Capture Data At Source

Arun Gupta, Group Chief Technology Officer (CTO), Shoppers Stop defines his Master Data Management (MDM) strategy. Talking to Biztech2.com, he elucidates that for a successful MDM project, it’s imperative to capture data at source, i.e., from the first point of entry and not from a system where it’s been copied from the source file.

What is the biggest challenge that enterprises face with respect to MDM projects? How can this be overcome?

A daunting challenge is to integrate multiple and disparate systems into a formal MDM solution. This can be addressed by systemically reducing the number of existing systems and thereby cutting down on the integration points, which will result in a leaner MDM solution.

What factors should the CIO consider before choosing a MDM solution?

Before embarking on a MDM project, review the problem and fix issues at source i.e., at the first point of data entry and look for an in-house solution. If the problem is too complex, then look at off-the-shelf MDM solution.

More from Biztech

Future Group - Reliance Retail Deal approved by CCI RBI ban on cryptocurrencies takes effect; prohibition could force investors to tap the black market

MDM is a discipline that can be enabled by technology. The technology solution alone will not solve the problem over the long-term unless the process of capturing data at source is addressed.

Complexity of the architecture with multiple source points or multiple systems editing the master data creates the need for a vendor solution. CIOs should consider simplifying the integration across multiple systems before evaluating MDM solutions.

Do you believe MDM will see major adoption in the near future?

MDM is becoming a focus area for industries who are in the transformation phase as well as where the complexity of integrating multiple systems is a challenge. Adoption has been slow but it should become mainstream for most enterprises in the next two-three years.

Most companies perceive that the challenge in MDM adoption is the lack of technology. But the real issue is that a process change is required, i.e. a ‘single point of entry’ needs to be created for data and systems should be then integrated.

However, this should not be done the other way round where the master data is pulled after the system integration process.

As the focus shifts from technology to what matters - clean master data, companies will begin to reap benefits from a mix of process and technology.

There are interesting MDM delivery models being introduced in the market. MDM–as-a-service is one of them. However I believe while implementing MDM-as-a-service, integration of various technologies that need to connect to the respective functions will be a challenge.

According to Gartner, by 2015, 15 percent of organisations will have added social media data about customers to the customer master data attributes it manages in its MDM systems. What is your view on the social media integration with MDM and how critical is this aspect of MDM going to be moving forward?

Social media integration is still experimental. Multiple identities and, in many cases, incorrect information in social media sites pose challenges. I believe this requires careful treading lest the insights generated are not aligned to what the reality indeed is.

In the coming years, customer-facing companies will attempt to integrate social media attributes to the master data to improve the understanding of the customer. Models will evolve but a ready framework that works across industries is a long way off.