eCommerce SaaS Economics - The Flaw of COCA:ASP

by Sam Mallikarjunan


June 6, 2013 at 7:56 AM

eCommerce SaaS EconomicseCommerce SaaS Economics

We once had an eCommerce marketing team come to us with a confusing and devastating problem: their competitor was massively outspending them per click in Google search engine advertising, which both this team and their competitor relied on to drive the majority of their business.

The team’s members were very confused because they knew that they had the same product supplier and roughly the same cost of goods sold (COGS). They also knew that their competitor’s website had roughly the same visit-to-customer conversion rate. There was no way that they could think of that their competitor was able to afford to spend so much to acquire each customer.

To explain this to them, we told them about a quote from one of our favorite eCommerce marketers, Matt Lauzon, at Gemvara.

Lauzon’s prediction is that effective eCommerce marketers will very soon have to think of eCommerce revenue models from the perspective of SaaS (Software as a service) economics, where the cost to acquire a customer is only one-half of the economic equation—and arguably not the most important half.

The Flaw of COCA:ASP – and How to Fix It

SaaS companies express their unit economics using the ratio formula COCA:LTV (cost of customer acquisition to long-term value). It’s the balance of the numbers in this ratio that defines the economic value of the enterprise. Since most SaaS firms are on monthly recurring revenue (MRR) models, the average LTV is a simple function of their average sale price and their revenue churn coupled with their ability to upgrade customers along whatever their secondary axis of pricing is.

However, most eCommerce firms express their unit economics using COCA:ASP (cost of customer acquisition to average sale price). This ratio formula is fundamentally flawed and overly simplistic. Although most eCommerce marketers understand and acknowledge intellectually that cross-selling, upselling, and reselling to existing customers is important, very few of them model or bake this fact into the way that they consider the unit economics of their business.

For our customer’s competitor, however, we were able to identify how they were using social media and e-mail marketing automation to very effectively stimulate repeat and recurring purchases from their existing customers, thereby vastly increasing the LTV of their individual customer contacts. Whether or not they were doing this on purpose, they’d effectively out-leveraged our customer, who was very focused on customer acquisition and price competitiveness, in the grand equation of COCA:LTV unit economics. Quite simply, they could afford incredibly slim—or even negative—profit margins on the individual transaction because they were being very effective at increasing the LTV of their individual customer contacts.


Long-tail eCommerce firms like Amazon have a fundamental weakness here because the best way to improve customer LTV is through content and engagement. Amazon, like many eCommerce marketers, relies on the somewhat lazy fallback of discounts and price incentives to drive buyer behavior. The traditional eCommerce method for extracting value from a list of existing customer contacts is to blast the list twice a week with special discounts and sales. This method is dangerous because it has the potential to turn customers that aren’t already highly price sensitive into ones that are by framing the entire conversation from the perspective of competitive price. Also, it fails to address the fact that consumer buying cycles aren’t a linear continuum. In fact, consumer buying cycles are a dynamic function of the many products and accessories that are related to their buyer personas, which we’ll discuss in-depth in a later chapter.

Amazon, because of its diverse inventory and product offerings, has a massive challenge in identifying and nurturing micro personas in a highly targeted and effective way. They’re fairly good at identifying related products or offering product suggestions based on historic purchase patterns, although they’ve been known to fail at this rather spectacularly as well. For example, Amazon sent an e-mail to a user recommending they buy a book called “Crafting With Cat Hair” – even though that user neither owned a cat nor was particularly crafty. While the specific factor of their algorithm triggered that suggestion isn’t clear, a real human would never have recommended that particular item to this user if they’d had a micro-persona defined for him.

However, they’ve historically avoided entirely using any kind of related-content nurturing in e-mail automation (much less social media nurturing) that helps address the dynamic buying cycles involved in selling interrelated or upgraded products. The sheer scale of what they’re selling makes it implausible that they’ll be able to compete with a vertically oriented eCommerce marketer who truly understands all the dimensions of his or her buyer personas and has an inherent expertise and experience in the factors and topics related to their product categories and accessories.


Click To Buy The Full Book Learn More About eCommerce Unit Economics  & How To Improve COCA:LTV

Download Free Chapter!

Buy Kindle eBook Buy Printed Book

Subscribe to Blog:

Recent Posts

Get Your Free Chapter!