Exclusives

Ramon Chen, Reltio –

We live in a world surrounded by data. Whether we’re conscious of it, we make decisions in our every day lives based on discreet pieces of information. From comparison-shopping, to determining our projected retirement funds, data is the critical fabric that we use to ensure the best possible outcome.

Data is being generated every day, in ever increasing volumes and varieties. As devices get smarter, so does the flood of inputs that is supposed to guide and help make better decisions, faster. Whether it’s getting a product recommendation that “you may like” while shopping on Amazon, to being remarketed and “retargeted” (a marketing term describing how you are shown a banner ad associated with a website you had visited), consumer data is being used to market products and services to us in a highly personalized way.

While some may consider it creepy, others see value when the timing and context are right. Personalized marketing can be a double-edged sword, however. An incorrect spelling of a name, the wrong conclusion of a preference can also lead to the reverse effect. “You don’t really know me at all” leads to the cliché of “you only get one chance to make an impression.”

As we ponder the impact data has on our personal lives, we might note that our own workplace environments and tools have seen their share of dramatic changes over the years. We can now capture and store more data than ever. Easy-to-use visualization software allows us to better understand data, and detect trends.

The term “data-driven” is used freely to describe any decision made with data. For many companies a data-driven strategy provides a foundation for strategic growth initiatives. Having better access to the right data leads to deeper insights which most companies use to drive increasing efficiency and revenue.

For IT professionals the challenge continues to be one of cost effective data management to meet the demands of business teams demanding agility and speed. To provide insights and to enable skills such as a “data scientist” – the hottest job du jour – is to uncover patterns and insights in a sea of numbers and facts, all while ensuring governance and compliant access to information.

Meanwhile, business teams continue to look to move away from their dependency on IT. They want democratized access to data. The popularity of consumer-facing data-driven applications such as Facebook and LinkedIn fuel their desire for self-service and independence. With so much technology available and billions being poured into backend IT management, and SaaS or cloud-based business visualization tools, how much better off are teams today?

Is it just your point of view?

The IT and applications landscape is littered with acronyms used to describe a new set of capabilities to solve the problems of a target group within an organization. SFA (Sales Force Automation), then CRM (Customer Relationship Management) for sales teams; Marketing Automation for marketing, ERP and finance, HR, the list continues.

Each application comes with its own database or store to capture, manage and maintain data to help automate and provide the information needed to achieve the objectives of each team and group within a company. Each one of these applications creates silos of data, with duplicate, uncorrelated and often-conflicting facts about customers, products, organizations and more.

Ten years ago, the need to bring these facts together itself spawned a technology, process and discipline called master data management (MDM). The goal was to create a 360-degree point-of-view by identifying and matching profiles across these siloed systems. Unfortunately, the complexity and cost of MDM made it affordable only to the largest companies. MDM’s fatal flaw was also that the identification and reconciliation of profile data (i.e. name, address, phone number) ignored two valuable facts that were needed to derive an actual 360-degree view.

Transactions or interactions related to profiles continued to remain siloed within applications. Without such detail that could include products purchased, social interactions, and other behavioral characteristics, MDM was largely unsuccessful and business teams saw limited value from the billions of dollars invested.

True collaboration remains elusive

With data continuing to be siloed, uncovering relationships across all the entities or actors that make up the important business ecosystem for a company has become a top priority for executives. Billions more has been invested into Big Data infrastructure (Hadoop anyone?) and data lakes have been popular projects for over 5 years.

The ugly truth about pouring data from multiple sources into a lake and allowing it to be analyzed by data scientists for macro-level insights through tools that offer predictive analytics, and machine learning, is that those insights don’t help the “feet on the street”. Those sales, marketing and other business teams and individuals, continue to rely on siloed data and applications.

In contrast, while consumer-facing applications like LinkedIn guide and provide customized and contextual insight to help in personal decision-making, workplace applications continue to languish. Note: Microsoft’s recent acquisition of LinkedIn ironically highlights that such issues are on the radar of major technology providers.

Data, specifically master profile data, by its very nature is born, and continues to evolve throughout its life cycle. Names change, new addresses are added and so forth. While MDM seemingly helps the problem of blending and matching profile data across systems, it does not offer the scale and real-time access, with teams and individuals unable to contribute to collaborate effectively.

Closing the loop is the missing link

Due to siloed applications, insights are analyzed and processed in new data lakes, in traditional data warehouses. This demarcation between operational apps and analytical insight continues to exist despite advances in technology. Macro-level insights rarely distill into the desired operational execution by business teams, and their actions are never accurately correlated to the origination of insights.

A closed loop of relevant insight leading to recommended actions, and such actions resulting in actual outcomes is the only way in which demonstrable, measurable value can be proven. It’s also the basis for which machines can really “learn” and produce better outcomes.

How to drive your data without really trying

If you’re an IT professional, MDM should clearly top your priority list, for reliable data is the foundation. You also have to invest in big data infrastructure, to tie an expanded suite of sources, relationships and transactions to deliver a 360-degree view–all while using predictive analytics and machine learning to automate and gain the insights your company needs.

As business users in sales, marketing or even compliance, you need a way of getting your job done faster, with greater efficiency. You covet the ease-of-use of consumer apps such as Amazon, Facebook and LinkedIn, and realize that collaboration is key to account-based strategies and initiatives.

The good news is that today modern data management platforms have built-in MDM. Like oxygen and water, reliable data is non-negotiable because poor quality data can drive you in the wrong direction. These platforms are architected with big data scale in mind. Analytics and machine learning are seamlessly integrated, distilling insights and recommended actions down to those relevant to individual users.

They also come with a new generation of data-driven applications as part of each deployment, which act on a single unified pool of data for realtime collaboration. With actions audited, tracked and correlated back to originating insight, loops are closed; there is continuous measurement and improvement.

No longer do applications have to silo data and deliver functionality that only meets the needs of a specific group. Today everyone can benefit from a full 360-degree view of everything across the enterprise. That is the true promise of being data-driven.

—

Click here to translate this article
Click here to download the complete .PDF version of this article
Click here to download the entire Spring 2016 M&E Journal