We’ve been tracking COCA for SaaS companies for about 5 years now. We use the metric to calculate benchmarks for customer acquisition expense by size of company, and by type of product offering, ie., whether a company is selling a low cost solution or a relatively high cost solution. We incorporate COCA into other benchmarks as well. For example, we can use COCA to look at whether customer acquisition expense for new customers has increased or decreased over the years, what the “pay back” period is for new customers, and what the customer profitability benchmarks are.
Calculating Customer Acquisition
One question I get asked a lot is how do you calculate “COCA?” Do you include all of sales and marketing expense? Do you include all of sales, and only some portion of marketing? For what period do you calculate COCA?
We recommend using all sales and marketing expense, trailing one quarter, to calculate COCA. For example, if you want to see your total COCA for 2010, then look at sales and marketing expense from Q4 2009 through Q3 2010. We do this based on the assumption that most SaaS companies have a sales cycle between 30-90 days. Some companies selling a larger, enterprise solution with longer sales cycles could look at sales and marketing expense trailing 2 quarters for the period being looked at.
We think that using all of sales and marketing expense to calculate COCA makes sense for small and mid-sized SaaS companies because most small and mid-sized SaaS companies are in growth mode, so almost all sales and marketing effort is expended in order to acquire new customers. With billion dollar companies, it would be fair to allocate some portion of marketing to overhead, general brand awareness and other marketing efforts, but with smaller companies, even several million dollar revenue companies, we generally find that it makes sense to assess all or almost all sales and marketing expense to COCA.
If you want to see COCA per new customer acquired, you divide the annual COCA by the total number of new customers added in that year. We suggest that you should use the net change in customers from one year to the next, ie., incorporating churn in the analysis, which gives you the true COCA for customers acquired that you could keep in a year. If you use the total number of new customers, and do not incorporate churn, your COCA will be slightly lower, but does not reflect the cost of acquiring paying customers and customers that you can keep.
Certainly some companies may define COCA slightly differently depending on their business model, but we have found over the past few years that more and more SaaS companies are using the general definition of COCA as all of sales and marketing. If you more thoughts on the subject, leave a comment below!