Note: this was originally published at https://alexoppenheimer.substack.com/
Generally accepted accounting principles were designed to create a single standard for comparing any and all businesses using the same terms and metrics. This is a very powerful tool, but it is a clear exchange for breadth over specificity. As a result, the economic reality of a business will likely differ from what the accounting metrics represent. By looking at a subset of business models, we can improve specificity - terms like ARR and GMV are both examples of this. While these terms have their own challenges with definitions, I think most agree what they represent: ARR is annualized recurring revenue for SaaS companies, and GMV is gross merchandise/marketplace value for marketplace companies.
One of my favorite examples of this is CAC - customer acquisition cost. It’s relevant for both SaaS businesses and marketplaces. Average CAC is the cost to acquire a single customer and Total CAC is the full spend for a cohort of customers. So what is the difference between CAC and Sales & Marketing expense? The truth is that Total CAC and S&M over a given period of time might be the same. That being said, the usually differ by a time offset that depends on sales cycle length.
The critical distinction is sales cycle length. Sales cycle is how long it takes for a customer to go from first contact to closing a deal. We measure sales cycle by segment and can use simple reports in Salesforce or another CRM tool to figure out how long it takes on average for customers to move through the sales funnel. Generally sales opportunities begin as leads, some of those convert to qualified for sales and then some of those convert to closed-won. The Sales Funnel shows the volume and conversion at each step in the process and is usually in the hands of the CRO or sales operations team. The two lengths of time that we care about in finance are: top of funnel to sales qualification (i.e. marketing cycle length) and then sales qualification to closed-won (i.e. sales cycle length).
There is a large variance in sales cycle lengths from business to business. For example, consumer software might have a <1 month sales cycle that runs entirely on marketing. On the other end of the spectrum, a complex sale to a large institution may take over a year and involve a number of different people and teams.
For the in-month sales cycle, New ARR in Month X will be attributed to the Marketing expense in Month X. Pretty straightforward.
As soon as you have a separate marketing funnel and sales funnel and things take longer than a few weeks, it becomes more complicated to measure, track and predict.
I often see operating stats that look like this:
The high level sales efficiency metric is New ARR / S&M: the idea being that we are investing in S&M in order to generate new ARR. The missing piece of information here is how long are the customers in the marketing funnel and how long are they in the sales funnel. Without that info, people often assume that it is in the same time period - whether that is a month, a quarter or a year. Switching over to a CAC calculation based on sales cycle length gives a more accurate picture of the sales efficiency.
For example: let’s say the marketing funnel takes 1 month and the sales funnel takes 3 months.
As a result, CAC would be the Marketing expense from 3 months prior and the average of the Sales expense over the last 3 months. So in the above example, new ARR acquired in April would be associated with the Marketing spend in January and the average of the Sales spend in Feb, March and April. Because this is a growing business, CAC will be lower than the S&M in the same month, meaning that efficiency numbers are actually higher when based on CAC. Quick note to operating executives: don’t use a basic formula and sell yourself short.
Now looking at it through the lens of sales cycle-dependent CAC:
Once you have the CAC line in the model, you can use it for SaaS Magic number, CAC payback, and a number of other metrics. (Of course, as the sales cycle changes, the CAC calculation must also change.) These metrics are important because they usually stay constant for a given product and business operation and allow you to project into the future.
The CAC calculation gets you one level deeper towards operating metrics from the high level financial metrics and is often a good bridge between the two. It represents the detailed funnels at a high level, and will ultimately drive projections with a higher degree of accuracy. For example, we now know that leads acquired through marketing spend in December will not yield sales until 3 months later in March. Based on historical metrics, we know roughly how many leads this spend will get us and therefore what kind of sales capacity we need to close on those leads and can plan sales hiring accordingly.
Coming from the financial perspective down into the operating perspective is a bit of a rabbit hole, so there’s an art in knowing where to stop, and it’s usually the point where precision starts to overtake accuracy in the model and digging deeper stops yielding better results. One step down from this, for example, would be to look into the sales expense - sales people are generally paid 50%+ of their total salary in commissions. If those commissions are paid in the month of sale, half of the sales expense should be weighted from that month and the other half weighted across the earlier months. If sales people are paid later on, then looking at it from a purely economic perspective as it pertains to commission rates will yield a much better result than looking at the accounting outputs for sales expense.