In his famous piece, The Art of Evangelism, Guy Kawasaki writes “Look for agnostics, ignore atheists”. This is particularly relevant to my recent blog post describing the harrowing experience of Open Source BI (OSBI).  In it, I point out that OSBI does indeed work for a vast number of very specific users, but most OSS-minded readers skipped right over that statement and interpreted my post as a blind bashing of OSS in favor of SaaS BI. Nothing could be further from the truth.

For if my intent were indeed to “convert” OSBI users to a SaaS approach, it would make me naive and likely frustrated. Because I strongly believe in Seth Godin‘s golden rule that you cannot change people’s worldview. Attempts to do so, especially in the technology realm, have too often proven futile. Especially since Open Source, in my opinion, is really more of a religion than a technology per se. You either have the faith, or you don’t. And these are not waters I wish to fish in.

No, what I am after are the people who are still on the fence between OSBI and SaaS. The decision makers still trying to map out the “right” BI strategy. The developers, consultants, business people and C-level leaders who seek a roadmap to what platforms and approaches are good long-term business investments. In these matters, technology is but one of many factors. At stake: job security, competitive advantage, and long-term business survival. Let me see if I can provide a little direction.

My first inclination is to describe typical users in each community. Because, if the profile fits your own organizational or personal traits, then chances are this is your camp. Here’s what I think the profile of a successful OSBI and SaaS BI user might look like:

BI USER PROFILE OSBI SAAS BI BACKGROUND Highly skilled software developer. Years of professional experience Highly skilled, but not necessarily from IT TECHNOLOGY PREFERENCE Likely Linux/Java or LAMP stack developer Language and platform agnostic LEARNING PATTERN Quickly reads and absorbs vast quantities of information No time or bandwidth for steep learning curves PROBLEM SOLVING Significant hacking and debugging skills Tinkerer PERSONALITY Self-starter, fend-for-yourself mentality Self-starter but collaborative as needed PAIN TOLERANCE Highly persistent and courageous. Willing to take the pain Tenacious but low pain tolerance BI EXPERTISE Medium to expert None to medium PHILOSOPHY “Get it right” “Get it done” TECHNOLOGY VISION Technology lover Technology just a means to an end BI USE CASE Often spills into the 20 of 80/20 Usually stays within the 80 of 80/20 SUPPORT EXPECTATIONS Relies on OSS community support and Google Expects dedicated real-time vendor support BI-SPECIFIC WORKLOAD Works on BI full time BI just one of many hats worn

 

Based on this, it’s fairly easy to recognize one of GoodData’s most common user type. Mind you, it’s not the only user type able and willing to use GoodData, but it is one we see in many of our customers. Internally, we call this persona “Norbert”. Norbert wants to provide reporting, analytics and data services for his customers. His customers can be either internal (same organization) or external (a consulting assignment). Norbert often shares many goals and traits with software developers (I can attest to this having been one for twenty years) namely, pleasing his end users, enhancing his career (or business) through enriching experience, and controlling his environment. Pushing a little further, Norbert is really not so far from your typical SaaS vendor or data provider (ISV) either (and often works there).

As it turns out, these three basic goals - end-user focus, career enhancement and environment control - are essential to healthy, productive and successful BI development. Unfortunately, OSBI only typically meets one of the three because it clearly offers a rich technical experience leveraging numerous tools and techniques. As such, OSBI can indeed be a career-enhancer. But sometimes, end users become disappointed due to over-hyped expectations, unforeseen technical hurdles, and delays. And controlling one’s environment in an Open Source soup of technologies, platforms, tools and communities can be challenging at best.

So if you are a BI implementer on the fence between OSBI and SaaS, you must ask yourself this question: I am the cowboy, commando type, or more of a “get it done” practical Norbert? And if you’re a business executive looking for a long-term BI strategy, you must choose your path wisely. Because it takes different kinds of people and skill sets to succeed in each model. Who’s on your BI team?

In the end, the challenge isn’t about deciding which is “better”, Open Source or Cloud. Rather, it is about determining which approach fits your organizational (or developer) DNA better. There’s a lot of hype in both camps out there undoubtedly. But it takes a healthy doze of realistic self-analysis and good, informative insight to steer the right course. To that effect, the following links might be a good start. Happy sailing.

Open Source:

Open Source BI Adoption on the Rise

Goodman on BI

Who Is Using Open Source Business Intelligence, and Why

OSBI Forums

SaaS BI:

Has SaaS Killed Open Source?

BI SaaS Vendors Are Not Created Equal

Business Intelligence: PivotLink v SAP, a tale of two worlds

SaaS BI Continues to Draw Interest, Traction