?
How do we exclude him from our list?
We’ll need to build a sequence segment (see in Figure C), then exclude this list in AdWords as explained in the article Advanced remarketing lists Excluding Audiences.

Segmentation excluding customers that removed from cart

Figure C: Customer exceeded $200 cart value, then he removed some items that lowered the cart value below $200.

Start the course now (free)