How We Define Influence

We recently wrote a blog post detailing the “Top 100 Influencers in DevRel.” This post received a lot of attention in the DevRel community and led to people wondering how exactly we measured “influence”.

eCairn defines an influencer as a person who, through their social media presence, affects the opinions of peers within a specific ecosystem. We specify a specific community for the following reason.

Take someone like Elon Musk (https://twitter.com/elonmusk) . He has 93 million followers on Twitter alone. He is very influential and does talk about Artificial intelligence from time to time. However when it comes to the data science community, his influence is low.

Here is our top 7 influencers in AI/Deep Learning

Influence is contextual and the context is a “group of people who interact and share”. So although Elon Musk is a celebrity with huge reach and some relevance to AI, he has no influence over the data science community.

Someone who works in and is more connected in the data science ecosystem would have a higher authority than Musk. We refer to these people as “micro-influencers.”

Although they are not your traditional celebrity influencer, they have influence in their specific community. That’s why we believe that when running an influencer campaign, it’s vital to first know the community that you target and second, know who actually has influence in a community.

It’s important to look at metrics outside of just follower count. Our metric of “influence score” does exactly this. You may be surprised to find out that the people with the biggest following are not always the ones with the highest influencer score.  Knowing the influence score of the people you are working with in an influencer campaign will lead to a more far reaching campaign.

In the rest of the blog we’ll detail the specific parameters that we use to calculate influence score.

Why Competitors Failed

The influence score that eCairn computes is local to an ecosystem. Some companies in the past have  unsuccessfully tried to create universal/generic influencer scores.

Klout is an example of one of these social media intelligence services that end up failing. They launched in 2008 and were shut down in 2018 after being acquired by Lithium Technologies. One of the main reasons that Klout ended up failing was that their system was easily gamed.

eCairn’s formula for calculating influence takes certain metrics into account that makes it much harder to game the system.  You can buy followers, it’s much harder to convince other people who are influential in your ecosystem to follow you!

Klout’s influence score, called “Klout score”, also ended up becoming something of a competition between people. People were starting to see Klout score as a way to measure their own importance, a vanity metric, rather than something that could benefit a business.

What we offer at eCairn is different in that sense. Another competitor that failed was Kred. Kred had similar issues to Klout, in that their system was being gamed by people so that they could artificially inflate their score. The biggest reason Klout and Kred failed though, is because they tried to determine “absolute influence.”

This means that they tried to calculate influence without any context or nuance. The way that micro influence really works is that people have influence over a specific group. This group is normally related to a common center of interest, although sometimes there are other criteria by which the group is defined.

Our service also differs from Klout and Kred in that our influence score is more tailored to spot micro & nano influencers. We don’t do much in the celebrity business. We define a micro influencer as someone with a reach that is in between the 1 thousands to 10 thousands. This is where influencer marketing really works, especially in B2B or for businesses targeting niches.

What eCairn offers is more specifically tailored to the micro-influencer group you see on the right.

A Closer Look at Conversation™

Below is a screenshot from our Conversation™ platform that gives some insight into some of the metrics that we use.

Here we use eCairn’s founder as an example of a micro influencer. The table you see shows how he is connected to other influencers in the same community (Digital Marketing)on Twitter. The amount of follows, likes, retweets, mentions, and replies are all metrics from Twitter that we use to help calculate a person’s influence. What is key is that we only account the “signals” that are within this ecosystem (in the example: Digital Marketing). This goes a lot deeper than just looking at an influencer’s follower count. An influencer can have a higher influence score than someone with a bigger following because they receive more engagement than that other person. Other existing methods of influence ranking only use follower count which can be artificially increased through the use of bots, fake accounts, purchase of followers and collusion with other users.

Below is a screenshot that shows a map of how people in a certain community are connected to each other. This is a great way of visualizing people’s influence.

Pictured: Community marketing map

We can also see eCairn’s founder in the middle of the map. This allows us to visualize how he is linked to other micro-influencers on Twitter within his community. We use the metrics we mentioned earlier to determine who everyone is connected with. You’ll also notice that the people who have the biggest circles are the ones with the most connections.

These are the people who would be ideal to target in an influencer campaign. Looking at a social graph/map and you can decide to target influencers that belong to different clusters or to focus on a specific cluster:

Using Blogs to Determine Influence

Another metric that we use to calculate a person’s influence score is blogs. Blogs provide the ability to broadcast original, in-depth and insightful content for an influencer’s readers. When compared to the small character count of social networks, we can see how blogging provides a unique avenue for developing insight in the community.

Final Thoughts

Building influence score by starting with a specific community ensures that you find influencers who are relevant. Say for example that you’re trying to find the influencers that are experts on movie production. You don’t want their interactions with the music recording community to affect the influence measurement on the movie developer community.

You’re trying to find the most influential people on a specific domain, not the most influential people who happen to have an interest in that topic. 

In eCairn Conversation ™, as you start by building your community, you control the level of noise that can come from irrelevant sources and can manage the objectivity of the ranking.

Uncategorized

Leave a Reply

Your email address will not be published. Required fields are marked *