When news broke on Friday that procure-to-pay vendor Coupa has filed IPO papers, SaaS investor Jason Lemkin, who organizes the SaaStr community and conference for software-as-a-service CEOs, tweeted a teasing rebuke:
Coupa is doing 69% YoY growth at ~$140m-ish ARR
What's your excuse for not growing faster? (sorry) https://t.co/R1ACxrY8ZE
— Jason M. Lemkin (@jasonlk) 9 September 2016
The target of the tweet wasn't Coupa but those observers who have commented that Coupa's $84 million revenues seem low for a company planning to IPO. Lemkin's point was that historic revenues don't tell the whole story when valuing companies like Coupa, which sell their cloud software to businesses on subscription contracts. When they're growing as fast as Coupa — 69% year-on-year growth — then the figure that matters is the total value of annual subscriptions as of the latest quarter, known as the annual recurring revenue (ARR).
The SEC doesn't require companies to report ARR, so you won't find the figure in Coupa's S-1 filing. But Lemkin estimates it at around $140 million, which leads to the tongue-in-cheek question in his tweet. Losses of $46.2 million may sound like a lot, but there's a school of thought that would argue it's quite a conservative number for a B2B SaaS business that's growing as fast as Coupa is.
That school of thought had its most public airing when cloud collaboration vendor Box was preparing to IPO. It all revolves around something that Lemkin at the time referred to as Box’s magic number, which led him to predict that Box would not merely thrive, but it would reach a $1 billion revenue run rate by 2019. (For the record, in its most recent results Box is forecasting $400 million revenue for the current year on growth of 30%, which leaves Lemkin a couple of $100m wide of the mark but still in the ballpark).
The magic number is a concept that's been around in SaaS circles for almost a decade. There are various ways to calculate it, but essentially it's the ratio between what you spend on sales and marketing in a quarter versus what you add that quarter in ARR. So let's do a quick back-of-the-envelope calculation for Coupa.
If we look at Coupa's S-1, we see that in its most recent quarter it added $2.4m in recurring revenue, which if we multiply by four to annualize it, implies a $9.6m increase in ARR. To achieve that, the company spent $19.3m in sales and marketing, so a 2x ratio (almost exactly, BTW, which may not be a coincidence if Coupa's management are monitoring this metric as closely as I imagine they are).
How does that compare to Box? Well, when Box filed its first S-1, early in 2014, another expert on SaaS business metrics, Zuora CEO Tien Tzuo, took a look at Box's figures and found that it "spent over $2 to acquire $1 of growth," at a time when its ARR was an estimated $148 million, very similar to Coupa's today.
So on this metric, Coupa is pretty much in line with Box. And like Box, its magic number is getting better, not worse. Of course there are other metrics to look at, which Tzuo goes into in his article on Box, but for the sake of brevity, let's assume Coupa's management has made sure these are on track too. [Update: for more detailed analyis of the S-1, see Dave Kellogg's blog today].
Selling to enterprise
The other factor that's worth emphasizing is that Coupa has a much more narrowly defined target customer than Box, which sells to smaller businesses as well as the larger enterprises that it's been targeting more recently. Coupa's customers start at 1,000 employees and are often much larger. Unlike Box's file sharing platform, its procure-to-pay processes, traditionally part of the classic ERP suite, are core to business operations right from the start.
All of this means that Coupa's product takes a lot more selling — but once it's in, it's very sticky. This, by the way, is why companies like Box and Salesforce, whose products do still appeal to smaller businesses, focus their most lavish sales efforts on larger enterprise deals. In Salesforce's case, the lifetime customer revenue effectively means each dollar spent to sign an enterprise customer is worth twice as much as a dollar spent to acquire a small business customer.
Finally, persuading enterprises to transform their existing procurement, contract and settlement processes by moving to a new, cloud-based application suite takes a lot of effort. Nobody wants the disruption, everybody takes the attitude that, 'If it ain't broke, don't fix it.' So it takes a lot of investment of time and energy to overcome that inertia and close each sale. It may not be fast, but the payoff should be worth it once the deal is done because in most cases the same inertia applies to the new incumbent.
At Coupa's annual conference in May, I collected a couple of stories from people with experience of buying Coupa solutions that demonstrate how much persuasion it takes. In fact, I started writing this post with the intention of recounting those stories. But it's taken me so long to set out the ARR background — which is important to understand — that I'll defer the stories to two separate posts. [Updated to add:] Both posts are now live. The first tells the story of the battles fought when Juniper Networks adopted Coupa, while the second sums up the five breakthroughs that vendors like Coupa must achieve to get enterprises on the path to transformation.