Designing the Data-Driven Enterprise

analyticsData increasingly has become an essential business capability just like the corporate supply chain. In today’s increasingly competitive environment, timely data that isn’t trustworthy or complete has no value. Or as my British friends like to say it, it is rubbish. But trustworthy data that is 3 months old is problematic too because it forces decision makers to rely on historical data extrapolation or even worse gut feel. Neither is good for business.

As I see it, the levers of data advantage are three: the timeliness, the completeness, and the trustworthiness of data. You need all three for your business to succeed in today’s hypercompetitive environment. Without current, complete, and correct information, enterprises are finding themselves like an athlete without the proper period of preparation. And while most firms agree data is strategic, data remains a business challenge. In fact, only 15% think they are actually good at data (Economist Intelligence Unit Study)– meaning 85% see others as being better at managing and using data.

RossJeanne Ross prescribes the nature of the problem as  “a company’s data, one of its most important assets, is patchy, error-prone, and not up to date”. Companies that do not play to win in how they manage their data are leaving their business subject to increasing risk. Over time they could even see their business advantages lessened as they confront data savvy competitors in the marketplace. This author believes that the 85%–if they do not change—could even see their core business franchises go away piece by piece. The time is now to establish data as a core business competency. The time is now to establish repeatable processes for data. The window of opportunity for being data driven is open as well is the business threat from not responding. In the words of Peter Drucker, “It is not necessary to change. Survival is not mandatory.”

Data represents the fuel for business leaders

For the data savvy, they are finding that their data orientation enables them to enter new markets at will and to take advantage of historical market participants and leaders. Think about how Google now is attacking the insurance business and how different their game plan is than the existing players. McKinsey has found that it pays to put data at the center of the business. Given this, I would like to suggest that there are four ways that data orientation fuels business.

Acting as one company

Mergers and Acquisitions are increasingly a way of life for global businesses. But how do you create one company that can truly act together? How can you create for example one sales force? It isn’t possible without one view of the company and its data. Without the ability to rapidly and seamlessly integrate data, you are left with multiple companies just working under the umbrella of one corporate master.

Decision Making

Historically, decision making has been made for the most part from backward facing data with the addition of forward facing projections and gut feel. This no longer works in a world where competitors can know rather than project or guess. What fuels today’s business decision making is pervasive access to timely, trustworthy data and the ability to discover business drivers. Here when you can trust data because you know where it came from and that it is accurate, it is a game changer. This enables you for example to get a complete picture of your customers, partners, employees, products, suppliers, assets, locations, etc.

Run Processes

Data is more and more the life blood of business processes whether they are for back office, front office, or even manufacturing. Data allows you to know when a process is working or when it is breaking down. Increasingly, fixing processes needs to happen in real time so when the process is breaking automated action can be taken. And this is the case whether the business strategy is one of process effectiveness or efficiency.

Control Businesses

Regardless of business type, data gives managers the ability to control their business and to take the actions needed to remediate a failure points from the planning, organizing, or leading phases of management. This is core to business management. Today, this needs to happen at the speed of business.

What stops folks

So what stops organizations from succeeding? There are three things.


  1. The most important is their data is typically not created with intention; it is created from the scraps of application design. This means that the right data for an improvement strategy isn’t often collected.
  2. Data is dispersed, this means that it cannot easily be put it together for let’s say an integrated picture of customer
  3. The wide variety of data that is possible to use today. Historically organizations just connected to internally developed or deployed applications or more recently to cloud applications which were externally developed and managed. This data was typically structured meaning that comes in a regularized format. Today, however, new sources of information are becoming more and more relevant to business decision making. This includes social media interaction data, mobile location data, and even machine or sensor data. Much of this data is not structured the same way as the above traditional sources. For example, sensor data could provide a number like 74. What does this mean? Is it a temperature and what type of temperature is it? This all means that without the ability fix data, data fails to become part of the operation of the business.

How then do we fix our data?

shutterstock_227687962 - CopyTwo things are needed. First is to think of data first before building new data sources. At the same time, we need a Data Centric Architecture that looks at the “information model” as opposed to the application “data model”.  The information model should put together information created-by and used-by business functions, implemented by or exchanged with systems, and stored in databases.

Next, we need to be able to manage data from any place that it comes from. Essential to this needs to be the ability to relate new data to existing data. Think of the potential benefit to a manufacturer of relating the temperature data described above to the desired baking temperature of cookies to yield of cookies to sales to be garnered. Or think about relating social interactions and mobile location data to sales data at a retailer.

Next, you need integrations to be done systematic instead of piece meal. Delivery needs to happen fast so business problems can be recognized and solved more quickly than in the past.  This means users get access to data when and where they need it. Traditionally, there was a wall between the developers and the business users of data. Business users did not necessarily know what they wanted but we would create requirements anyway. Today, we have an opportunity to allow business users and data scientists to experiment with the data and then with this knowledge push off to the developers exactly what they want. This allows business users on their own to combine existing and new unstructured data sources themselves to test hypothesis. Users in this mode will assemble data and put it together from different sources at different times. This is all about solving the business problems that have held the business back from achieving agility in responding to change. Specifically, it is about providing data by design and not as an afterthought of application design. It is about enabling better management of data that is dispersed across many systems and many formats. It treats integration systematically instead of in piece meal fashion. And lastly, it enables data discovery before investing in data.

Parting thoughts

As we have discussed, data is now an essential business capability. It enables businesses to run better in four ways. But most businesses are held back on delivering the potential of their data. What is needed is to fix things are four things. Together these can provide the business capabilities to re-establish or establish an enterprises right to win.

Further Reading

Solution Page: The Informatica Intelligent Data Platform

Blogs and Articles

Twitter: @MylesSuer