Throughout the years, I have been witness to many Data Warehousing projects that started as an effort to help the business users make business decisions using information and along the way
transformed into data centralization exercises. Somewhere during the project execution, the team lost focus on what the organization needed to be more efficient and productive, and became tangled
in the complexity of the data. The Data Warehouse became an information bank, storing all valuable data elements into an electronic vault.
The expectations from the project teams were very similar, all making assumptions under the principle “If we build it, they will come”. The teams believed that if they could put enough valuable
information in a single location, the business users would come knocking on the door, begging to be given access to the information. Reality prove them wrong, as the expectations from the business
users grew up disproportionately given the amount time it was taking IT to complete the project. The business users expected a silver bullet, kind of “click here and your information problems will
be solved button”, but instead they got a mountain of disorganized data. As a business user colorfully put it: “(finding information in the Data Warehouse) is like going through a huge bin of
laundry trying to find your favorite pair of socks. You know they are in there somewhere, but when you can not find, you really start getting frustrated.”
If anything, the early Data Warehousing efforts did make the business users realize the value that information could bring to the operation of the business. The business community also concluded
that waiting for IT was a long and tedious process and at the end they were not getting what they needed. The solution was simple, many business teams started hiring their own “shadow” IT
organizations, bringing people who had the skills and knowledge of the systems into functional areas bypassing IT controls and governance. The irony came as the Data Warehousing effort was supposed
to bring the data into a single location, but these spawned initiatives looked to solve specific business problems by creating small functional data marts across the enterprise. Some other
organizations went as far as to look for outside help, asking 3rd party companies, and some times their own vendors to provide insight into the operation of the business by looking and analyzing
the data. Although “more responsive than IT,” these functional silos became very expensive to maintain and operate as they did not encourage knowledge sharing, resource pooling or
reusing components across departments. Furthermore, as people were using different methods and sources to retrieve and calculate the data, it was a rare occurrence for the numbers to match.
After some time, someone invariable would question how much money and resources were dedicated to these internal/external reporting and analysis teams. The findings were always shocking; in many
organizations, these informal IT groups were bigger than the official IT Business Intelligence team. People would slowly start to realize that they could merge all these teams into a single
Business Intelligence (BI) team with a significant impact on the bottom line. However there was still some hesitance to form a core team as they did not want to repeat the errors from the past. A
new, fresh approach was needed, and so the concept was born: what if IT could provide “The Platform” that enabled the Business for Self Service across the Enterprise. The idea was intriguing,
what was lacking was the know-how to make it from concept to practice. This article will focus on defining the concept of enabling a Business Intelligence Self Service Platform in the Enterprise
and then laying the foundation to make such an idea a true possibility in any organization today.
Defining a Business Intelligence Self Service Platform
from a business perspective
We can easily define “Business Intelligence Self Service” as a service provided by an open Business Intelligence platform that enables business users to access the information they need by
themselves, using an easy to understand User Interface (UI) that is defined in business terms and not IT jargon. Perhaps the most critical feature of such a service is to eliminate the “IT touch”
for the users to create new reports; the platform should support the capability to combine existing data elements into a custom report. It goes beyond providing ad-hoc capabilities to truly enable
a creativity engine that makes it possible for the business users to follow the Information Analysis Cycle (Figure 1) quickly, by themselves, and without calling for IT support.
Click to enlarge
In order to better explain the concept of a self Service BI platform, let’s follow the Information Analysis Cycle with a typical Retail Merchandising scenario:
A cake mix vendor, MiaCake, is in a grocery retailer buyer’s office arguing his company needs to raise prices 15% because of high fuel prices. The buyer is skeptical as he does not know the
effect this price increase will have on the consumer behavior. He starts conceptualizing how to measure the potential effect. He remembers that three months ago the competing brand was on sale and
the different in price was about 15%. He writes a report comparing MiaCake’s sales and the competing brand sales during the week the competing brand was in promotion using the company’s
BI platform, that by the way enables him to create his own reports and analysis on demand. One minute later with the seller still at the buyer’s office, the report comes backs and shows
sales were down more than 30% for MiaCake, as consumers quickly shifted to the other brand. The vendor is shocked by the results, but can not argue against the data being show in front of him. He
then calls his manager and decides to raise prices only by 5% in order to maximize revenue.
As depicted in the example above, timing is everything. If a Self Service Platform had not been present, the report might not have been possible to build without IT intervention. The vendor and the
buyer would have had to resort to non-factual arguments probably leading to a less profitable scenario.
Why enable a Business Intelligence Self Service Platform
in the Enterprise?
As discussed in the previous chapter, enabling a Business Intelligence Self Service Platform across the enterprise can have a positive financial impact in the organization. Once business users get
more familiar with the platform, and realize how easy it is to create and execute reports on their own, they will start changing their habits to use facts to make decisions instead of gut feelings;
thus enabling the next level of enterprise maturity by laying the foundation for creating repeatable/reusable strategies that can be leveraged and shared across the organization in a systematic
fashion. A common framework can be created that allows interdepartmental communication and collaboration to socialize the findings and strategies that resulted from the new capabilities; providing
management with the tools they need to establish common goals, and define the base methodology to achieve and measure those goals.
In many companies, this Business Intelligence platform has become the engine for positive change reaching into areas that were not traditional information customers, transforming their day to day
operations, and making them more efficient and objective.
IT will also perceive a significant change. As it becomes a capability provider, it is teaching the business users how to fish, instead of IT fishing for them. This paradigm change will free the
resources that were previously consumed by servicing individual information requests and redirect them on growing the number of services offered by the platform, thus enabling IT to become a
strategic partner with the business.
How to Enable a Business Intelligence Self Service Platform
in the Enterprise?
Enabling a true Self Service BI platform has to be done across three different components: People, Process and Technology (Figure 2).
Each component highlights a different focus that complements the others:
In order for the platform to be functional, it has to be implemented with a complete Business focus in mind; if preferable someone from the business should lead the initiative and clearly define
how the new platform is going to impact the operations of the business. Furthermore, change management strategies should be clearly outlined and the project plan should allocate sufficient time for
them. The business lead should establish a vision for the future that energizes and set the expectations as to what to come from the platform. A training plan should be designed that takes into
consideration the different skill levels of the users to make sure all the users will be able to get the most of the new capabilities and they won’t resort back to old habits.
Once the direction of the initiative has been set, it will be critical to align the technology components to support it. Establishing a common infrastructure, both on the database platform and the
BI toolset suite will be vital to the success of the initiative. The database platform will have to provide a stable, scalable environment that can grow with the users, adding capacity as the
business matures and as the data volumes grow. A robust data model will be required that can capture the nuances of the business, and most important how do they relate and fit into the Enterprise.
In order to overcome the technical complexities of the data and facilitate the use to the business community, there will be a need to establish a common metadata business layer. This metadata layer
will isolate the user community from technicalities and enable them to use the data based on its business meaning.
Canned reports will have to be set-up as examples on the capabilities of the platform; these will be used as the base for subsequent analyses, thus providing the first level of analytics out of
Last, but not least, the BI platform will have to be robust enough to provide a graphical report builder that facilitates the business users to create their own reports. Furthermore the BI platform
will need to accommodate advanced capabilities, such as:
- Advanced drilling: the capability to navigate across the data either up/down or across hierarchies
- Real time/ near real time data: the concept streaming information as it happens to minimize the time the users need to wait for the data being available.
In order to bind the technology and the people together a process has to be established. To successfully kick-off a Self Service BI platform, a common, well defined process will be required to
provide the initial support to the end users. As IT won’t have the business knowledge to help the business users translate their information needs into reports, a hybrid group, the power users,
will need to be created. This group will be composed of business users who are early adopters and understand the technology, as well as, the business needs. They will be the liaisons between the
business community and IT. IT will be responsible for maintaining the platform and adding additional data elements from different functional areas.
A straightforward process should be published that outlines the different levels of support, and so the business community knows who to contact and what to expect from this interaction.
How to measure success implementing a
Business Intelligence Self Service Platform in the Enterprise?
Defining clear goals is perhaps the single most important action item while implementing a BI Self Service platform. Key Performance Indicators (KPI) will have to be baselined against
organizational goals (e.g. Out of my user universe how many use BI daily, weekly, monthly at all?) These measurements will highlight strategies to promote the use of the platform, and benefits
across the enterprise.
One the favorite strategies to promote the use of the platform is to focus on specific initiatives, measuring before and after, and if possible have the champion of the initiative give a lunch
& learn to the user community; reporting first hand the positive impact the platform has had in her/his business. (e.g. a deli Vice-President decided to measure the sales of rotisserie
chickens, she found that the stores were not making them as they did not know how many to produce at a particular time, after running a couple of reports on sales by hour she was able to put
together some guidelines as to how many chickens and when to cook them. The results were amazing, sales went up by 5,000% in one week, the store associates were impressed by the knowledge the VP
had shown about the business, and their specific store) The rotisserie chickens story became an urban legend within the company, inspiring, and energizing people to match the deli success in
There is no more powerful way to measure results than in old, plain dollars, if possible try to quantify the benefits in hard dollars. This way everybody across the Enterprise understands the
scale, and there will no doubt about the Return On Investment of the initiative.
Implementing a Self Service Business Intelligence platform in the Enterprise is not a simple task. It requires aligning People, Technology and Process to a Business vision that needs to be strong
enough to bridge the gaps among different departments and have them collaborate single mindedly. However the benefits of undertaking such as effort can be substantial and can positively influence
the performance of everyone in the organization.