Graphs

If you want to learn why we use the graphs we use I wrote a blog post about it here

Railgun

The Railgun graph is designed to display an anologue view of a group of companies and their interests.

Sentiment

The idea of the Sentiment Graph is to quickly get a grasp of the behaviour and interests of a company from the social media activity of its employees.

Final

This graph will be deprecated at some point in 2022 {.is-warning}

Our final data graph is a visualisation of our last data mining step after filtering posts and employees. All further graphs from this point on are different ways to view the final data graphs such as the Sentiment graph.

Correlation

Shhhhh this is an alpha product, but if you ask us nicely we may be able to incorporate it into a campaign. {.is-warning}

Our Correlation Graph is designed to reduce time exploring links between colleagues of a company. The idea behind this is to help you traverse a company through internal relationships. This is better served once you already have a relationship with someone in this company or for planning a route through people to build a network with them.

This graph shows how much colleagues interact with each other around the target keywords e.g liking the same posts, liking a post they made, etc. So if you're targeting a company for a specific area you know the people that are talking about the same topics and it should be easy to build relationships by leveraging that people know each other and are interested in the same things.

How to read

We have tried to make it as easy as possible to read our graphs but there are some not so obvious aspects of them that are worth pointing out.

Node Proximity

The closer the employee is to a node (cloud for example) the more interest they have in that node. Say the employee liked two posts about cloud and only one about google, the employee would appear to be closer to the cloud node.

Node Line Colour

Sometimes on a large sentiment graph, we will highlight the nodes and their lines will be a different colour. This is just to help show exactly which employees are interested in certain nodes.

Node Size

As you can see in our examples, nodes are of different sizes. Their size is directly proportional to how many incoming links they have, A.K.A. the more interest they have the larger they appear.