We are witnessing a dramatic change in the way SaaS businesses are managed. While SaaS financial metrics, such as recurring revenue, acquisition cost, service cost, churn, growth and lifetime value have dramatically increased our understanding of the economics of SaaS businesses, they have proven inadequate for managing them. As useful as they may be, SaaS financial metrics look at the past, not the future. They can tell you that you have a problem with churn, but they cannot tell you what you should do about it. Motivated by the need to better understand churn, many SaaS businesses have been independently exploring a new class of customer success metrics and have begun to embed them in SaaS customer success workflows with the hope of preventing churn before it occurs. We are witnessing the emergence of The Metrics-driven SaaS business.
Two weeks ago, I kicked off a new blog
For those of us that work in SaaS, we feel the customer success metrics pain when we try to bend Web marketing tools like Google Analytics, Marketo or Eloqua to the purpose of SaaS customer success. Unfortunately, they are just not up to the task. They don’t integrate the critical subscription, product usage, and account engagement data required. More importantly, they don’t have the necessary analytical power to enable the Metrics-driven SaaS Business. At best they supply simple historical reporting and heuristic scoring systems that have little basis in reality. There is no short-changing the math. If you want real predictive analytics; you have to use real statistical methods.
Real Stats. Real Easy.
I started my software career at SPSS, a very successful Chicago-based software company acquired by IBM. SPSS made predictive analytics under the tag line: “Real Stats. Real Easy.” The reason I bring this up is that I think a lot of folks believe that real predictive analytics is something that is hard to do and even harder to apply to everyday business. Well it is hard to do, but it can be very easily applied to everyday business. A SaaS customer success manager doesn’t need to know a system is using logistic regression or survival analysis to produce health scores and churn alerts. She just needs to see the red light go on and get the alert in time to keep a customer from churning. Plus, statistical visualization methods can be incredibly intuitive and powerful, providing the ability to zoom out to see high level root causes and drill down to investigate account-specific issues.
The Bluenose platform uses real statistics to create SaaS customer success metrics,
root cause analyses and predictive analytics that enable fact-based churn reduction.
I was lucky enough to get a preview of Bluenose back in November, and they got two things right that I have been waiting to see a long time: powerful statistical visualizations and predictive analytics. Of course the system has the baseline SaaS customer success capabilities, such as work flow management, surveys and broad data integration, but these are just a SaaS customer Success ERP module without the right analytics. Simple heuristic scoring systems just doesn’t cut the mustard. SaaS customer success metrics need real stats, real easy.
I think the Bluenose management team gets the potential of predictive analytics for SaaS customer success, because their roots are in serious big data analytics, specifically anti-virus software. That and I know they interviewed more than 50 potential customers in the development of their requirements. Having announced a significant pre-launch $11 million A round in December, the production release is imminent and the company is actively seeking and working with pilot customers. Anyone who follows my blog knows that I don’t do advertising and I rarely give product recommendations. For me, blogging is a labor of love, not commerce. The reason I am collaborating with the folks at Bluenose is that I think they get it, and I want to see the Metrics-driven SaaS Business become the standard in our industry.