|
June 23rd, 2011 by Quaid Saifee
I recently talked to one of our customers who recently implemented QlikView. They bought the software in October 2010, hired us to install the software, training and initial data modeling etc. They have very bright people who took over from there.
I called to see if they were happy with QlikView and how it is being used in the organization. I had no doubt about the answer to the first one, which was “very happy” with QlikView. IT Director at the client was very pleased with how things were going. He said this tool has given the abilities to their sales force which they never had. He told me that normally sales people are very tough to please and are very resistant to change, but in this case things went very smoothly. He said one of the important feature in helping them transition to QlikView was its export to Excel feature.
When I asked him about ROI. He said it is difficult to come up with the specific numbers. He said “Benefit of using a tool like QlikView is an ROI in itself”. He said our implementation was pretty quick and one should also consider “time to market” in the ROI. He promised to delve further and see if he can come up with a tangible ROI.
When a tool has fulfilled unmet need and is helping the business become more profitable and users love it, I wonder in that case is tangible ROI even necessary?
Posted in Business Intelligence, General, IT, QlikView | No Comments »
May 10th, 2011 by Quaid Saifee
Data Discovery is the new-term, which is being used by Gartner for new breed of BI. This breed is faster, cheaper in terms of TCO (Total cost of ownership) and better.
In its latest BI Magic Quadrant report, Gartner has bifurcated BI into Traditional (read old) BI Platform and Data Discovery Platform (read new). Data Discovery platforms are all the rage and that’s where all the excitement is.
Three main heavyweights (though they are pretty lightweights compared to Traditional BI platforms) in the Data Discovery Platforms are
- QlikView
- Tableau
- Spotfire from TIBCO
QlikView is the leader of the pack and has great momentum behind it. One can also see that in its stock performance.
From what I see in the marketplace, most of the Fortune 1000 companies are not going to switch to this new BI platform but is in the process of adding that in their BI tool mix.
I have seen it again and again, that when these tools are shown to IT and business users, they get dreamy eyes. They have not seen anything like this before.
Once they play with it, they demand that to do their job they need a tool like this and cannot live without it.
Though I must admit we have run into situations where IT has considered the new BI tools as a threat to their job and hence kept it out.
One thing is for sure traditional BI vendors are not standing still, they are coming up with data discovery tools of their own, example being PowerPivot from Microsoft and Explorer from SAP.
Are you seeing the same excitement around “data discovery platform”?
Posted in Business Intelligence, General, QlikView, SAP | No Comments »
April 26th, 2011 by Quaid Saifee
BI appliances:
Everybody (well almost everybody) BI vendor has a BI appliance now. Most of the independent appliance vendors have been acquired in last couple of years by mega vendors.
- Microsoft acquired DATAllegro in 2008 and recently released HP Business Decision Appliance
- IBM acquired Netezza in 2010
- EMC acquired Greenplum
There are plenty of other appliances out there, just to name a few prominent one.
- Teradata
- Dataupia
- Oracle Exadata Database Machine
- Business Intelligence Accelerator from SAP
- HANA from SAP
- Icebreaker BI appliance from Ingres and JasperSoft
I found this very useful presentation on appliances, if you are interested in finding more, click here.
Though I wonder with memory becoming cheaper and cheaper and with 64 bit platform, why do you have to have a special appliance? Why not use an in-memory tool like QlikView with tons of RAM ?
Data Mining and Predictive Analytics
There are very few BI vendors who have been able to integrate Predictive Analytics with their BI tools. SAS has been a leader in this for long time and continues its #1 position. Business Objects (before SAP acquired it) had started working with SPSS (before it was acquired by IBM) to bring Predictive Analytics to its reporting tools (WebI), it never got off the ground. Spotfire from TIBCO is another BI / data visualization tool which has done a good job in integrating Predictive Analytics with BI tools.
All other tools are pretty standalone Predictive Analytics tools; R, KXEN, SPSS, Portrait Software, Angoss Software etc.
Despite the demand for Predictive analytics, the key reason companies have not been able to widely adopt these tools, is the lack of qualified people (Masters and PhDs) to do the job. Open Source R tool is being integrated by several major BI vendors in their toolset and in the next couple of years we should see those efforts come to some fruition. If your organization has been involved in some data mining, you can help the industry and yourself by taking this survey from Rexer Analytics
Open Source
Open source BI tools like Pentaho, Jaspersoft, BIRT (open source project of Actuate), Talend (ETL) have come a long way, but they still cannot match the functionality of regular licensed BI products. Software might come for free, but the cost to implement depending on the scope of the project, can run from tens of thousands to Millions of dollars.
Also it is hard to find talent, who know these tools well. Most of the organizations do consider open source tools but eventually decide to go the other way because of lack of feature and functions and availability of talent.
Posted in Business Intelligence, General, QlikView | No Comments »
May 14th, 2010 by Quaid Saifee
Sorting wheat from chaff in BI? What’s must, good to have or useless depends on your needs.
Well like everything 2.0, BI has its own 2.0 features and functions. Some of that is old stuff in a new package. Let’s go over the list of new features / functions / technology in BI.
- Mobile BI: With the advent of smart phones and particularly iPhone, this is one technology feature which BI software leaders cannot afford to miss. Remember just showing the same reports/dashboards on the mobile screen is not Mobile BI. Mobile BI is still in its infancy. Here is a good summary on mobile BI progress over past few years.
* added on April 30th, 2011 (Mobile BI to become the Norm – Information Management)
- Data Visualization: I wish we had the dashboards when I was working in the railroad industry years back. I remember most of the dispatch staff would print tens of pages to look at few numbers, to highlight them and store them in a special folder. Charts, graphs etc on a dashboards are not only making executives interested in getting the information they need but are also helping middle level manager and operational line worker. Why look at 500 page report to find a number, which can be shown in a dashboard, finally management by exception is becoming the norm. No BI stack is complete without dashboard and there are many niche dashboard vendors in the market now.
- BI in the cloud: Like everything else BI can be on the cloud as well, but should you put it there and why? Bob Mitchell has a good piece in Computerworld on this topic
- Analytics: Business Analytics Vs Business Intelligence, most of the BI vendors have tried to blur the lines between the two. I have rarely seen a power user without the power of Microsoft Excel even when they have the latest and greatest BI tools at their disposal. The fact remains that their analytics users in the companies who need to analyze data never analyzed before (not in the datamart) and ask the questions never asked before. Rock Gnatovich wrote an article regarding this in CIO magazine.
- Search: Search should of course be a natural part of any BI tool. It is easier said than done. Most of the enterprise BI tools have basic search functionality like searching for a report with specific title or even a field in the report. Also there is also a whole new category defined by Gartner called “Information Access Technology”, where the products bring the structured and non-structured data together for analysis, there are specialist tools like Endeca which are leaders in this category.
- In-memory BI: This is one of the hottest term in the BI field and for a good reason, who does not want faster response from their BI applications. Nigel Pendse has written a great commentaryon this, very interesting and it puts things in context.
Coming up in part II
- BI appliances:
- Data Mining
- Predictive analytics
- Open Source etc.
Quaid Saifee
Posted in Business Intelligence, Cloud, General, IT, SaaS | No Comments »
February 12th, 2010 by Quaid Saifee
All of us have seen quite a few articles written on best practices to accomplish Business Intelligence projects. Couple of years back when we were involved in a BIG BI project for a Fortune 100 company, those articles did not help.
As we went over lessons learnt from that project, I realized that it is more important to emphasize DON’Ts rather than DO’s.
I am making the assumption that you are embarking on a decent size (5-8 resources, multi-hundred thousand dollar project, fixed cost project). The key points should be applicable whether you are doing this project internally or hiring a consulting company to do it for you; doing this on fixed cost terms or on time and material basis.
- Don’t start without exact specs and timeline (clarify that fixed cost depends on these two factors), this does not mean that one cannot modify the scope or implement iterative development methodology, but broad specs should be well defined before the start of the project
- Don’t start with more than couple of business processes (Ralph Kimball)
- Don’t start without blessings from the business users and IT staff (adversarial relationship with anybody in either of these groups can result into a disaster)
- Don’t start without identifying key business sponsor and key IT decision maker (and daily/weekly time commitment from them)
- Don’t start without a functional requirement document in place
- Don’t start without a designated / dedicated project manager (qualification and quality of the project manager is key)
- Don’t compromise on the quality of the resources (delay the project if need be if the needed resources are not available)
- Don’t follow the big bang theory (show incremental progress, this should be part of project plan)
- Don’t staff fully right from the “get go”. (Technical: PM, Data modeler, ETL architect, Sr. reporter; Business: Power users)
- Don’t say Yes to everything client (internal or external) wants, make sure to convey your opinion about the implication of changes on cost/timeline (in writing)
Got more Don’ts, let me know!
Quaid Saifee
Tags: BI, Business Intelligence, dashboards, data warehouse, ETL, reporting
Posted in Business Intelligence, General, IT | No Comments »
January 22nd, 2010 by Scott Smith
As the first blog of hopefully many, I thought I would write about my chosen line of work – Business Intelligence, or more comfortably, BI. Given that you’re taking the time to read this, you likely already have your own ideas of what BI is all about: data warehouses, reports, dashboards; basically any set of processes and technology that provide timely business information to end users.
But what does BI really mean? Let’s see what my good friend www.dictionary.com says about BI:
Business Intelligence
- the process of gathering information about a business or industry matter; a broad range of applications and technologies for gathering, storing, analyzing, and providing access to data to help make business decisions; abbr. BI
Interestingly, I see the word “business” a couple times, but no mention of “intelligence” anywhere. Hmm…I wonder why? I think it’s because intelligence is in the eye of the beholder. Let’s go back to the online dictionary to see what they say about the word intelligence:
Intelligence
- the capacity for learning, reasoning, understanding, and similar forms of mental activity; aptitude in grasping truths, relationships, facts, meanings, etc.
So my read on that definition is that intelligence is about mental capacity, the ability to interpret facts, and aptitude. Nowhere in the definition does it say anything about what to do with intelligence, assuming you’re blessed with it in the first place. In other words, intelligence is merely the foundation for good decision-making.
And therein lies the rub: To quit the wordplay for a moment, this definition of intelligence is exactly why we, as BI practitioners, seem to struggle with some of the basics of justification for what we do and sell, like return-on-investment for BI, and its value compared to other IT investment options. Too often, the details of a BI investment are left to IT, focusing on providing BI capability back to the business – without the necessary partnership to ensure that BI actually produces net positive business results and real ROI.
Let’s take one last trip to the dictionary and perhaps look for a more appropriate moniker:
Wisdom
- the quality or state of being wise; knowledge of what is true or right coupled with just judgment as to action; sagacity, discernment, or insight.
Wisdom, using intelligence as a foundation, provides access to the truth, good decisions, and the experience to take appropriate actions when intelligence alone in not enough.
Business Wisdom: I like the sound of that. Want to buy some?
Truly Yours,
Scott
Posted in General | No Comments »
|
|
|
| |
© 2005-2011 Wit Inc. All Rights
Reserved. Site by:
|
| |
|
