SaaS economics push Salesforce towards bigger deals

Phil Wainewright Profile picture for user pwainewright December 17, 2014
Salesforce is focusing on high-end enterprise sales because it believes the SaaS economics of these longer-lasting contracts drive better margins.

© Olivier Le Moal -
Salesforce is targeting large enterprise accounts. That was the overriding takeaway from last week's Analyst Summit, which brought a few dozen industry analysts together in Menlo Park, California, to hear about the vendor's technology and go-to-market plans.

While Salesforce still touts the "democracy" of its offering — small and midsized businesses (SMBs) make up the bulk of its 150,000 customers — it's clear the company is currently directing its efforts to building up sales to large enterprises.

In a presentation on SaaS economics by John Cummings, VP investor relations, the most telling metric was a comparison of attrition rates and the consequent impact on lifetime customer value. This proved to be the key to understanding why Salesforce is trumpeting its growing tally of seven- and eight-figure contract wins.

He revealed that the median attrition rate — the percentage of customers that stop subscribing each year, for whatever reason — is one in every ten across the entire customer base. But the figure varies dramatically according to size of customer.

Larger customers stay longer

The average attrition rate for enterprise customers is around half that of the smaller 'commercial' customer base. The larger the company, the more likely it is to remain a subscriber once it has signed up.

Translated into the length of time Salesforce can expect a customer to remain on the books, this skew in the attrition rate suggests that a large enterprise customer will typically stay several more years than the 10 years implicit in a ten percent churn rate. Whereas a commercial customer will typically have gone away before the ten years are up. Or as Cummings summed up:

Not every dollar of annual contract value is created equal.

In other words, each dollar spent to acquire an enterprise customer (a metric known as the 'cost to book') can be thought of as worth twice as much in lifetime revenue as a dollar spent to acquire a small business customer.

Therefore Salesforce believes it's worth spending extra to acquire those high-value customers. In the past couple of years, that spending can be seen in the recruitment of company president Keith Block and the team of high-performing salespeople he's added; the launch of a dedicated 'pod' for US government customers; the initiative to develop and market vertical industry solutions; and the creation of the Ignite team of business transformation consultants, tasked with helping customers develop ambitious new projects. These have all added to costs but are designed to increase the number of seven- and eight-figure deals the company lands.

As Salesforce's continuing growth takes it up the ladder of the world's top ten software companies — another billion or so added to annual revenues will bring it into the top five — it increasingly compares its performance to that of SAP and Oracle. It's a comparison that leaves it wanting more. Of its 150,000 customers, said Cummings, just 800 pay more than $1m per year. The top 100 customers account for 20 percent of total revenue.

Cummings expressed dissatisfaction with these levels of enterprise penetration compared to its more established rivals, revealing the hunger within Salesforce to up its game at this end of the market.

Margins rise when sales slow

Cummings' presentation perhaps goes some way towards explaining why Salesforce continues to post losses despite the cautions expressed by some observers, including those published just as the summit began by my diginomica colleague Dennis Howlett in his two-part analysis of SaaS economics.

Cummings dug into the framework of thinking that reveals why Salesforce believes those losses are manageable in the event of a market slowdown. The key metric here is the ongoing 'cost to serve' each customer. As sales decrease, the commissions paid to salespeople are expected to reduce proportionately. Since this 'cost to book' is so much higher compared to a customer's annual contract value than the 'cost to serve', the framework suggests a corresponding uptick in operating margins when growth is slowing. Cummings said the 'long-term' operating margin "at maturity" under the framework is in the mid-30s percentage range (based on Salesforce's usual non-GAAP financials).

Even more importantly from the point of view of the stock price, said Cummings, operating cash flow also rises as margins rise, and at some point becomes higher than the slowing rate of revenue growth. This, he told me, is why the use of stock options as part of employee compensation would not be a significant source of exposure in the event of slowing growth, as cash flow is what matters to investors, he said. The inference one takes from that is that the stock price would hold up, although of course that's never guaranteed. In any case, he added, there isn't a direct financial correlation between cash salaries and stock incentives.

My take

I'll leave it to Dennis and to others better qualified than I to pass comment on the financial metrics outlined by Cummings. Though given the lack of precedents for businesses of such size relying almost entirely on subscription revenues, any analysis in the end depends on subjective judgement at least as much as on objective fact.

I'm more concerned by the implications of such a heavy focus on large enterprise deals. One of Salesforce's strengths, it has always seemed to me, has been its appeal across-the-board to organizations of all sizes. This encourages Salesforce to ensure the growing complexity of its offerings doesn't get in the way of maintaining ease of adoption and use.

Much of the product development, however, that was presented at last week's Summit is high-end in nature, whether that's the Wave analytics cloud, the Marketing Cloud's JourneyBuilder or the industry solutions. I'll be writing in more detail on some of those in future posts, by the way. For now I'll simply observe that there was very little in all of this for the SMB sector.

The financial metrics are clear enough — it doesn't make sense to invest a lot of dollars in acquiring customers that may only be around for a few years. But to me that's an argument for investing in more automated ways of bringing them on board and servicing their needs — especially in reducing those somewhat high attrition rates.

In Wave, the company has a powerful analytics engine it can bring to bear on this problem. Those customers may have lower value, but they're still important — and a handful of them will grow into tomorrow's large enterprises. It's better to keep them in the Salesforce fold than let someone else build a rival customer base here.

Disclosure: Salesforce, SAP and Oracle are diginomica premier partners and Salesforce paid my travel costs to attend last week's analyst event.

Updated Dec 18th and 19th following clarification from Salesforce of some of the content of the presentation.

Image credit: © Olivier Le Moal -

A grey colored placeholder image