Saturday, January 22, 2011

Guided Ad hoc 2.0 with Pentaho 3.x

A barrier that may be encountered when adopting a Business Intelligence tool is...ease of use. If the BI tool is too difficult for business users to use or understand, they may resort back to using antiquated desktop databases and spreadsheets. If the new BI tool is not being used to the best of its abilities, then the organization is not leveraging their BI investment. Therefore its dreams of consolidating information and delivering one version of the truth just went out the window. **CRASH** Sure training, mentorship and education can help with this barrier, however there are many individuals that are simply resistant to change. What if there was another way to provide robust reporting capabilities without a steep learning curve? Possibly with the creation of templates designed for a specific purpose? Let's leave the Dashboards, Ad hoc query and OLAP tools for the experts and provide a simpler way for the technically challenged to run and create reports.

Read more about it here and watch the tutorial and download the sample:

Wednesday, January 12, 2011

Pentaho Reporting and Pentaho Analysis

The Pentaho BI Suite consists of the following BI modules: Reporting, Analysis, Data Integration, Dashboards and Data Mining. The modules can be deployed as an entire package or as individual components that can be integrated and embedded. Traditionally, each module is used in conjunction with a specific business need. The focus of this article will be to highlight the differences and similarities between Pentaho Reporting and Pentaho Analysis. Most organizations already have some form of Operational Reporting and Analysis tools. These tools are used for tracking business performance, trends and uncovering potential problems that require action. Business questions usually fall into a few categories. Questions that are asked on a regular basis for certain time periods (years, quarters,months, weeks) and questions that are asked for a purpose - usually random in nature, posed to uncover potential problems or outliers and can be commonly referred to as Ad hoc queries or OLAP Analysis.

Read more here:

Thursday, January 6, 2011

See you in San Francisco at the Pentaho Global Partner Summit

It is only the 6th of January and already this has been a crazy month.

"Crazy" as in crazy busy. I'm sure you have heard the phrase before, but then again it depends on what industry you are in. A down economy has certainly not affected the Commercial Open Source space, I can tell you that.

To add to all the excitement, on January 19th and 20th is our Global Partner Summit in San Francisco at the Presido Golden Gate Club.

CTOs, architects, product managers, business executives and partner-facing staff from System Integrators and Resellers should attend this event. You can register and find out more here: Global Partner Summit

There will be technology tracks, business tracks, Q&A discussion panels and more for all to take part in. This year I am honored to join the team to present a couple of topics that surely should not be missed.

Sales Engineering will be holding sessions that will show you how you can brand and customize the default Pentaho User Console. I will also present how adding "Guided Ad hoc" to your applications can provide business value to those who are not so accepting of the out-of-the-box tools.

You can view the full agenda here

I look forward to speaking with many of you as well as, once again ,visiting my home away from home...San Francisco.

See you there.

Michael Tarallo
Director of Sales Engineering

Wednesday, January 5, 2011

Pretty is not a decision maker

"Beauty is only skin deep." "Physical beauty is superficial." Blah, Blah, yeah I know, I am sure you've heard it all before.... and "No" they are not phrases coined by some "ugly duckling" in an attempt to make it feel better about itself. However, physical characteristics will always play a part in how we as humans are initially captivated and intrigued. It's true for how most of us consider our mates... it is even true for how some organizations consider BI software.

When I was with a proprietary BI vendor (before the explosive disruptive model of Commercial Open Source BI), I spent 3 weeks on site with a prospect conducting a POC (Proof of Concept). It was well received for both its data integration and information delivery functionality. However, even though we had specific data integration capabilities that surpassed the competition, we still lost because the business users liked the competition's "Prettier Dashboards". The first thing out of the IT Director's mouth was... "Well, we went with "Vendor B", their dashboards just looked better." Completely disregarding the specific EDI integration that we brought to the table. Huh..I pondered for a few minutes...scratching my head with a puzzled look. Cocking my head slightly to one side I rebutted, "That's disappointing, since when did 'Pretty' become a decision maker?" The Sales Rep looked over at me and smiled. That was over 5 years ago.

More recently, inspired by a colleague of mine, Gabriel Fuchs and his web post Data Visualization – Cool is Not a Key Driver! - I am still overwhelmingly surprised how much emphasis organizations put on the importance of "having nice looking dashboards" without really knowing what is involved under the covers. Further more they have a tendency to not know what charts or visualizations should "go" with "what" data. (You'll be surprised at how many simple line charts are used incorrectly or when to use or not use a pie chart) I have heard so many colorful descriptions I had to wonder if they really understood the business value behind a BI solution at all. From dashboards that are "Nice and Friendly" to those that are "Fancy, Sexy, Sizzle and In your face". At times I was wondering if they were describing their ideal mate or the latest and greatest automobile.

All too often, IT or the occasional business user will start researching BI solutions and stumble upon a software package that appears to do what they need. Perhaps they were able to get a "Fancy" dashboard up and running quickly. Soon they may find that the proposed solution is either too costly, not scalable, only runs on Windows, cannot access all their data easily or perhaps only provides dashboards and lacks other critical BI functionality. They may have been initially captivated by the Siren's music but soon realize that the "Fancy" dashboard was just skin deep. 1 out of every 10 calls that I am on reveals that the prospects are only looking for just dashboards. When further discovery takes place, it is also learned that the "dashboard only" deployment is usually for just a few users and localized departmental data, not exactly an Enterprise wide solution. The rest of the prospective calls are looking for Dashboards as well as Reporting, Analysis and more often than not, Data Integration. I mention Data Integration as well because these organizations have disparate data sources on many different platforms. They are looking to easily access, optimize and visualize this data that will be able to answer today's questions as well as tomorrow's questions, perhaps across the entire data set - not just a small slice.

Here are some important facts to remember:

  • Most business users do not understand the value of BI
  • It is important to show how BI can help knowledge workers do a better job
  • IT cannot just throw a BI application at the wall of business users and hope it sticks
  • BI is NOT a technology tool
  • BI involves specific business processes
  • BI applications can both drive revenue growth and can also reduce costs to optimize profits
  • Do NOT assume that Subject Matter experts understand BI and its potential
Here are some key questions an organization should ask before they begin the process of finding a BI solution provider:

  • What is your definition of a successful evaluation?
  • What data is needed in order to…..?
  • Where is the data I need in order to….?
  • How easily can I access all that data?
  • Do I have the proper skill sets to deploy a BI Application?
  • Do I want my business users to ask Ad hoc questions?
  • What questions do I or my business users want to ask of the data?
  • Do I need Operational reporting including schedule and distribution?
  • What have I found from my existing BI application(s)?
  • What actions do I want to take from my findings?
I hope you now realize that there is a lot more involved than just that "Pretty Dashboard".