We humans are social animals. Our lives and survival depend on relationships with others. Though such an intrinsic part of our existence, managing relationships takes work. Building, maintaining, and nurturing relationships depends on deep understanding and openness. Relationships are dynamic; new ones are established and old ones fade away constantly. Things quickly take a turn for the worse when we stop working on our relationships.
Business is the same. Business is an ongoing exchange and engagement that makes relationship management one of its core tenets. Companies that can effectively discover, nurture, and leverage relationships have a tremendous advantage over those that cannot.
Relationship Status: It’s Complicated
In the business world, we work with a complex set of relationships, no different than some of our personal ones. Traditional enterprise systems are developed to manage a single business entity and relationship. CRM systems for customer relations, SCM for supplier relations, HRM for employee relations, ERP for products and accounts. These siloed systems lock up business entities (customers, accounts, products) and make it difficult to establish a network of connections across all systems. Each system owns an object, but without the complete information about that object or its relationships across all systems. As a result, there are misrepresentations and trust issues across systems, and each system starts creating its own versions of customers, products, and accounts to perform its task.
Relationships in business are not one-dimensional. Business entities do not have one-to-one but many-to-many relationships across people, products, and places. The legacy transactional applications don’t provide the complete view of many-to-many relationships spanning all business entities because they are not designed to do so.
This is where graph technology comes in.
Graph technology allows you to describe an unlimited number of disparate entities and associated relationships, connecting people, products, organizations, and places in many-to-many relationships. With a graph, you can easily visualize which sales person manages which account, who are the key influencers in an account, the products they use, and in which location. And once you establish the relationships between all people, products, and places, you can pivot and make any entity the center of the examination and visualize its relationships. For example, with the product as a center, you can traverse all suppliers, customer accounts, and users across locations.
Graphs play an important role in entity resolution (record linkage) as well. They bring additional evidence via relations and help decide if two objects maintained in different data sources are the same or not. Graph clustering algorithms can be used to infer new properties or relations such as grouping family members into a household. Try doing that with your CRM system.
Put Spark into Your Relationship
Understanding the complex relationships across many-to-many business entities at a big data scale needs sophisticated analytics technology like Apache Spark. You can utilize advanced algorithms such as triangles, page ranks, and node connectivity to learn more about your relationships. Spark algorithms help you calculate scores, relationship strengths, and the influence of entities relevant to your business. Data scientists are using these techniques for ranking entities – for example, finding the most influential doctors by analyzing their publications, speaking engagements, citations in research papers, and participation in clinical trials. Once this information is compiled in the form of a graph, a PageRank algorithm is used to calculate the numerical value of the influence.
Spark provides SQL, machine learning and graph libraries to build custom algorithms. If you need to figure top-selling products within each of your target segments, you can use Spark SQL. If you want to group your customers based on their social connections, locations, and purchased products, you can use Spark GraphX library, and you can use Spark machine learning for predictive analytics.
When combined with predictive analytics and machine learning, relationship management becomes agile. You can discover new relationships quickly using intelligent recommendations and find influencers in the network to strengthen the bonds with customers, suppliers, and employees.
Spark helps you understand the value of your relationships and influencers in your network, and provides guidance on relationships that need work and ones to drop. If only it were this simple in our personal lives.
Trust is the Foundation of All Relationships
Trust is essential for any relationship to survive and thrive. To establish meaningful relationships across your data entities and to glean relevant insights from relationships, you must have trustworthy data. Before you jump into Spark big data analytics, make sure that you have your data in order. A reliable data foundation is a must. Make sure that you have a modern data management platform that brings data together from multiple internal and external sources and helps you match, merge, clean, and relate information. Put necessary processes and governance in place to ensure data reliability and integrity. With a reliable data foundation in place, the graph relationships will be real and Spark insights will be meaningful.
Relish Your Relationships
Advancements in graph technology and access to a powerful Spark environment have offered us an opportunity to leverage data, our strategic asset, like never before. We can uncover relationships across business entities that were not visible with legacy transactional applications. Big data analytics and creative visualization technologies have armed us with tools to help us discover, manage, and nurture relationships, and to make well-informed, data-driven decisions.
Ajay Khanna is the Vice President, Product Marketing, at Reltio and an enterprise applications expert. Prior to joining Reltio, he held senior positions at Veeva Systems, Oracle, and other software companies including KANA, Progress, and Amdocs. Ajay holds an MBA in marketing and finance from Santa Clara University.
Subscribe to Data Informed for the latest information and news on big data and analytics for the enterprise, plus get instant access to more than 20 eBooks.