What is audience segmentation, and why is it so important?
Audience segmentation is identifying groups of people within a broader audience, which allows us to uncover the insights that make them unique, what they care about and what trends hold them together.
Before we get further into the topic, it is important to understand the difference between segmentation and filtering.
Segmentation is an output that happens when an audience is grouped together by unique characteristics and connections that are data-driven and unbiased.
Whereas, filtering is often an input method used as an initial step to define an audience, or as a step to select an audience post-insights, the results of which will provide a predetermined audience.
At Audiense, our 'unit of work' is the initial audience, defined by specific criteria you choose to input, such as people found in a specific geolocation. Once your audience has been defined, Audiense segmentation allows you to dig deeper into the niche communities that make the audience and provide niche insights on each of these groups, enabling you to action your findings whether it be optimising content type/reach, increasing campaign ROIs, increasing lead generation or uncovering new market opportunities.
How does Audiense segment?
We believe that on Twitter there are two main types of behavior in the follower/engagement relationship, they are not completely exclusive one from another, but there is a clear distinction:
- You follow/engage with accounts that you don’t know personally or are not work-related, hobbies, or peers, but they are interesting to you. Usually, they are going to have a bigger audience than the average person. Extreme examples of this behaviour are celebrities accounts.
- You follow/engage with accounts that you know personally or are your work/hobbies, peers. Extreme examples of this behaviour are following your boss or your university colleagues.
Historically, Audiense segmentation has been based on interconnectivity, which is more connected with the second behaviour. Now we offer two types of segmentation, Interconnectivity, and Affinity, so that we can use the most relevant one depending on your audience or use case.
The Twitter handle of a user represents its personal branding. Follows are indications of their affinities, hobbies, interests, when properly registered/analysed. When segmenting by Affinity, we are running interest-based segmentation because it focuses on buckets of users and their interactions with the top 1-10% of the influential accounts. When segmenting by Interconnectivity, we are able to see how audiences naturally come together at a more granular level, and it's particularly useful in certain use cases like B2B, and overall for non-global-B2C audience studies.
Interconnectivity segmentation
Audiense segments based on interconnectivity by clustering individuals based on “who knows who” i.e how these individuals are interconnected. We take into account who engages with who and cluster them together - for instance, if person A follows person B then they’ll be clustered together. Our data team found that clustering based on this method creates consistent and relevant segments, as well as identifying more unique characteristics for each segment. Once we have identified the clusters/groups, we can ask “how do they know each other?” i.e. what are the hidden trends that the platform has revealed, including interests, psychographics, influencers.
A much more granular look at the audience and the sub-communities that represent it.
Best for:
- Finding very niche segments within your audience. In this context, niche refers to a smaller, specialized section of the audience, perfect for a targeted approach.
- Identifying ‘odd one outs’, potential non-desired segments within your definition, such as bots: they tend to not be interconnected among themselves, and therefore stand outside the segments.
- It works for both B2C and B2B. In B2B, a group of interconnections can appear even if they do not explicitly express their job role. If we were to look at Facebook employees, the segmentation would cluster the audience into departments like engineers/product, and sales/reps, without the members having to have declared that data publicly. We believe that B2B cases can find Interconnectivity segmentation quite valuable in breaking down their markets into beachheads to conquer.
Affinity segmentation (Beta)
Audiense’ new Affinity segmentation clusters your audience into segments based on commonalities of affinity and interests they have expressed.
When segmenting by affinity, Audiense will look into all the accounts that the members of your defined audience are following. With that information, the algorithm will group the audience members who follow similar sets of accounts and therefore have similar interests.
Best for:
- Looking for fewer but more widely defined segments: this segmentation often results in a lower number of segments that provide a larger % of the audience within them, and at a higher-level segmentation of your audience
- When you need your segment sizes to represent the total audience size
- Analysing big audiences in order to understand a broader audience (often linked to a consumer media strategy use case)
Note: Users with a Free plan or Twitter Marketing plan will only be able to select Interconnectivity segmentation. Users with an Audience Insights plan will have both segmentation methods available for use.
How to select Affinity segmentation?
1 - Start creating your report as usual: choose audience type, define audience type.
2 - Click Next, check the box “Affinity Segmentation”
(this segmentation type is now set by default)
3 - Those reports created with Affinity segmentation will be identifiable with the purple Affinity segmentation label:
4 - If you don't want to segment your audiences by Affinity, just select Interconnectivity segmentation.
What else you should know about our Affinity segmentation
- Affinity segmentation reports count towards your monthly report credit. Please contact your Account Manager if you have any questions about running the different types of segmentation. We advise you to look at both types, and learn what is more useful for your particular use case (a more granular look into an audience, especially a local audience will be well-fitted towards Interconnectivity vs a broader targeted campaign may be best suited towards Affinity).
- Bear in mind this feature is in Beta and you might encounter some unexpected behaviours. If so, we kindly ask you to report them via help@audiense.com or product@audiense.com.
- If you would like to provide feedback on your experience, either directly contact your account manager or get in touch via productrequest@audiense.com.
Newly improved Segment names!
Read more about our AI generated Segment name improvements that you can find on our audience breakdown view once you've created a new insights report in this FAQ.