Gartner Releases 2007 Magic Quadrant for Data Warehouse & Database Management Systems

 
My previous engagement had a propensity for buying every possible BI enterprise grade platform on the market, so I can attest to Microsoft, IBM and Teradata’s entry. My exposure to Oracle is minimal, but that which I am exposed (CDC), is really among the best out there. I am biased to Microsoft’s BI stack, though they recently slighted me regarding their Customer Conference in Sun Valley this year.
 
Personal feelings aside, for what you save in cost with the MS licensing and server fee, you make up for in prerequisite software requirements, implementation fees and consulting services (mainly the latter) because you have to forcibly make it work the way you want it to or in the way you were sold it would work, yet were disappointed when it didn’t turn out to be the be-all end-all product you thought it would. ROI and BI is almost a joke, where the latter is starting to feel way too ERP-like or CRM-like for my taste. MS offers a familiarity that the others do not if you are not accustomed to Crystal reports or MYSQL, because most of us in corporate America understand concepts like the "breadcrumb" trail…But, that is a misnomer. Another urban legend sold by $160-200/hr consultancies while the product was still in beta. Buyer beware…Though participating in a Beta can be fun and lucrative, you will most likely get to know and love the Beta right before MS releases the product to market with a fraction of the features that you really cared about. To make matters worse, there is a lost cost in the customization of the Beta product when you have to invest a laborious amount of time uninstalling and reinstalling the components (in the correct order no less), often involving registry keys. DO I need to say more? I certainly wouldnt want my grandmother calling regedit32 – would you?
Oh and did i mention that often those custom apps that were built by the fancy shmancy consultant often break after they leave, and while most of us are used to the "maintenance contract" or when you have to pay more money to have them return to research and fix what they built in the 1st place. I wouldnt mind paying for the fix, as organizations change metadata so often, that I can see how a custom built solution would need updates time to time. But to come back and research for a cost, shouldered by the client as well? Don’t I get fries with my combo meal?
 
Before you get excited about the lower price point, remember to ask the consultant who you will inevitably need the following stack of questions:
 
-How much is your maintenance contract, if not included in the price?
-How many times on average did you have to return to a client to fix a solution you developed?
-What was the average duration (in days) of those fixes?
-Do you have any reduced rate maintenance plans if we contracted multiple years at once?
-What happens if you cannot solve a feature set promised in the scoping (i.e. too difficult, can’t, won’t admit they can’t etc) or you don’t get those expected results and feature sets as promised by the vendor product material? Will you comp that since we took a risk at your recommendation to work with a beta.
(Get that in writing).
 
Don’t get me wrong. I am a consultant too. However, the sweet spot for me if the many organizational hats I have worn./ BI, Scorecards, Dashboards, Six Sigma, Call Center, R & D, Statistical Modeling, Planning, Analysis and Budgeting are all areas I have worked in my career. By understanding both verticals and horizontal streams, one can scope and analyze the true nature of the problem being solved. SO many times, consultants, having more knowledge than yourself, (let’s face it, if they didnt, they wouldnt have been called in by your boss) or having the bandwidth that you dont (but mostly the former), so they assume they can tell you a technical solution, throw together a loose rendering of the star schema they plan to implement, possibly a data flow diagram if you are lucky and even more remotely, a security/auth logical flow model for your IT security ops team, so that you are wowed by the time they spent in Visio making the boxes "look pretty". Oh and slow down the pen from signing if they rendered the drawings in 3-D. I have seen more 3-D process maps and data flow models drive the pen to the ‘Sign here’ line, before the CIO really could take a look at the content.
We all have to make decisions at a rapid pace. And believe me, I love our MS platform (PerformancePoint Server 2007) delivered via Sharepoint/MOSS 2007 to both internal and external customers/partners. But it came at a cost. A $1.5 million cost when all was said and done. And most of that (94%) was consulting services, which of that cost, nearly all but 1 scorecard was broken within the 1st and 2nd years post deployment.
Not to mention to add. cost we had to spend on top of other add. costs from addendum statement’s of work for the documentation on how to use this customized non-functional thing.
 
Moral of the story is do your homework. The right consultants are out there with experience in deploying performance management solutions. Don’t just go based on customer testimonials. Believe me, they really butter you up so that you will give a good reference. Ask the questions I listed, .GET THEM IN WRITING., and do your homework (read blogs, search the Internet), and if you are going to go by testimonials, dig into their maintenance and frequency of down time. We were unable to sustain the 90% up-time goals we had set in year 1 prior to paying more money to have them return to upgrade the work to a better, more stable architecture, blaming the database dev who initially scoped it out.
 
If you get the right group of consultants, you will end up the winner with the MS toolset because of cost. If you have a limited budget, go with MS, but only buy a component and it’s prereq. one at a time. Get it to work. Train your end users and measure adoption. Then, add on to the stack by procuring other licenses.
 
If you have a decent sized company or even a medium to larger size org, go with Business Objects Edge (medium company size) or Enterprise (like the name sounds size). And dont forget to add on the BO Voyager…It allows you to use their presentation tools on top of existing OLAP structures (Analysis Services cubes), so that you do not throw away previous work. Presentation wise, hands down, BO is the winner.
 
The Performance Management suite (Analytics), and Dashboard tool (which is now part of the IDD product), are far more robust and rich with data visualization schema’s out of the box. And the BO BI stack has just plain been doing it for so much longer than MS, who really was one of the last big companies to delve into this long-outstanding technology sector. Props for what they have done. Just apply usage carefully. Build your roadmap and analyze whether you should stay with one stack or diversify. And keep a watchful eye on anyone building anything for you.
 
 
 
 
 
 
Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s