HP Business Intelligence Solutions

Main | April 2010 »

March 30, 2010

Innovation, Transformation, and the BICC

A recent tweet from an industry analyst intimated that he was meeting with a client who had spent $11 million on their business intelligence implementation over the last year and that they were still uncertain of the value they would be getting from the investment.  

This is a wake up call for anyone not currently pursuing a business intelligence competency center. The verdict is out: companies who started their BICC efforts 3 to 4 years ago are now using BI to transform their companies.

Since we've been helping companies align IT and business using the concept of a Business Intelligence Competency Center for more than 5 years, I've had the privilege of talking to companies who have mastered the art.  Just for the record, anyone can write a book, anyone can write a white paper, but I prefer talking to people who have done it and learning from them. That's exactly what I did; I talked to people who are seeing the benefits of a working BICC. Here are three trends that get me excited about BICC success.

TREND ONE: Business Intelligence Tied to Corporate Objectives

A common theme for BICC success is the right level of business sponsorship and the right mix of business and IT participation. With these two elements in place, we are seeing companies consistently tie their entire business intelligence program to very specific corporate objectives and aligning with the overall mission and vision of the company. In some cases, companies are even mapping every single BI initiative to a specific business processes or set of processes. This means that nothing is done in BI that doesn't have a direct impact on the business with measurable results.

TREND TWO: Move from BI Maintenance to Innovation

I'm surprised at the number of organizations that are still approaching BI as a means for driving more reports for the business. I'm equally surprised at the number of business users that are operating in a world of reports that stems back from the day of green bar reports.

With a working BICC in place, companies are seeing the walls between business and IT come tumbling down. IT is discovering more of what the business actually needs to operate more efficiently. The business is getting a better understanding of what can be done with business intelligence technology. The end result: amazing efficiency.

Imagine what it costs to produce 6,000 reports. Consider what it would be like to reduce the number of reports produced to 180 and exceed the expectations of the business. Even better, think about the resources you just freed to work on innovative uses of business intelligence.

TREND THREE: Emergence of Trusted Information

Because a working BICC drives toward data ownership, data stewardship, and data governance, companies who are doing it right are experiencing significant improvements in data quality. Companies who have been driving this kind of governance program are actually changing the information culture of their companies.

In today's world, the first thing you do when you get a report is question the accuracy of the figures. In fact, it's likely that you will get a second report on the same topic, from another part of the company, and you will find conflicts in the results.

With the right governance in place, you can maintain data heritage in a way that informs anyone viewing a report the sources of the data and the details of any transformation and cleansing that impacts the end result. We are seeing companies move toward certified reports or certified sets of data, stamped with the mark of the BICC, and therefore trusted by the business users.

The culture shift is simple to understand. When people get used to getting trusted information, they will immediately question or reject anything that doesn't have the BICC stamp. It takes time and discipline to change the information culture of a company, but it's worth the journey.

So, these are just a few of the trends we are seeing from companies who have taken governance seriously. And this is a wake up call for anyone not currently pursuing a business intelligence competency center. Or perhaps, we should be talking about a Business Intelligence Innovation Center!

I'd love to hear about your success with Business Intelligence Competency Center's. What have you done that's worked well? What results have you seen? Either leave a comment below, or send me an email at john.santaferraro@hp.com.

Download this white paper (0.70MB, PDF) to read more about how to implement a Business Intelligence Competency Center.

You can also dig deeper into the Top 10 Trends in Business Intelligence for 2010.

Post by John Santaferraro
Twitter: santaferraro

Posted by HP Business Intelligence Solutions at 12:02 PM

March 29, 2010

The Rise of the Single-stack Vendor: Modern Breakthrough or Retro-trend?

Does anyone else see the irony in Oracle Corp. advocating buying the entire system – from chip to application – from a single vendor to “help spare the customer from the frustration of having to build business IT systems from piece parts?” (Oracle being the company who practically invented the concept of portable software in the mid-1980s to help spare the customer from the tyranny of dependence on a single vendor).

Of course Oracle is not the only company consolidating many components into a single IT stack. Witness the mega-acquisitions of the past several years.

The question is: does the single-stack concept resonate with 21st century buyers (who have more of a choice than buyers did in the '70s and early ‘80s)? During the past 25 years, did we learn that the effort to integrate heterogeneous components, best of breed though they may be, is counterproductive? Or is that effort outweighed by the value of innovation stemming from competition among many small vendors?

Here is research that indicates that the latter is the case. But there is plenty of evidence that both approaches have their supporters.

Either way, what makes the single stack appealing today is:

• The same advantages it once offered
• A common operating system to facilitate point product integration and vendor switch options
• Standards for allowing the stacks to interoperate

Interesting that Linux has helped to take us back to where UNIX helped take us from.

Posted by HP Business Intelligence Solutions at 10:03 AM

The Rise of the Full-stack Vendor: Modern Breakthrough or Retro-trend?

Does anyone else see the irony in Oracle Corp. advocating buying the entire system – from chip to application – from a single vendor to “help spare the customer from the frustration of having to build business IT systems from piece parts?” (Oracle being the company who practically invented the concept of portable software in the mid-1980s to help spare the customer from the tyranny of dependence on a single vendor).

Of course Oracle is not the only company consolidating many components into a single IT stack. Witness the mega-acquisitions of the past several years.

The question is: does the single-stack concept resonate with 21st century buyers (who have more of a choice than buyers did in the '70s and early ‘80s)? During the past 25 years, did we learn that the effort to integrate heterogeneous components, best of breed though they may be, is counterproductive? Or is that effort outweighed by the value of innovation stemming from competition among many small vendors?

Here is research that indicates that the latter is the case. But there is plenty of evidence that both approaches have their supporters.

Either way, what makes the single stack appealing today is:

• The same advantages it once offered
• A common operating system to facilitate point product integration and vendor switch options
• Standards for allowing the stacks to interoperate

Interesting that Linux has helped to take us back to where UNIX helped take us from.

Posted by HP Business Intelligence Solutions at 10:03 AM

March 26, 2010

Informatica, 29West, and HP Business Intelligence

Informatica this week announced its acquisition of 29West, a pioneer in ultra-low latency messaging, signaling their continued push toward real-time data integration. 29West is currently being used by companies to send and receive millions of messages per second with microsecond delivery times for zero-latency data delivery. For Informatica, this means the provision of the industry's first zero-latency data integration platform.

For HP, Ultra Low Latency Messaging (ULLM) is right in the sweet spot of HP Neoview Advantage, the only data warehouse platform built with out-of-the-box extreme availability and designed for a mix of analytic and zero-latency transactional workloads. As a reseller of Informatica's Enterprise Data Integration platform, our customers benefit when streams of information bring new insight into their decision environments.

We see the Informatica and 29West relationship as a positive move and a possible future opportunity. The HP connected intelligence approach looks at tapping into new sources of data such as sensor data, RFID, logistics, smart meters, clickstream data, digital oilfield data, and a whole new world of information in motion. We are currently working with companies to harvest low-latency data to provide them with competitive advantage around customer intelligence and operational excellence. For example, smart meter data includes amazing insight into how utility customers use power, allowing utility companies to do micro-segmentation and builder closer relationships with their customers. New smart meter portals are also rich sources of clickstream data that provide more interaction points with key customers. I talked a little more about this in my Smart Meter Intelligence podcast back in July 2009.

We anticipate that new insight from streaming information could allow companies to sustain competitive advantage over time. As the world of information continues to grow in complexity, our data integration services continue to lead the way toward solving our customers' most difficult data challenges.

Posted by HP Business Intelligence Solutions at 12:59 PM

March 24, 2010

Gen X Business Intelligence: Where Design Meets Intelligence

What happens when modernist thinking meets traditional data warehousing? You could end up with a Picasso of information, or a Dali database. But personally, I'd rather look for new architectural design that would transport us into the 21^st century of digital intelligence. I'll take Le Corbusier's "form follows function" over modernism any day. Why not consider Buckminster Fullers's approach of "applying the principles of science to solving the problems of humanity"? There might even be something to learn from Frank Lloyd Wright, the organic architect.

So, imagine design ethos infecting our world of business intelligence. The perfect business intelligence solution, like a modern touring car, responds to demand in real-time. For long trips, the suspension is loosened for comfort. For handling on mountain curves, the car moves closer to the ground and the suspension tightens. When more power is needed a turbo turns on. When the car goes into a skid, intelligent power and braking help steer the car.

Great design is visionary and vital. When we think of great architects and designers, we find people who bring great vision and problem solving skills to the world.

"Form follows function" was a term attributed to Le Corbusier, otherwise known as Ludwig Mies Vander Rohe, and others who established modern concepts of design that veered away from "style" towards practical solutions to design challenges. Ergonomics, human factors, and other engineering analogs could transform the way we think about business intelligence.

Connecting to great design principles bolsters the brand of business intelligence around style, innovation, and engineering.  Consider what ergonomics might mean to how people interact with information. Reflect on the possibilities of introducing human factors to the design of information architectures and end user interfaces. This new thinking can differentiate business intelligence solutions with the idea that we design solutions with real business functionalities and real business value.

Another architectural great, Buckminster Fuller, loved moving from science to real world solutions. Perhaps, most notably, his scientific discoveries lead to the design of the geodesic dome. The patterns and strength of molecular structures lead him to answers about everything from architecture to human existence and social order.

Again, consider the benefits of getting outside perspectives to impact the way we do business intelligence. Think about the way we move particles to near light speed in a linear accelerator. Is there a way to use information to accelerate business across the different divisions in organization? What would happen if we took the 40 principles of TRIZ and applied them to our complex information challenges? You can take almost any principle of science and find innovative ways to think about solving our business intelligence challenges.

My favorite architect has always been Frank Lloyd Wright. According to Wikipedia, "Organic architecture is a philosophy of architecture which promotes harmony between human habitation and the natural world through design approaches so sympathetic and well integrated with its site that buildings, furnishings, and surroundings become part of a unified, interrelated composition."

So, what happens when we stop thinking of ourselves as old world information architects and see ourselves as integrators and developers of the unified reality of information? We become aware and considerate of unique business surroundings and the real world surroundings of different information workers and front line decision makers.

BI modernization is not about finding new, free from, information architectures with point solutions and pop up data marts that end up looking like modern art. It's about understanding what the business needs to do (form follows function); its about decision science (principles of science solving human problems); and its about sympathetic, integration of intelligence into the world of the front line worker.

BI modernization is only one of the "Top 10 Trends for Business Intelligence." To take a look at the other nine trends, download the white paper at hp.com/go/bi.

Post by John Santaferraro

john.santaferraro@hp.com

Twitter: santaferraro

Posted by HP Business Intelligence Solutions at 2:39 PM

March 23, 2010

What You Should Know About Measuring Data Warehouse ROI

Six or seven years ago, the idea of measuring the return on investment (ROI) of your data warehouse was all the rage. Vendors suggested it. Consultants suggested it. Perhaps your boss did more than suggest it. There was no shortage of help, from meaningless equations on websites, to courses, to whole books on how to do it. I too wrote and spoke about measuring the ROI of the data warehouse, or more appropriately, the business application it enabled. A presentation that I delivered at the Computerworld Premier 100 IT Leaders Conference reviewed the basics, along with practical wisdom from leading data warehouse users.

The basic equations (including Payback Period, Internal Rate of Return, and Net Present Value) are straightforward enough:

 ROI = (Gain from Investment - Cost of Investment) / Cost of Investment

The difficulty, however, is in determining what values should be used in the equations. Assessing the gain - business benefits and/or cost savings – is tricky. Even assessing the cost is not that easy.

But That Was Then

At the time, data warehouses were largely discretionary. So it was important to estimate and measure the return on what was a large investment, to justify making it here vs. somewhere else, IT or otherwise. A few organizations even calculated the Internal Rate of Return (IRR) to compare against their company’s hurdle rate. Even those who didn’t do anything, usually believed that they should.

At the same time, not much attention was paid to measuring the ROI of order processing, ERP, or other mandatory systems needed to run the business. Alternative solutions were evaluated and compared on the basis of total cost of ownership (TCO). Unlike the situation with the data warehouse, there was no debate about whether or not the investment should be made.

And This Is Now

As I indicated in a 2004 Computerworld article, data warehouses were destined to become more business critical. As they have in fact become much less discretionary and more necessary, the inclination to forecast and measure their ROI has decreased. In a BI survey that Hewlett-Packard conducted with BI and data warehousing professionals in 2009, fewer than 16% of respondents said that they measure DW/BI ROI. Over 41% said they had no plans to do so, an indication which increased over the course of the survey [see chart]. Given the value of staff time, it appears that this exercise is not viewed to be as necessary or worthwhile as it once was.

            

 

 

Other Approaches

It seems to have given way to measurements that are more meaningful such as including the data warehouse cost in measuring the ROI of the business solutions(s) it enables.

When organizations evaluate vendors for new implementation, enhancement, modernization or upgrade, being able to make a meaningful TCO comparison among alternatives is as important as it ever was. This is by no means a straightforward exercise. The advent of new configuration approaches such as appliance packaging and use of analytic data marts on one hand vs. in-database analytics on the other has made it more complicated to make an apples-to-apples comparison. The approach that represents the lowest cost in the first year may actually be the most expensive in years two or three based on the variations in maintenance pricing, staffing and training requirements over time, as well as the approach to upgrades, etc.

Given the flexibility and variety of data warehouse deployment approaches, perhaps the most useful approach is to focus on quantifying the benefit of the business solution while separately developing a rigorous comparative TCO measurement of the data warehouse, data integration and BI infrastructure required to deliver it.

Posted by HP Business Intelligence Solutions at 6:49 PM

March 19, 2010

BI Trends and the New Generation of Business Intelligence

It is interesting that each new computing era seems to emerge from a recession. Interactive computing was born when people began using a CRT and keyboard as a computer input device to replace punch cards in the mid-1970s. Amid another recession, the 1981 introduction of the IBM PC is generally credited with beginning the Personal computing era (or at least being the first commercially successful implementation) . The Networked computing era took off circa 2001 as we emerged from the burst of the dot-com bubble.

But then, recession often fosters innovation. (Out of the Great Depression came chocolate chip cookies, the laundromat, xerography, and invention in a Palo Alto garage in 1939 went on to become Hewlett-Packard).

What’s even more interesting is the extent to which the computing industry has developed software to effectively exploit each new platform to glean valuable information from business data:

• Interactive computing – Computers like Digital Equipment Corporation’s PDP-11s were so cost effective that companies could afford to, once a month, load summarized production data onto one of these machines for the purpose of what was then called decision support, or executive information systems. Programmers coded in 3GLs (3^rd generation languages) like Cobol, and newly-invented wombat-friendly 4GL (or possibly transitional 3½ GL) DATATRIEVE. This allowed different cuts at the data, and the ability to sort results in multiple ways, giving business managers a view they hadn’t had before.

• Personal computing – The PC was the perfect home for The Spreadsheet, originally invented as Visicalc on the Apple II. The spreadsheet became not only the user interface of choice for accessing business data, but the killer app of the era, allowing business people to manipulate data themselves, do what-if analysis, and even combine data from multiple sources.

• Networked computing – In the 1990s, the World Wide Web made use of the Internet pervasive. By 2000, the dot-com bubble had burst, but the networked computing era had taken off and companies developed successful business models that took advantage of the new style of computing – Amazon.com, Travelocity, and Google to name a few. Their business is analyzing massive amounts of data so they can provide tailored services to customers they’ve never met. This represented a whole new way of doing business where success was based on how well you analyzed data, and on applying analytic algorithms and concepts from operations research. Whereas the previous era linked people within a division or perhaps a company, this now allowed B2B and B2C.

The new generation of business data reporting and analysis, now known as business intelligence (BI) came into its own, characterized by the rise of the power user and user flexibility – OLAP cubes allowing users to slice & dice data, dimensional models for easier use by non-technical business users, visualization, web-based query, use of portals for information sharing, etc. But even now, analysis is still primarily done on historical data, and BI still largely lacks the ability to improve business processes and performance, and to support better operational decisions, and more automated decisions and action in real time.

We are now entering the next computing era, often known as “ubiquitous computing,” characterized by proliferation of embedded technology - where computers are available throughout the physical environment but effectively invisible, a mobile revolution, digitization of content, simulation, sensors, and the emergence of location and context within the data. This era too will inevitably shape BI direction, with the prospect of being able to detect opportunities and threats and take action based on the specific situation, rather than relying on a pre-programmed scenario which not be appropriate. BI systems will evolve to take advantage of the new computing environment to increase the ability to exploit data to improve business management.

The new generation of BI systems will have to make use of multiple event data, be able to find the data that is needed, extract what is relevant, assimilate data from multiple sources, analyze it and incorporate the resulting information into applications as appropriate. This will increase the need for enterprises to better manage data and information assets, among other things.

HP has published our view of the most significant BI trends in this year’s “Top 10 trends in Business Intelligence for 2010.” In this blog, I would like to further probe the trends that are shaping BI, the objectives, the challenges and ways to overcome them, ways to exploit the new computing platform, along with opportunities presented by new products, technologies, architectural approaches and delivery models. Which ones are most promising? What are the highest priorities? What are other organizations doing? I welcome your discussions and look forward to thought-provoking dialogue.

Posted by HP Business Intelligence Solutions at 3:34 AM

March 17, 2010

Collaborative Business Intelligence is Here!

I'm on Twitter (twitter.com/santaferraro) almost every day. My LinkedIn profile (http://www.linkedin.com/in/johnsantaferraro ) continues to grow. And my friends from all aspects of life are finding me on Facebook (I won't post that link. It's more of my personal life.) Along with all of that, I live in the world of business intelligence from 9-5.  I find myself sitting right in the intersection of social media and business intelligence.

The connection that I see most of my colleagues making is around analyzing the social media conversation to discover things like sentiment and identify new tribal leaders.

But what really interests me is the opportunity to change the way companies do business intelligence. What happens when the analyst and the business user are connected through online tribal communities? What happens when the rookie analyst is set free to produce and publish analytics that are 100 times more impactful than the one's produced by veterans.

We are already seeing the emergence of companies like Lyzasoft (www.lyzasoft.com) who are embedding analytics into a blog and twitter-like environment for collaboration. Analysts can embed an analytic widget right into the text of a blog and have it tweeted to other colleagues in order to get their input. Another interesting approach is the Google Fusion Tables, a pilot with analytics on the web and the ability to embed apps in blogs and other online media.

This move toward "community-based" intelligence brings a unique set of challenges and compelling number of opportunities. The challenges include loss of control and loss of traceability.

• Loss of Control. Because of information quality issues and budgetary confinement, many business intelligence groups are keeping a tighter reign on the flow of information. The emergence of a collaborative environment could counter this trend and mean a loss of information control. The balance point is that the community tends to quickly identify poor uses of information.
• Loss of Traceability. When communities of people are tweaking reports, changing views of information, and adding in their own equation to the mix, there is the danger of ending up without a clear sense of data heritage. If the end result is interesting data without context, there is no way to measure the quality or value of the end result. Products like Lyzasoft are building traceability into their product.

The collaborative opportunities far outweigh the challenges. Consider better alignment between business and IT, quicker time to value for the use of analytics, and the emergence of star analysts from among business and IT ranks.

• Better Alignment Between Business and IT. With business users and IT working together in a collaborative environment, both groups will quickly come to an understanding of each other's world. The business will gain an understanding what can actually be done with business intelligence tools and constructs. The IT analysts get a clear picture of what the business actually needs to drive growth and profit.
• Quicker Time to Value. It's as simple as "two heads are smarter than one." It's the nature of collaboration. Immediate access to information and immediate feedback will drive more value and quicker results.
• Emergence of Star Analysts. In today's world, you get noticed if you have the title "Senior Business Analyst" and you have been producing business intelligence for years. In the world of collaborative BI, you can show up on the scene, have no title, be on the business or IT side, come up with an amazing look at information and suddenly be noticed by the community. This may be threatening to the old line BI analysts, but it's good for the business. It challenges everyone to drive more value out of BI.

My recommendation: If you jump into a collaborative business intelligence project, make sure it's aligned with your BICC (Business Intelligence Competency Center). You might want to download the white paper on Building a Business Intelligence Competency Center (PDF, 1MB) for more details on how to do that.

This is just one of the "Top 10 Trends for Business Intelligence." To take a look at the other nine trends, download the white paper at hp.com/go/bi.

 

Post by John Santaferraro
Twitter: santaferraro

Posted by HP Business Intelligence Solutions at 1:30 PM

March 12, 2010

Are we back in the 60's?

Yes, I was alive in the 60's.  And yes, blogging reminds me of the 60's.  I was recently up in the Haight-Ashbury district of San Francisco and I reminisced about the free concerts, sit-ins, and general feeling of openness and community that used to thrive on the edge of Golden Gate Park.   I always loved the view from Buena Vista Park and felt like something was happening in this place on top of the city, where you could see the world.

So, here we are 50 years later. The web is in full bloom and web logs are the way of expressing yourself to the world. (I had to say "web logs" because I have this thing about remembering from where we have come.)  The view from here is amazing!  You can actually see the whole world (wide web). Is it true that there are over 200,000,000 blogs on the web now?

Anyway, welcome to my little corner of the internet.  In my "connected intelligence" blog, my hope is to communicate openly, invite you to dialogue (yes, I want to hear from you), and spark the kind of conversation that can potentially open new doors to the world of intelligence.   We have barely plumbed the surface of what can be done with the insight that is still locked away in oceans of data.  Kudos to the men, women, and companies who are driven to dig deeper and find the nuggets of knowledge that will change the way we do business and live life.

So, why a "connected intelligence" blog?  I thought of doing a blog on "business intelligence," but I wanted to go beyond business.  Business intelligence is only one piece of what I want to talk about.  Customers still matter most.  Just in time delivery of goods still drives down costs and increases profit.  And business performance delivers value to everyone that matters to your company. But there is incremental value to be created when we connect intelligence beyond the boundaries of traditional BI.

We all understand that intelligence can make a dollar, but I'd like to go beyond commercial intelligence. What about intelligence that makes a difference?  Think about intelligence that could lead us to the cure for cancer; product data that could reveal the way to a cleaner, greener planet; social intelligence that could give us insight into how to meet the needs of third world countries; or, at a personal level, productivity intelligence that could give us back two extra hours a week to be with our families.

That's the connected intelligence journey I'd like to take with you...

Posted by HP Business Intelligence Solutions at 6:53 PM

Welcome, Business Intelligence fellows, clients, analysts, and enthusiasts!

The HP Business Intelligence blog is the home of news, opinions, and conversation about Connected Intelligence and its power to transforms the enterprise's ability to compete and thrive in an information-centric world.  We’ll be exploring market trends, comment on strategic information technologies, and examine what’s happening in the business intelligence space. 

We look forward to our conversation and to sharing views and insight!

John Santaferraro, HPBI Marketing Communications & Industry Marketing Lead

Vickie Farrell, HPBI Marketing Strategy

 

Posted by HP Business Intelligence Solutions at 1:59 AM