So, check this story now.

It is a Sunday evening on the Diwali eve, and I am at a pharmacy store run by a brand new, but very ambitious department store chain. I buy some strips of Crocin, Calmpose and Nice, and a box of Vicks and Amrutanjan (A balm for headache). I ask for the bill, and the attendant (who works with 2 computers, wow!), checks his system, & then after some confusing minutes reports apolegetically that he cannot provide a bill for Vicks & Amrutanjan because his system does not have those entries yet.

Now, this store had certainly bought it’s supplies of Vicks & Amrutanjan, and what this means is that these entries are not yet part of their billing system.

Now, I go and ask some executive in this organization about this, and I am sure they would tell me that, they we expect the software systems to handle this. We are not sure what the problem is, we’ll call the network or the system guys, they might fix this in some time.

I have heard similar horror stories in some of the banks in India, which are newly getting computerized. The bank execs are sometimes clueless about their own systems.

With massive computerization & automation, the business volumes have increased tremendously. Organizations use sophisticated systems to run their operations. But the issue is that all these systems are all very Geeky, sometimes requiring a high level of technical expertise.

So, the system executes the Business operations, and the executives run the Business.They will know that if the company has a new supplier, then the billing system MUST be updated with products from this supplier.But, how will they know if their software systems are really doing that?

Before, I go ahead, here are some examples of master data. I am not going to go into defining Master Data Management here. There are some good sources of info available on the web for that

Sales & Marketing

1. Customer Information
2. Service Type Information
3. Warranty information
4. Distribution & partner information

Supply Chain

1. Products Information
2. Item codes
3. Supplier Details

Finance

1. Cost centers
2. Department codes
3. Company hierarchies

The problems normally are the following

1. Data is normally maintained in multiple places.This leads to data quality & consistency issues
2. Missing & inconsistent data can lead to operational inefficiencies
3. In the case of financial data, can lead to compliance issues with SOX.
4. In organizations with multiple departments, enforcing a common vocabulary across the organization is itself a challenge.
5. Different Systems MUST use the same definition for Data as defined by the enterprise
6. These data validation rules MUST be visible to the enterprise to allow for an Audit
7. Outgoing& incoming business information MUST be validated to confirm to the agreed upon standard in the organization.
8. Incomplete information MUST be corrected or flagged as incomplete as per business guidelines in force, uniformly across all systems
9. Most important, all these validations MUST be done in a fashion that is visible to the business and not just to geeks.

BRMS - A Way out?

There are well established methodologies to MDM in an organization. Whereever business policies and standard guidelines need to be enforced, using a BRMS in conjunction with the other MDM tools will mean that the organization has far more flexibility & control in managing it’s core data. Specifically

1. Data Validation Rules can be captured in a form that are visible to the Business
2. Allow the business to participate in writing the validation rules, instead of relying on middlemen
3. Cross system data validation rules are easily captured using a BRMS
4. BRMS helps enterprises prevent costly errors.

Take Away

Organizations where data quality is critical must think about how their data quality guidelines are enforced, and how to go about establishing standard vocabularies across their organizations. BRMS provide the right set of tools that enable the business to work along side their IT in achieving these goals to achieve maximum operating efficiencies