Max Tomassi

engaging on social media is hard? we’ll give you a tip!

Engaging audiences on social networks is extremely important for a company with millions of followers/friends, as a broadcasting company is likely to have. Analysing that “social engagement” allows for the extraction of really useful information like the topics people were more interested in, or when during day they were more active. With such a huge audience though, it isn’t easy to have a clear overview of what’s going on. This is exactly when our social network analysis tool, Engage, comes to the rescue!

Engage allows tracking of multiple Twitter accounts and Facebook pages, providing statistics and useful temporal graphs: in a glance you can immediately have an idea of the level of engagement of your audience over a specific period of your choosing (maybe there’s a big event this week and you want to see how your audience is reacting to it?). There’s more, though…

one tip a day keeps the doctor away

We decided to support the engagement level graph with a new featur we called “tips”: we will provide small tips with specific and useful statistical information. The first one we are going to release soon is a tip for each tracked Twitter account and Facebook page saying, for example:

‘Week from 26.08.2013 to 01.09.2013: sent 50 updates per day with peak of 70 on 29.08.2013, received 1200 updates per day with peak of 1700 on 30.08.2013’

For both “account actions” (updates sent by the tracked Twitter account or Facebook page) and “audience actions” (reactions received from the audience) we provide the following metrics:

  • Average of updates sent in the period of interest (e.g. 50 updates per day)
  • Maximum number of updates sent (e.g. peak of 70)
  • Time when the peak happened (e.g. on 29.08.2013)
Given that we are analysing a period of one week, the granularity of the data provided is at day level. The same is true if the period is a month, while for quarterly and yearly periods the granularity will be at month level (e.g. sent 1200 updates per month with peak of 1500 on February 2013).

We think this can be an useful tip for our customers. It makes understanding which specific time was the best in terms of both produced and received updates easier (I have an higher engagement on Tuesdays!) as well as if there’s a correspondence between the “account actions” and the ones made by its audience (was the audience more engaged when I produced more updates?)

some technical details, please

Technically speaking we created two new RESTful services in the backend side of Engage, one dedicated to Twitter and one to Facebook. When invoked, these provide a json document containing the data related to the metrics we’ve seen before, so that the frontend side of Engage can process them asynchronously and compose the tip.

For computing the metrics of the tip we reused (reusability is beautiful) the same external services that already provide the data we show in the engagement level graph, but with a slightly different granularity: as we saw before if we are analysing a week we want a level of granularity of days for the tip instead of hours as the graph needs.

lots more to come, stay tuned!

More “tips” and new interesting features are going to be released to Engage in the near future, so I suggest you keep tuned for our next blog posts! In the meantime let us know if you find this feature useful or if you have one to suggest that you’d love to see in Engage. How? Leave a comment below or tweet us!

blog comments powered by Disqus