Coverage of this session by Peter Wiley of SocialMedia.org.
3:28 — Lutz: Hopefully my presentation will change the way you see influencers and what you need to get content trending.
3:29 — Lutz: The promise of social media is the “boom” of a single post, share, tweet. It is difficult to attain that boom and a lot of pressure to get the effect. In order to work with data, we need a certain framework. The framework we came up with was to ask the right questions, measure the right data, and to take actions and learn from them.
3:30 — Lutz: We only use one framework? We need to adapt the framework to different departments. Can be used in sales, marketing, customer care. All of these instances will need different metrics, channels.
3:31 — Lutz: I will be talking about marketing and “the influencer.” About the influencer: they have impact on the distribution, “reach,” of content.
3:33 — Lutz: The longer the news travels, in social media, the more effective it becomes.
3:34 — Lutz: Reach does not equal influence. Reach is awareness, influence is getting people to do stuff.
3:35 — Lutz: Non-human traffic on the web is about 50-60 percent robot.
3:36 — Lutz: Analyzing all of the traffic on the web, many companies do not play by the rules and use bots to boost traffic. 50% of what we believe is influence is actually “homophily.”
3:37 — Lutz: Homophily: people have a pre-determined idea and are just sharing this, they might not actually be influencing other people. Influence is dependent on content. What does it take to create a trend?
3:39 — Lutz: Trending depends on how much other news is out there, the content, the sentiment, the emotionality (anger is the best thing to use in order to get something trending).
3:42 — Lutz: There are no truths in this game, only algorithms. If you try and game this algorithm, they’ll change it and it won’t work anymore. You have to cut down texts into different pieces and figure out what is trending and what is not trending.
3:43 — Lutz: [reach] x [frequency] x [content] = engagement. Engagement is probably the best metric to have in order to measure influence. However, engagement is not all equal.
Q & A
Q: How do you apply all this to LinkedIn?
A: I apply my basic knowledge of how to work with data. As a data scientist, you take a lot of data and make it small.
Q: Have you done any work trying to figure out what is the best kind of engagement to measure?
A: Yes! So you have re-shares, likes, and commentary. Commentary has the least engagement, but it’s the best indicator that somebody read the content and understood it. The problem is that you can’t compare it. But it is the best to use.