Archive for the ‘Business Intelligence’ Category

Oracle OpenWorld is next week – the annual look-at-me showcase event.  In their own words, Oracle describes it as “the world’s largest and most important conference for Oracle technologists, business users, and partners.”

There’s a lot to anticipate, especially with the recent turnover in the President’s office at Oracle.  Charles Phillips has been stricken from the agenda and Mark Hurd inserted in his place (with Safra Catz remaining in her co-president role).

It should be an interesting show. Below are some key highlights and information I’ll be eagerly anticipating, along with some of my predictions:

• Ellison / Hurd / Catz
  • Questions: Will Ellison address anything about how the new co-president pair will differ from the roles Phillips and Catz played? How will Mark relate to Catz?  Will they appear together?  Their keynotes are currently scheduled on separate days.
  • Prediction: Ellison won’t mention Phillips at all, but he will emphasize how well Catz and Hurd will complement each other.  He’ll be nice to HP, instead aiming his barbs at IBM (with some well-placed insults directed at SAP, because he just can’t resist the temptation).

• HP and Oracle
  • Questions: Ann Livermore, Executive Vice President, HP Enterprise Business, is scheduled to deliver a keynote address immediately after Ellison and Catz speak on Sunday evening.  Will last week’s war of words affect what she says?  Will she address the issue at all?  Will Ellison be confrontational about HP?
  • Prediction: It’ll be all love and kisses between Oracle and HP…at least on stage.

• Extension of the Exadata approach
  • Questions: Exadata is Oracle’s product family of high-performance data warehouse appliances.  They’re extremely expensive, but incredibly fast.  Will Oracle announce extensions to this approach, making more of their software stack available on appliance hardware?
  • Prediction: I expect that they’ll make some significant announcement here, possibly including a medium-sized acquisition.  Maybe this is where Hurd will get his first moment in the spotlight – the announcements here are likely to be made by Hurd in his Monday morning keynote, followed by Oracle Executive VP John Fowler, who was CTO of Sun’s software organization before Oracle acquired Sun.

• Fusion Applications
  • Questions: Ellison announced Oracle’s Project Fusion on January 17, 2005 and he discussed the next generation application suite – the Fusion Apps.  His keynote on Sunday evening will come just 2071 days after that initial announcement.  In those 5 ½ years, Oracle’s communications about Fusion Apps have been all over the map: sometimes vague and sometimes detailed; occasionally effusive and optimistic but often silent; cryptic but occasionally clear and concise.  Consistently, the vision has always been ambitious.

    There are more than 30 Fusion Apps sessions on the agenda.  Clearly, momentum is building (at least from a hype standpoint).

    Will Oracle finally commit to a release date?  Most analysts think they will.
    Will they provide clarity on how Fusion Apps will co-exist with E-Business Suite, PeopleSoft, Siebel, and JD Edwards?

  • Prediction: They’ll announce a release date for Fusion Apps, but it’ll be vague (as far as what apps will be included).   They’ll also be vague about interaction with existing apps – there will be lots of platitudes about “Applications Unlimited” promising eternal support for the legacy applications coupled with more platitudes about “co-existence” and “gradual migration” and “SaaS architecture providing flexibility.”  In the end, lots of customers will leave still scratching their heads.

• Oracle BI 11g
  • Questions: Oracle BI, on the other hand, is real and available now.  The newest release of Oracle’s flagship business intelligence product (Oracle Business Intelligence Suite Enterprise Edition Plus, for those who want to use the complete and proper name) was announced in July and became generally available a few weeks ago.  This new release has gotten a lot of positive press from analysts.  Will it be as good as advertised?  Will we learn whether some of the limitations of earlier versions have been corrected? Will Oracle acknowledge any flaws or shortcomings at all?

    I’ll personally be attending quite a few sessions on this new product, looking to learn its details.  I’m hoping that this turns out to be the most interesting part of the show.

  • Prediction: Yes, this will be impressive.  However, once we look under the covers, it won’t be quite as impressive as Oracle wants us to believe.  I’m sure I’ll have more to say about this in weeks to come.

For now…I’ll leave you with this thought:

Don’t expect fireworks or surprises at OpenWorld next week.

It’ll be yet another tightly choreographed, “Oracle is ready to rule the world” event.

That Ain’t Right

Whenever I agree with Larry Ellison, I reconsider my position to make sure I’m thinking clearly.  This time, as I reconsidered and dug a little deeper, I decided that I didn’t agree with him after all.  Larry is wrong.

Ellison, the Founder and CEO of Oracle Corporation, is notoriously direct, confrontational, and controversial.  He frequently makes outrageous claims about the software industry in general and about his competitors.  Earlier this week, Ellison attacked the decision by Hewlett Packard’s board of directors to demand the resignation of Mark Hurd, CEO of HP.

The Hurd story has been reported broadly over the past week.  In short, following an inquiry into sexual harassment allegations against Hurd, the HP board found no evidence of sexual misconduct, but did find that he had filed incorrect expense reports.  The improper expense reports were (at least partially) to conceal his relationship with the woman involved (who was not an HP employee, but instead was a consultant hired by the company).

In typical Ellison fashion, his critique was blunt and unequivocal.  “The HP board just made the worst personnel decision since the idiots on the Apple board fired Steve Jobs many years ago,” said Ellison in an email to the New York Times.  He added that, “In losing Mark Hurd, the HP board failed to act in the best interest of HP’s employees, shareholders, customers and partners,” pointing out that the board’s inquiry had concluded that the allegations of sexual harassment were false.

Ellison is, if nothing else, consistent.  Earlier this year, Oracle’s president, Charles Phillips (who is married), admitted to having an affair for more than eight years with another woman.  That affair became public in January 2010 when billboards highlighting the relationship were displayed in New York, Atlanta, and San Francisco.   Although the revelations were embarrassing for Phillips (who stated at the time that the affair had ended in 2008), there was no apparent lasting impact at Oracle.  Phillips remains in the same position he had when the story broke in January, with his role unchanged.

Ellison clearly values results above everything.  He places business performance at the top of the list for evaluating business people.  He is satisfied with Phillips’ leadership at Oracle.  Similarly, he makes it clear that he believes HP’s board (along with its employees and customers) should be satisfied with Hurd’s performance.

Ellison is right that Hurd has done a remarkable job in turning HP around.  As a business leader, he’s done well for the company.  He’s also right that Hurd’s behavior, regarding the possibility of harassment, was investigated and the board concluded that the allegations weren’t true.

There’s more to it than that, of course.  The expense account improprieties are (by all accounts) real and significant and deceptive.  That clearly destroys some level of trust between Hurd and the board and between Hurd and his subordinates at HP.  It’s wrong.

In addition, Hurd is a married man.  Regardless of whether you believe Hurd’s claims that there was no sexual relationship with this woman, there clearly was a pattern of deception.  Hurd’s marital commitments might not matter to Ellison (who has been married four times himself).  But they matter a lot to me.  That deception is also wrong.

Finally, there’s a concept that might seem quaint and outdated in today’s business world, especially to hard drivers like Ellison: The HP Way.  Bill Hewlett and Dave Packard, the founders of Hewlett Packard, professed a management strategy and style that permeated the company throughout its first six decades.  Packard outlined their core values in the book The HP Way: How Bill Hewlett and I Built Our Company, explaining their commitment to openness, honesty, and flexibility throughout the organization.  The HP Way was a core foundation of values they had established (and adhered to) in running the company from its founding in 1938.  They formalized The HP Way in a written set of core objectives distributed to division managers in 1957 (just a year before the company went public).  Bill and Dave were respected (and eventually revered) for creating a work environment that emphasized trust and valued integrity, while still pursuing profitability.

I worked briefly for HP back in the late 80s.  My wife had joined HP as a summer intern following her junior year in college and she joined as a full-time software developer a year later.  I can assert, from close, first-hand experience that The HP Way was real, genuine, and immensely valuable.  I have a number of close friends who’ve spent all or most of their careers at HP, including a few approaching their 30^th anniversaries.

Mark Hurd and his predecessor, Carly Fiorina, completely transformed the company culture.  Their focus was to maximize growth and profitability at the expense of virtually everything else.  If necessary, the long-term company values espoused by Bill and Dave would be sacrificed in favor of financial objectives.  To the dismay of most long-time employees, The HP Way virtually disappeared as a guiding force and, fairly or not, Fiorina and Hurd were blamed for the change.

Hurd’s recent actions confirm, on a personal level, his antipathy to The HP Way.  He hasn’t acted with integrity.  He has violated the trust of his employees and the board of directors.  In his own words, Hurd “…did not live up to the standards and principles of trust, respect and integrity that I have espoused at HP… .”  He deserved to lose his job.

Maybe, with this change, the HP board is taking the first step to try to restore The HP Way.

I hope so.  I think Bill and Dave would like that.

For now, I’ll leave you with this thought:

Business leaders need to act with integrity.  Always.

Every Business Intelligence (BI) platform starts out being pretty useless because it has nothing inside.  It needs to be “filled up” with some content before it becomes valuable.  Let me provide a few analogies to explain what I mean.

Imagine that you just bought a new GPS unit (or SatNav, if you’re from Europe).  You know you’ll love it because of all the great features: it knows exactly where you are, it can talk to you, it can download traffic information, and it can calculate your speed (and direction of travel).

Now, imagine how frustrating it would be if your GPS unit didn’t come with any maps or location information.  While it can tell you where you are, it doesn’t have any way to tell you how to get to your destination and it doesn’t have any way for you to look up the address of a restaurant or hotel.  No one would sell an “empty” GPS device like that, because no one would buy it.

Let’s apply that same analogy to a slightly different device.  How many phone numbers do you have memorized?  If you’re like me, it’s far fewer than it used to be (and that’s not just because we’re getting old and forgetful either!).  Years ago, we all had to memorize the phone numbers we wanted to call (or resort to looking them up in cumbersome phone books, something that today’s teens have heard of but never actually used).

Telephone manufacturers simplified the process by adding memory and redial features to our phones that we all take for granted now.  If you have a smart phone, you can even synchronize with the contacts in your email system.  That “content” (the lists of phone numbers) makes the device much easier to use.

What does this have to do with business intelligence?  Actually, quite a lot – it turns out that similar issues exist in the realm of BI.

Lots of vendors sell business intelligence platforms.  The most successful and most extensive offerings come from the “Big 4” software vendors: IBM (after they acquired Cognos), Microsoft, Oracle, and SAP (after they acquired Business Objects).  There are some other key players, including MicroStrategy and open source vendors like Pentaho and JasperSoft, but most of the revenue (and, as a result, most of the industry focus) belongs to the Big 4.

Each BI platform, by default, comes with no content.  It doesn’t know where your data is or how best to present it.  It doesn’t know how to organize that data into Financials, Supply Chain, Payroll, Manufacturing, or similar subject areas that will benefit your business.  It doesn’t know how to connect the data in ways that are meaningful for you (like linking an invoice to a purchase order or an employee to a supervisor).  It’s like the GPS device with no maps or the telephone with no address book.

Of course, every BI platform comes with tools to add some content.  It has wizards to point to your database and automatically load details about all your tables (and even find some complicated things like primary keys and foreign keys that help connect those tables).  After running those wizards, you’ll have at least some content.  However, it’s kind of like having your GPS loaded with a bunch of streets.  It could connect them together, but it won’t know a freeway from a country lane, and it won’t know where Starbucks is, and it won’t know anything about speed limits or construction zones or gas stations, let alone the fastest or most direct route to your correct destination.

To make your BI platform work as smoothly as your GPS device (or your cell phone with speed dialing), you need some well-designed “content” that matches your business.  For example, a budget to actual report could tell you whether you’re spending too much or an on-hand inventory report might confirm that you can fulfill a big order that just came in.

Not surprisingly, that custom “content” represents a pretty big market of its own.  The industry experts call it “BI Applications” and we’ll explore that in a future post.

For now…I’ll leave you with this thought:

Your BI tool isn’t very useful if it has nothing in it.  You need to fill it up with some content that will help you run your business.

Lots of my posts here will be about the Business Intelligence (BI) industry.  That’s not surprising, since I’ve worked for BI reporting company Noetix for more than 9 years, so my recent professional focus has been all about BI.  I’ve studied the industry, watched our competitors, worked with our partners, and (of course) paid careful attention to our own customers.

The BI sector of the software industry is notorious for failed projects and dissatisfied customers.  That reputation closely matches the reputation that the CRM (Customer Relationship Management) and SFA (Sales Force Automation) sectors experienced 5 or 10 years ago.  In fact, for many people, that impression persists, even today. [Hmmmm, before I joined Noetix, I was VP of Engineering for a CRM software company.  I wonder what that says about me?]

Obviously, many BI projects (and CRM projects) are successful, but quite a few are failures.  Why?  There are, of course, many different reasons that BI projects (or CRM projects) haven’t been successful.  It would be foolish to try to generalize with a blanket “here’s why” explanation, however there are some common mistakes that many failed projects share.

When organizations face challenges, technology solutions are appealing.  [That’s good for me – I’ve spent my whole career working for companies that build those solutions]  It’s enticing for an organization to just write a check, acquire some software (and, probably, hardware), set it all up, and wait for magical results to happen.  It doesn’t work that way, of course.  Solving difficult problems usually requires organizational focus, careful planning, process change, and managerial skill.

In the late ‘90s, CRM systems were positioned as somewhat magical.  Vendors (especially industry leader Siebel Systems) presented their software as out-of-the-box solutions that would provide customers with immediate benefits.  “Get a single view of your customers, including what they’ve bought, what they might buy, who the contacts are, what support issues they’ve reported, and more,” the vendors promised.  It sounded wonderful and many customers immediately jumped at the chance to implement these new offerings.

In order for the new systems to succeed, the company needed to make some core business changes.  Sales needed to enter the correct information and keep it up to date.  Marketing needed to track (and measure) their campaigns and initiatives.  Everyone in the company needed to share a common “customer” and make sure the data entered was consistent across the board.  So, if one department entered Acme Corporation and another entered Acme Corp and yet another knew them as ACME International, it would appear that there were 3 different companies in the system, rather than just one.

There were many other changes that these systems demanded as well, but for many CRM customers, the importance of those business process changes wasn’t recognized, so the changes didn’t happen.  As a result, the new (and often very expensive) CRM systems failed and the customer was left with an empty bank account and little to show for it.

For now…I’ll leave you with this thought:

Don’t expect a new technology to solve a major problem by itself.

Often, process changes (and attitude changes) are as important (or maybe even more important) as new technology.

The CEO of business software maker NetSuite made news this week when, in an interview in London, he said “I think in 10 years we’ll be bigger than SAP.”   Zach Nelson, CEO of NetSuite, cited the growth of software-as-a-service, the foundation of NetSuite’s offering, as the driving force behind his optimism.

I’ve got nothing against Zach Nelson or NetSuite.  In fact, I agree that he should be optimistic for their future outlook and that the company will continue to succeed against SAP (and Oracle and other big ERP vendors).  They’re growing, they have good customer success stories, and they’re riding a technology wave that’s really hot.

But, “bigger than SAP” – that’s just ridiculous.  SAP is more than 80 times bigger than NetSuite.

NetSuite’s 2009 revenue was $166.5 million, up 9% from 2008.

SAP’s 2009 revenue was $13.63 billion, down 9% from 2008.

Clearly, NetSuite is doing well.  They’re growing at a steady rate, in a troubled economy.  Just as clearly, SAP is struggling and the challenges they face are well-documented.

Let’s make some outrageous projections to see how NetSuite might be “bigger than SAP” in a decade.  If SAP revenue declines by 10% every year for 10 years, it would be about $4.7 billion.  [Reality check: this could never happen.  If it DID happen, don’t you think SAP would do something to reverse that trend somewhere along the way during that decade?]

Here’s a graph that shows some very optimistic growth possibilities for NetSuite, compared to that unrealistic precipitous decline for SAP:

Why make outrageous claims that can never be supported by facts?  NetSuite is doing very well and they deserve a lot of positive attention.  Don’t screw it up by making claims that are just ridiculous.

For now…I’ll leave you with this thought:

Don’t make baseless claims.   They make you look silly.

I’m guessing that many of you are not familiar with data visualization expert Stephen Few.  However, you probably do know who Martin Tyler and Ian Darke are (even if you forget their names before the end of the year). Stephen Few has one thing in common with the other two, and I admire them all for the same reason: they aren’t afraid to tell the truth.

Martin Tyler and Ian Darke are English football commentators (that’s soccer for us Americans).  They were hired by ABC / ESPN to broadcast the 2010 World Cup from South Africa.  Millions of Americans (and others around the world) were thrilled and captivated by Darke’s breathless call of Landon Donovan’s late goal against Algeria that moved the U.S. team on to the next round.

Throughout the World Cup (and, I’m sure, in their regular broadcasts of the English Premier League), these announcers are honest and direct in their assessment of the players and the officials.  After a disputed call in the game between Slovenia and the U.S., Darke called it “one of the stupidest decisions I’ve ever seen.”  Tyler described a disastrous mistake by the English goalkeeper as “one of the softest goals you’ll ever see at this level of football.  It doesn’t often happen in schoolboy play.”  Other memorable quotes from commentators throughout the tournament included “Mexico is right to be indignant – a huge injustice” and “No need for that – very ridiculous.”

Stephen Few operates in a completely different arena.  Data visualization is the science of communicating information clearly and effectively through graphical means.  He emphasizes clarity and simplicity.  In fact, the home page of his company website, Perceptual Edge, includes scholarly quotes about simplicity from Henry David Thoreau, Leonardo da Vinci, and English mathematician Alfred North Whitehead.

Few comments frequently on charts, graphs, and dashboards presented by the media, advertisers, and by Business Intelligence (BI) vendors.  In those comments, he pulls no punches.  When he thinks something is bad (which happens a lot), he says it’s bad.  More importantly, he supports his scorn with a detailed critique describing exactly why it’s ineffective and how it could be improved.  As an added bonus, he writes clearly, filling his detailed explanations with well-placed sarcasm.  [I have to admit - I’m a huge fan of well-placed sarcasm!]

Quotes from Few on his firm’s website are just as direct as those from the English football commentators:

• “How can a vendor that claims to understand data and presumes to teach people best practices in its use know so little? Oracle, you should be embarrassed.”
• “I don’t blame MicroStrategy’s customers. They’re working within the constraints of the tool and emulating impoverished examples, which is probably all they’ve ever seen. I mostly blame the folks at MicroStrategy, who should know better.”
• “SAP BusinessObjects and most other Big BI companies haven’t taken the time to understand data sense-making in general, data visualization in particular, or even the real needs of their customers.”

Too often, broadcasters and commentators share a timid streak with analysts and consultants. They’re reluctant to criticize anyone.  Maybe they fear legal reprisals.  Maybe they’re hesitant to offend a potential future client.  Maybe they’re too closely tied to the people or subjects they’re commenting on.  Maybe it’s a combination of all of those.  In any case, it’s a shame.  The very experts we’d like to rely on for truth don’t deliver—instead they give us sugar-coated platitudes.

In contrast, Stephen Few, Martin Tyler, and Ian Darke provide refreshing honesty and frankness.  I wish there were more like them.

For now…I’ll leave you with this thought:

Don’t be wishy-washy—tell the truth.

In Les Misérables, little Cosette is a 10-year-old girl living a life of indentured servitude.  Cosette is a cleaning girl for the innkeeper, Thenardier, and his heartless wife, who brutally punishes Cosette whenever she slips up.  As an escape, Cosette fantasizes about her “castle on a cloud”—a perfect place where there aren’t “any floors to sweep” and “crying isn’t allowed”.  Instead, a lady all in white holds her and sings lullabies saying “I love you very much”.  Of course, Cosette’s fantasy is just that and is shattered as Madame Thenardier cruelly returns her to her real world.

The emerging excitement of “BI in the Cloud” seems a lot like Cosette’s castle.  Businesses are struggling to deploy new BI systems quickly and efficiently (and maybe the CIO or IT director can empathize with poor Cosette).  A lot of BI vendors are promising that their SaaS (software as a service) BI solutions will come to the rescue.  It sounds perfect— no more “floors to sweep”; instead, the hard work will be done for you.  Unfortunately for many customers, it’s just as much a fantasy as Cosette’s castle.  In fact, the BI industry can’t even agree on what “BI in the cloud” means.

In June, Gartner research published a report titled: ‘Cloud Analytics’ Means Many Different Kinds of Opportunity, co-authored by Bill Gassman and Rita Knox.  In the report, Gartner names six elements of cloud analytics and SaaS BI: data sources, data models, processing applications, computing power, analytic models, and storing of results.  In short, the report’s conclusion is that BI vendors use many different definitions when referring to SaaS BI and this can cause BIG problems for buyers.

Gassman encourages enterprises embarking on SaaS BI to start with vendors that collect and analyze data that already resides in the cloud.  That makes sense, because one of the biggest challenges of any BI system is getting access to data.  The big BI vendors, with their big BI platforms, aren’t really solving the hardest problems when they simply deploy those platforms “in the cloud”.  The Gartner report faults the vendors’ marketing hype for getting in the way of customers understanding what they’re actually buying. [No surprise there…who’s ever heard of marketing causing extreme confusion like that?!]

So, if you’re expecting to quickly get SaaS based BI on the cheap, you’re probably not being any more realistic than little Cosette.  If you start naively down that path, you’ll be snapped back to reality soon enough.

For now…I’ll leave you with this thought:

BI in the cloud is a promising idea, but (at least for today) it’s a dream, not a reality.