In-Depth

Pentaho Beefing Up Enterprise Support

Open source business intelligence player Pentaho has had a lot to crow about.

• By Stephen Swoyer
• 02/28/2012

Open source software (OSS) business intelligence (BI) player Pentaho Corp. has had a lot to crow about thanks to releases of new versions of its flagship BI and business analytics offerings, Pentaho BI 4 Enterprise Edition and Pentaho Business Analytics, respectively.

As 2011came to a close, however, Pentaho focused on burnishing its enterprise support bona-fides. In December, for example, Pentaho unveiled a new worldwide customer support site -- the Pentaho Global Support Portal -- which it positioned as part of an ongoing effort to beef up its sales and support resources. Pentaho in 2011 also introduced a new customer-focused support scheme -- the Pentaho Customer Success Team -- which assigns a support point person to every account. Finally, Pentaho says it has doubled the size of its global customer support team.

Support is a big concern for customers of all sizes. For enterprise customers and enterprise-scale deployments, it's key. When it comes to support, OSS players such as Pentaho have had to clear an additional hurdle: pervasive fear, uncertainty, and doubt-mongering. Closed source competitors (and open source skeptics) have raised concerns about everything from OSS intellectual property liability to OSS coding quality to OSS security. They've likewise sought to raise concerns about the quality, consistency, or reliability of OSS support.

That's why Pentaho's recent push -- which explicitly takes up the theme of support in an enterprise context -- is a big deal. The new Global Support Portal, for example, is based on the Zendesk customer support platform. Zendesk, for the record, is the same company that handles customer responsiveness for deal-of-the-day Website GroupOn and social media powerhouse Twitter.

What's ironic is that -- as of late 2011 -- Pentaho's combined OEM and SMB customer bases were still larger than its enterprise base.

"About 65 percent of our business comes from direct customers, 35 percent [comes] from the OEM side," said Richard Daley, Pentaho founder and chief strategy officer, told BI This Week in August. "From a positioning standpoint, OEMs, along with SMBs, have been the primary purchasers of Pentaho products. That's not to say that we don't have big enterprise deployments, and that we don't get brought into big Fortune 500 implementations. We've just been stronger in these [segments]."

Focusing on Performance, Too

Pentaho's strength in both of these segments shouldn't be surprising, especially considering the bulwarks that have traditionally confronted OSS adoption in the enterprise.

In tandem with last year's push to burnish its support, however, Pentaho also sought to beef up its performance profile, too -- particularly with respect to the kinds of high-end, advanced analytic applications that are highly desired by enterprise customers.

At the TDWI World Conference in Orlando, for example, the company announced a new version of Pentaho Business Analytics, an analytic platform that combines Pentaho's reporting, ad hoc query and analysis, data integration, data visualization, predictive analytic, and Hadoop assets.

The new Pentaho Business Analytics shipped with several end-user-oriented -- and enterprise-IT-friendly -- amenities, such as a drag-and-drop report designer (which Pentaho first released back in June, with its BI Suite 4.0) and a self-service dashboard design facility.

It likewise packed several Big Data-oriented features -- including support for in-memory caching (to improve availability and accelerate recovery in the event of a failover) and in-memory data aggregation (to better manage query loads).

Interviewed prior to Pentaho's announcement of its Big Data-ready Business Analytics suite, Daley said that most customers weren't yet kicking the tires on Big Data. Most, in fact, are still doing warm-up exercises.

"Realistically, you're still in the early, early adopter stage. Until Hadoop itself [becomes easier] to onboard and onramp for customers, ... it's a self-selection of customers that are able to get that [up] and running. We have customers today ... who are buying our Hadoop products. I'd still say it's in its infancy, however," he comments.

"The ones with the biggest pain points are the ones that are going through and pushing fast and furious on Hadoop. You go through and look at the use cases on Web logs [and] sensor data ... these customers will tell you that they have to push this [information] into something other than a relational database."

Back to Basics: Bread and Butter BI

In spite of the high-end focus of its Pentaho Business Analytics offering, Daley says that traditional reporting and analysis -- the bread and butter of BI -- will continue to be critical to Pentaho's fortunes in the enterprise, with SMBs, and in the OEM channel. The release of Pentaho 4.0 last June gave Pentaho an improved story to tell in this regard, he says.

"One of the biggest enhancements in [Pentaho] 4.0 was an all-new ad hoc reporting interface. It supports drag-and-drop design: over the Web, inside a browser. It's very like a standard WYSIWYG designer: it gives the user full control; they don't have to know SQL, they don't have to download a desktop designer, and they don't have to learn reporting paradigms."

Regarding self-service, Daley cites the new data-source access wizard that also debuted with Pentaho 4.0 "A lot of [BI] is people just going through and setting up basic reporting and analytics. Always the big pain [in this case] is getting to the data, so even though we give [users] great interactive reporting capabilities, if they can't get to the data, it doesn't make any sense," says Daley. "We created a brand new data source access capability [in Pentaho BI 4] that, as a business user, lets you sit there and click on a couple of steps and identify what data you want to work with."

For large shops with existing data warehouse practices, this isn't a huge deal, Daley concedes. "They just need reporting and stuff on top of it, although most large organizations tend to have sources other than the data warehouse [that] users need access to, too," he continues. "We're also dealing with a lot of small and medium-sized businesses that never really had that [data warehouse] infrastructure in place. This [capability] helps them get the information they need."