Business Intelligence

Best Practices in Business Intelligence

This article also has some best practices for Business Intelligence from the Director of Continuous Improvement at The Scotts Company, Ram Balasubramanian:

1. Focus on common sales data. The approach we took was to concentrate on common sales data collection activities that would work for all five of our major customers and meet roughly 80 percent of the needs required to capture that data. Once we ironed out these baseline processes, we focused on customization efforts to cover specific situations, such as specific attributes related to each retailer, automation of data capture, and mapping based on individual retailer systems, and so on.

2. Establish repeatable processes that you can reuse over and over again across your customer base, and you will run a much more efficient and less costly BI operation. The initial investment will be high, but the resulting infrastructure will be worth it. It will be quick and easy to bring additional sites on board for a small fraction of the original investment cost.

3. Ensure the accuracy of your data. Because data accuracy is clearly essential for every BI infrastructure, we maintain a close collaborative relationship with the master data team (all bad data eventually surfaces in the reporting area). In fact, we often are able to help the team take proactive steps to maintain clean data by establishing standards.

4. Expect your operational teams to use the SAP BI platform on a regular basis, but don’t expect that senior management will be eager to log on to a BI platform. You will have much more success directing this group of users to access the BI platform through a portal, which offers executives standardized access and one-stop shopping for all reporting activities.

5. Train power users from all participating departments early on in the project. These power users can then go back and train their coworkers (business users) and also help with the formulation, execution, and trouble-shooting of queries. This frees up IT teams to focus on more technical issues and to step in only when the power users are experiencing difficulties.

6. Encourage regular dialogues among your SAP BI power users. We hosted weekly open forums for power users so that they could share tips and tricks with one another for building better queries. This practice has built a vibrant BI community across Scotts and has been a catalyst for self-sufficiency among the user group. (could use wikis on corporate portals to interact and build a knowledge sharing platform)

7. Keep in mind that SAP BI offers a sophisticated data modeling environment. It’s up to you to strike the right balance for your data models. Make them too simple, and you can’t get everything you want out of them. Make them too complex, and you effectively saddle yourself with a point solution.

I think these points from Ram are extremely useful for any BI system plan! 

Tangible Benefits from BI Implementations!

Very often the tangible benefits of implementing a successful BI platform are not clearly understood. I am reading the Netweaver Magazine (Primier Issue Vol. 1) and it has a case study on The Scotts Company. The benefits that the company experienced within 24 months were amazing!

Before BI Implementation:

- Limited Visibility into its channels
- Unreliable Forecasts
- Excessive Stock Levels
- Annual Inventory Turns less   than 2
- Customer Fill rates less than 90%

After SAP BI Implementation:

- Annual Inventory Turns Over 4.5 (likely to touch 5 soon)
- $99 million reduction in Inventory in the channel
- $158 million increase in Cash Flow
- Doubling of Utilization of Scott's Manufacturing plants
- Reduction, by nearly half, of the number of mixing warehouses
- Customer Fill Rates over 95%

The transformation is quite obviously tangible as well as substantial that adds to the overall Shareholder Value of the company!

Now, how did the BI help the company achieve all this?  Due to the following:

1. Enabled accurate and real time enterprise-wide view of sales, inventory, and customer service metrics at the customer, regional and brand level
2. Enabled detailed evaluation of all areas of order processing and customer service.
3. Provided near real time sales data to merchandisers helped them with timely execution of promotions.
4. Effective Use of data helped the company to replenish inventory for customers and strengthen strategic collaborations with retailers.
5. Analysis of data helps target media spending and marketing efforts at regional and national levels
6. Helped understand the sales trends during a season better due to season-breaking BW reports.

Very often the companies want the new BI system to fit into their existing processes!  Little realizing that such a time is the best opportunity to take the efficiency and value of the business several notches up!

Dichotomy between Legal and Management Reporting

 

"That's Ok.... our profit centers roll upto a Company Code" he told me in response to a question on why the Legal Accounting Consolidation process was using a Profit Center as the basis?

Profit Center represented Business Units (BUs) and the BU managers used to "game" their management reports to give a good "view". Well, but the Legal Accounting data could not be "gamed" - as it had to be accurate. But they told me that everything annuled at the Company Code level as the Profit Centers rolled up. May be I thought.

The company was trying to implement SAP's Consolidation product called Strategic Enterprise Management (SEM)-Business Consolidation (BCS) and had deliberately created just one "Consolidation Unit" called Profit Center! That may look fine and dandy in a set up where the Accounting Manager did not want any "redundant" organizational levels to be used for consolidation purpose than were absolutely necessary, but was it the correct decision? Such an action increased Reconciliations at the close time!

One point that everyone missed in the entire episode was the basic and foundation principles of Reporting and Consolidations!

1. Management Reporting, by need had to be Flexible to accomodate the needs of the line managers!
   

2. Legal Reporting could not Afford to be flexible!! It had to be accurate to a degree where the CFO could sign his name upon!
   

3. That is why, Legal Reporting should not be held ransom to the Management Reporting blocks!

   At the outset of your reporting initiatives, please do remember that Legal and Management information needs and frameworks are different and it is not prudent to mix the two worlds and processes much! Or it can come to haunt you big time!

 
  Technorati : Accounting, Legal Reporting, SOX, Sarbanes

Tactical Follies in Implementing BI Projeects

Interesting case study on the "Change Management" initiative at John Hopkins.  Read carefully the issues that are foremost in the minds of the employees.  Reassurance is the main one!

About 20 Johns Hopkins IT managers clustered around a Mount Washington conference table recently for one purpose: to learn how to prepare their employees for their user roles and training in the new SAP software system being implemented by HopkinsOne.

They angled their faces toward Pam Somers, senior specialist from Organizational Diversity and Development, as she tried to allay their apprehensions: "What do you really want during that 'doom and gloom' feel [as the go-live date approaches]?" she asked.

"Chocolate," someone blurted out.

"Reassurance," said another.

Somers didn't have any chocolate, but she and the HopkinsOne team did provide support and reassurance during the two-hour meeting on July 13, the second of more than 85 HopkinsOne manager preparation sessions taking place across Johns Hopkins between now and mid-September.

These "Nuts and Bolts" sessions give directors, managers and supervisors the information and materials needed to understand how SAP will impact their organizations and to prepare employees for fall training courses for SAP access.

Sadly, these days, Change Management is getting to be forgotten in the large scheme of the project implementations.  Many companies are bypassing the bigger consulting firms and going to the smaller implementors or simply undertaking the projects on their own through employees! 

That has meant that the rigor that was inherent in an implementation taken by a Big Four is gone. Business Intelligence is an area where the change in the system is the largest when you go to the larger ERP softwares.  Hitherto, the folks had been using the good old MS Excel and now suddenly they are led into the world of OLAP and Infocubes and Datawarehouses!!  Easy to scare them!! 

Moreover the entire paradigm changes in such scenarios!  Performance Management project - comprising of Reporting and Planning - in such a scenario should be approached first as a BPR project to completely revamp the processes.  If that does not happen, then the tendency of the consultants - even the larger companies - is to slam the Excel model into SAP or Oracle or whatever have you!  So, basically, your data model had been decided more than a decade ago!  Does that sound right?  But that is precisely what happens!

Business Intelligence is a greater exercise than just plugging in the technology.  It is about getting the maximum out of the data that a company and its environment generates and can lay hands on to improve the Shareholder Value of the company! 

Sadly, the CIO office which really implements such projects are devoid of any strategic finance thinking.

Ineffectiveness of too many choices and data!

 

Are a lot of choices for the consumers always a smart marketing strategy? Apparently yes as the traditionalist American marketing Gurus may have us believe. But research tells us otherwise. Sheena Iyenger, Assoc Prof at Columbia University, did research on the affect of increasing the flavors of jams in a grocery store and its sales figures. Similarly the satisfaction from a choice of the same Godiva chocolate seemed to be more when taken out of a pack of 6 rather than a pack of 30!

To explore consumer responses to extensive options, we conducted a field investigation in an upscale grocery store, Draeger's, in Menlo Park, Calif. On two consecutive Saturdays, a tasting booth for Wilkin & Sons exotic jams was arranged. As consumers passed the tasting booth, they encountered a display with either six or 24 different jams. We observed and calculated the number of people who stopped at the tasting booth as well as the number of people who chose to purchase the jam in question.

The results are striking. They demonstrate that although extensive choice is initially more enticing than limited choice, limited choice is ultimately more motivating. In fact, 60 percent of the passersby approached the table in the extensive-choice condition, as compared to only 40 percent in the limited-choice condition. However, 30 percent of the consumers who encountered the limited selection actually purchased a jam, whereas only 3 percent of those exposed to the extensive selection made a purchase.

In subsequent studies we found that people are actually less satisfied with the choices they make if selected from a larger set of options. For instance, the same Godiva chocolate chosen from a set of 30 chocolates is considered to be less delicious than if it is chosen from a set of six. Moreover, we found that the negative consequences of too much choice extend beyond consumer contexts to work contexts. An examination of individuals completing a task chosen from a larger range of options as compared to a smaller set of options revealed that people performed better at their chosen activities if they have chosen the activity from a smaller range of options.

It seems that data and choices have a similar relationship to decision-making and its effectiveness when it comes to the volume. Increasing data and choices initially increase the effectiveness and accuracy of the decision. But after a certain point - which comes fairly early in the volumes game, the effectiveness and accuracy starts declining. So basically its a classic bell curve!

It is therefore not always the smartest of moves to go for higher amounts of data to gather more information for your decisions.

As Tom Peters also discusses in his blog-post that strategy is not as important as execution is. Which essentially points to the habit of analysis-paralysis in most companies. The amount of time taken to analyze a situation is just crazy!!

Remember, there are two things important in any decision - Time and Analysis.

Do not Analyze so much that the Time becomes irrelevant.

 
  Technorati : Choices, Columbia, Iyenger

BI opportunity in Retail Users in the Next Two Years

 

Is Business Intelligence coming of age? The retailers are going to invest in a big way. Retailers need to data mine the transaction and PoP data they generate to judge and predict trends. The first step for that journey needs to be a robust and reliable data warehoue and analysis system. Needless to say that the opportunity in BI is huge!

Seventy-six percent of retailers use or have active, budgeted plans to use business intelligence in some form or another within their organizations within the next couple of years, according to a new report from Aberdeen Group. Yet only 21 percent of retailers indicated that they have had enterprisewide business intelligence systems for at least two years.

The report, "Business Cohesion in Retail: Bringing Cohesion to a Fragmented Enterprise," states that average retailers have largely missed opportunities to make better decisions by viewing data on an ad-hoc basis, rather than at regularly scheduled intervals.

 

Report author and Aberdeen research director Greg Belkin says in the report that many retailers are still siloed in terms of their data, so enterprisewide business intelligence is impossible until enterprisewide views of data are available. Belkin adds that traditional merchandising processes are in the need of a serious overhaul to add more in terms of predictive analytics "to ensure last year's results don't become this year's self-fulfilling prophecies."

Breaking down the silos is no simple matter. "Smaller companies have as many as 10 different data marts, while firms over $1 billion will have as many as 50 data marts within the company," adds Leslie Ament, Aberdeen channel director for the customer intelligence practice. Large retailers tend to be near the upper end of this range due to the array of different suppliers.

Supply chain complexities are a big part of the challenge in breaking down these silos, Ament adds. Because different retailers have different supply chain intricacies for discounts, billings, item names and other data records, there's no simple way for a business intelligence vendor to meet the needs of the market.

Belkin recommends that customer centricity should drive any further investments in business intelligence. Those retailers who are "best in class" are already using business intelligence across the enterprise, but these retailers must avoid stalling at current levels by continuing to invest in the technology.

The first step for any retailer looking to implement business intelligence, according to Belkin and Ament, is to "get the house in order"; in other words, to understand the value proposition of business intelligence and to determine where it can provide immediate return on investment.

 
  Technorati : BI, Retail