Audience Profiling – Example & Learnings
If you work in marketing or sales there is a good chance that you are working with lists: lists of clients, lists of prospects, lists of followers, list of key opinion leaders, list of influencers, list of competitors …
All these lists are audiences that marketing and sales need to understand in order to:
- segment markets & identify personas
- identify content that resonates
- spot trends & influencers
- structure campaigns
- assign territories
The good news is that thanks to social intelligence and eCairn Audience Profiling service, it is now possible to do this without spending an arm and a leg. Instead of a long description of what we mean by audience profiling, we are sharing below an audience profiling performed on the followers of the @slackapi account on Twitter.
Similar profiling can be done with lists of emails, LinkedIn profiles. If you’re interested, just fill the form at the end of this article.
Starting with these 14.6K followers, we automatically remove organizations, media, bots & accounts that are no longer active.
Then, we enrich the list of valid personal accounts with data points including LinkedIn Profiles, bios, pictures, content, and “links shared” and we performed analytics and text mining on the resulting data.
Here is the outcome:
On this map, the circle on a city is green, yellow, blue, and white depending on the presence of top, high ann medium influencers in that city.
2. Professional Profiles
Looking at what companies these people work for we see, no big surprise- that a huge number of slack employees follow their company account. We then find people working for large US tech then, freelancers and entrepreneurs.
This is actually quite interesting for sales. Zooming on the Microsoft people, here are a few people following slackapi:
Since we provide the Linkedin url for these people, it is easy to send an intro message/connection request, mentioning “since you’re following our Twitter account” …
Looking at what people put in their bio, we see that @slackapi followers are clearly software and tech people. The bios are 90% filled with tech keywords with only a few keywords like “husband”, “dad” then “travel” and “music” showing up.
Interesting to see that “formerly/ previously” show up a lot. Maybe a sign that engineers keep connections with the companies they used to work for. That is something that could be used in a referral program.
There are many positive keywords too (love, enthusiast) and people writing for this audience should use a positive tone.
Mining job titles, we find a lot of executive titles (CEO, CTO) and developers. However, titles span the full ranges of seniority.
3. Centers of Interest
Here are the top keywords that these people use (removing politics and coronavirus), clearly these people talk a lot about Tech. “Share your idea and build amazing apps” is a tagline that would resonate with the audience.
The text cloud below is the same, but with a focus on Tech. This audience uses hashtags a lot and #ai, #blockchain and #IoT are clear top trends.
These are the articles shared the most by this audience. A mix of tech and political articles.
In terms of domains, the audience is sharing mainly articles from medium, nytimes and github.
When we look at ho these people follow the most. Politicians, Tech gurus and VCs.
4. Network, Influencers & Personas
Looking at how the people within this list interact on Twitter, we calculate the influence of people and end up with this list of top influencers. The number is the influence and the radar logo is an indication of their reach.
When we plot the map i.e visualize it based on people following or engaging each other, we clearly see a few clusters:
Some sub-groups may be interested to manage as a separated audience:
- API specialists, people who are interested in all APIs and how to combine APIs to build products
- Community practices: More interested in the use case, business case, and the benefits of this API.
- Large corporations: These groups can be used for ABM.
We also deliver Affinity Map based on the similarity of people’s social graph i.e grouping people following similar people.
Here is the Affinity Map for the SlackAPI followers:
When diving into one cluster, we clearly get a better understanding of the people in that group and can define personas.
On the left is a text mining of the people’s bios. On the right a TF/IDF analysis of who these people follow most or less, compared with the people in other groups.