No-code graph authoring for Linkedin network analysis
Beginners guide to GraphXR for market research
Linkedin is a social networking site with a mass store of connections between working professionals. Primarily thought as a platform for job hunters and recruiters, its graph-like backbone is a great basis for market research. If you were looking to understand how you were connected to cybersecurity experts — for example — or if you wanted to know who makes up sports integrity (which involves law enforcement, sport governing bodies, intelligence, and so on), a high-level view of your Linkedin connections could provide some market insight.
It is rather a difficult task to identify the appropriate persons to contact, especially when you are searching for subject matter experts outside your field. That’s where building a knowledge graph of your existing and potential connections on Linkedin comes in handy. Visualizing how people are related to one another via their job and expertise can provide much needed context that is iterated and expanded upon for market research.
A visualization-first approach to looking at connections
Without technical expertise, it is possible to readily build a custom knowledge graph. In this blog, we will walk through using GraphXR to begin visualizing answers to questions like:
- Who are thought-leaders across a certain industry?
- What companies do they work for?
- Who are their connections?
Start with downloading your data
It can be of interest to explore your own Linkedin connections on GraphXR to start. Get a copy of your LinkedIn connections data by following these quick steps below:
It takes around 10 min to receive this data in a CSV file via email to download. It will provide you the name, profile, position, company, and connection date. While the data is rather limited, you can manually scrape mutual connections under Profile names of your connections. In this example comprised of fake profiles and companies for demonstration purposes, we create a new column for mutual connections, separating profile names by comma as such:
By dragging-and-dropping the connections CSV into GraphXR, you can quickly group people by company with a no-code Extract data transform. You can also link multiple columns in the spreadsheet shown above. In the short video below, we walk through how to link the “Name” and “Mutual Connections” columns by creating a new category, Contacts, and a new relationship, CONNECTED, under the Extract transform. We also check the Split checkbox to parse a list of names and the Key checkbox to deduplicate our data by merging on the key property, Name.
Expanding your knowledge graph on an ad-hoc basis
As you discover more mutual connections, you’ll often need to add nodes or edges to the graph. You can do this on an ad-hoc basis by manually creating a new node and its properties for every new person or company you discover, and creating one or more edges connecting them to existing people / companies in your knowledge graph.
Tracing a path of connection
Now that your network of interest is expressed as a graph, you can use graph algorithms to discover and highlight patterns of connection rapidly and effectively. Suppose you want to know the people who could connect you to a person of interest. You can use the shortest path algorithm to select source and target nodes, and highlight the connections between you and someone else. Applying the shortest path algorithm on GraphXR can be particularly useful when the person of interest is beyond your first or second degree connections. Spotlighting the path tells the story of how someone is connected within the context of the larger network.
Overview
Market research is an essential part of successful business planning and strategy. The process of gathering and analyzing information on a particular market or industry through a knowledge graph is fundamental to understanding the people and organizations that make it up. It contextualizes our relationships with specific professionals in the field, and perhaps helps us find bridges to other fields that we would not otherwise see.
The goal of market research is to help people make informed decisions. With GraphXR, it is possible to target research by unleashing the power of connections through artful visualization. Learn more at kineviz.com/graphxr or contact the Kineviz team today!
