Social relationships hold a wealth of information about your customers, employees, and suppliers that (if you have the tools to harness them) can provide you valuable insights about the factors that influence behaviors that ultimately impact your bottom line. The challenge with analyzing and visualizing social relationships is that the underlying data doesn’t match well with the relational data structures that most data warehouses are designed around. The use cases that your company needs to embrace to create competitive advantage are too complex for most relational databases to handle, and as a result, you may need to look beyond your traditional RDBMS to object databases designed for social analytics.
Why Relational Databases Struggle with Social Data
Relational databases are designed for data that looks like a spreadsheet – rows and columns of data about clearly defined concepts. Relationships are primarily structured between tables, and only occasionally do parent/child relationships get defined between records in the same table. When you do have a parent/child relationship in a relational database, the nature of that relationship is clearly understood and defined in the database structure. Social data doesn’t fit this mold very well. With social data, the relationships are where the meaningful insights lie, not in the data stored in a table. Person x directly influences persons y and z in the context of a post, and groups A, B, C, and E all are influenced indirectly through some network of relationships that exist. Somehow, this influence impacted the behavior of certain individuals, but sorting everything out and visualizing it in a way that makes sense and is actionable using a relational database would be nearly impossible – they just aren’t designed for this type of data.
What Types of Questions do you need to Answer about Social Relationships?
Social relationships have existed since the dawn of humanity and have had a growing impact on the business world ever since. What has changed in the past few years is the existence of a digital representation of these relationships through social media platforms like Facebook, Yelp, Pinterest, Twitter, and LinkedIn (there are many more). Since the data now exists for companies to discover social relationships, leaders and decision-makers are now asking their IT departments to help them answer some fundamental strategic questions?
- If a social message is generated, how far will it reach throughout the network (who will see it)?
- How are social relationships and messaging impacting buying behavior?
- How can we influence (leverage social relationships) the perceptions of the company, products, and services in the eyes of potential customers (aka, get them to buy more)?
- How can we use social relationships to identify and mitigate reputational risks quickly before they have a significant impact?
- How can I use social relationships and feedback data to identify new market opportunities?
These questions all have a couple of things in common – they require an understanding of social relationships, and they involve combining social data with transactional data (like marketing messages, sales records, customer data, etc.). Social data doesn’t exist separate from your business data; it is part of your business data.
Object Databases are Ideally Suited for Social Data
Relational databases struggle with both modeling social relationships and analyzing the complex use cases that executives are increasingly requesting. Object databases, on the other hand, are designed specifically to handle the types of data and relationships involved in social analytics. The insights in an object database are centered around the complex web of relationships that exists between objects and the object data itself is secondary. Does this mean that an object data can’t handle relational data? No, but relational processing is the easy part. Object database are optimized for difficult questions and complex use cases surrounding relationships as a context for processing data content about the objects.
Answering complex business questions requires an understanding of both content and context – bringing them together so leaders can visualize the big picture. This is what object databases are designed to do. Additionally, object databases are architected in a way that enables big data sets and complex transactions to be broken into smaller parts and analyzed using parallel processing. Meaning difficult data problems can be analyzed more quickly, giving leaders the answers they seek fast.
To learn more about how object databases can be used to help your company leverage social relationships to drive better business results, visit https://www.actian.com/data-management/nosql-object-database/