Master data management: Creating a better customer experience


Do you know who your customers are?

Customer satisfaction is key to business success. If your organization wants to get ahead in a challenging market and/or difficult economy, you could do much worse than understand who your customers are, what they want and how best to deliver that — or a close approximation of that.

Unfortunately, for many organizations, even that first step — understanding who customers are — is much harder than it ought to be. See if the following scenarios seem familiar to you:

  • You receive a flyer in the mail inviting you to join a DVD rental service. You, however, already subscribe to that service.
  • You turn up for an appointment at a healthcare institution. There you are required to fill out many forms, even though you filled out the same forms the last time you were there.
  • You order a product over the Web. The company’s site asks you each time which of several shipping addresses you’d like to use, and the default answer is not the correct one or even the most recent one you entered.

What these organizations have in common is a failure to reconcile customer data in an accurate and timely manner. It’s a common problem, and the consequences go far beyond temporarily annoying customers (as doubtful a business strategy as that might be). Among other potential outcomes are diminished brand strength, lost revenue and needless expenditures in the form of printing, mailing and staff overhead.

In certain industries, even these consequences are trivialized by still more daunting possibilities. In healthcare, for instance, the failure to identify patients accurately in an emergency situation could result in a failure to render the proper medications, treatments or procedures required by a patient’s condition. Even something as simple as an allergic reaction to a normally correct medication could lead to a truly catastrophic result.

A puzzle with no easy solutions

The root cause of the problem is difficult to address. Typically, an organization develops many databases for many purposes; each organizational silo may have its own version of a customer’s information because the customer has interacted with the company in different ways, at different times, for different reasons. A hospital, for instance, may have patients’ billing records, which are completely separate from medical records.

And this problem of record duplication, in turn, makes improving customer satisfaction much harder than it should be. If you don’t know who your customers are, how can you align the services you provide with the services they want?

Fortunately, this issue is common enough that the IT industry has addressed it via master data management (MDM) solutions. These try to reconcile customer records across multiple databases, even fragmented databases, in order to provide a single version of “truth” — information about the customer, which can then be trusted and acted upon by business processes in order to yield a better outcome both for the organization and the customer.

Not all MDM solutions are created equal, though — a necessary reflection of the complex problem involved. Imagine that you have multiple databases, each of which should have at least one record representing a given customer… but possibly each database will have more than one record, and possibly none.

Now consider how many stumbling blocks exist for the software that attempts to sift through the information and arrive at an accurate reconciliation. The different records may not even represent the same individual at all, and even if they do, there may be varying field data concerning critical information such as street addresses, phone numbers, order history and other factors.

The software may not recognize that two or more records represent the same person or, arguably worse, it may create a false association that leaves the organization in an awkward position.

A logistics company, for instance, may find it is better not to deliver a package at all than deliver it to the wrong location. For healthcare, or law enforcement, still worse outcomes may apply.

De-duplication allows business to develop customer centric approach to interacting with their customers by making certain that all information about a customer is identified and collected together!

IBM InfoSphere Master Data Management can create a single, trusted view of your data

Those seeking the best available solutions to this problem should consider IBM’s MDM portfolio. Our offerings are representative of IBM’s commitment to smarter computing in an almost literal sense: their algorithms are unusually sophisticated, and the mathematics behind them is unusually rigorous. But they’re also smarter solutions in a business sense: they can be integrated more easily and completely into an organization’s infrastructure, they can be applied in more ways and the benefits they generate are more extensive as a result.

One clear example of such a solution: IBM InfoSphere Master Data Management. It’s built around a matching engine that identifies not just likely matches of records across and within databases, but also estimates the probability of the match in order to de-duplicate as many records as possible, as automatically as possible — even given a large set of exceptionally large databases.

Furthermore, its capabilities can be centralized and extended — via service-oriented architecture (SOA) — in as many ways, to as many applications and services, as the organization requires. And it includes collaborative tasks that can be used to create tailored workflows that closely match an organization’s needs, as well as pre-built, extensible data models that are specifically optimized for master data management.

A final note: it’s worth considering that IBM is not just a leader in the MDM space — we are in fact the world leader, as measured by both market share and revenues (which, as of last year, were twice those of the nearest runner up).

And for independent confirmation of IBM’s MDM strengths, there’s the fact that IBM was given the prestigious Magic Quadrant award by Gartner Group. Gartner took note of our broad information management strategy, observed that our solutions are strong in a wide range of industries and geographies and singled us out, beyond any competitor, for our comprehensive ability to execute.

Scott Schumacher is an IBM Distinguished Engineer and Infosphere MDM Chief Scientist. He is a technology expert specializing in statistical matching algorithms for healthcare, enterprise, and public sector solutions. For more than 20 years Dr. Schumacher has been heavily involved in research, development, testing, and implementation of complex data analysis solutions, including work commissioned by the Department of Defense.

Smarter Computing Analyst Paper - HurwitzTo effectively compete in today’s changing world, it is essential that companies leverage innovative technology to differentiate from competitors. Learn how you can do that and more in the Smarter Computing Analyst Paper from Hurwitz and Associates.

Subscribe to the Smarter Computing Blog

Recent Posts

Goodbye IBM Smarter Computing, hello IBM Systems!

Catherine Nicholson

IBM Smarter Computing is now IBM Systems! Read on to visit our new IBM Systems blog, as this blog is sunsetting and will no longer publish new posts.

Continue reading

VersaStack expands to more applications and workloads

Eric Herzog

Today, IT departments are caught between two conflicting forces: huge demand for new applications and flat IT budgets. VersaStack can efficiently reconcile the conflicting forces of application and workload demand and cost, while enabling more business value and expansion. IT departments will transform from cost burdens into enablers of innovation and business growth.

Continue reading

Leave a Reply

Your email address will not be published. Required fields are marked *

* Copy This Password *

* Type Or Paste Password Here *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>