Audience analysis is defined as the process of gathering and examining data about a certain audience (audience insights) in order to best craft the communication.
It used to be part of the tasks of technical writers for theatre, while it’s today extended to all the products and projects that have an online voice and presence.
It is, as we argue here, the most crucial part of a digital strategy, and the reason 95% of marketers we met used to struggle in digital marketing performance: delivering the right communication is, in fact, the most determining factor for success, and the way to go to improve marketing and ad spend.
Two are the main problems of today’s marketing and advertising facing online audiences:
These two problems kept us awake at night for several years while working on online marketing strategies with different clients and industries. Nobody is safe when it comes to crafting messages that are going to generate — almost instantly — feedbacks and ROI data.
CEOs are there to check straight away if this or that campaign is performing and whether people are getting it right.
I remember well the long meetings and complex evaluation that involved the best minds of the company deciding what was the right message to push out, in the hope it would resonate with the audience.
We felt the pressure from the board team, while we felt overwhelmed by the complexity of data analysis and data silos.
My mission is to help marketers and advertising agencies overcome these feelings while taking brave decisions for the brand.
What 95% of marketers and advertising agencies were missing out on was a solid and thorough understanding of the audience, or better said, the multiple audiences, they wanted to reach. Even the most tech-savvy CMOs would focus on the accessible data (owned media channels analytics) and guess the rest, missing out on the whole big data/social data scenario.
Let’s see what a thorough audience analysis looks like, and what all the things are to know in order to have a marketing message stick with the audiences.
(If you wish to get the article in your inbox, get it here.)
Wrapping up and Instagram Board
Once upon a time, the audience was constituted by viewers of theatre, television and cinema, by radio listeners, by readership.
The audience analysis in the classic definition of the audience involved identifying the audience and adapting a speech to their interests, level of understanding, attitudes, and beliefs.
Up to now, most of the literature on audience analysis refers to this.
Even, ironically, if you google it.
What changed massively the concept of audience, evidencing the need for a redefinition of audience analysis, is the participation of audience today and the structure of communication itself, caused by the advent of network communication (internet, social media, and mobile technologies).
The vertical paradigm of the “one-to-many” message turned into a “many-to-many” horizontal flow.
Audiences are now proactive and independent rather than docile and accepting.
Audience participation in the narrative of the official communication gives marketers and storytellers both a great challenge and a huge opportunity that aligns with the prosumers’ revolution for brands (we talked about prosumers here)
These are the reasons why audience analysis must find new, proofed ways to solve the old but still good problem of crafting a message that would stick.
The 95% of marketers and advertising agencies’ problem was not that they didn’t understand this. The problem was that they weren’t sure about where to find and analyse the multitude of audiences they were trying to reach. They would limit, in most cases, the audience analysis to the owned analytics, a.k.a, the brand website analytics and social media channels analytics, missing, in fact, all the places where their customers (especially future customers) hang out.
Which is missing out on the biggest portion of the cake.
That’s why I set myself to write down a comprehensive list of audiences that we analysed in the past for brands and projects, and I hope with this to make you feel less overwhelmed while exploring new territories for data analysis.
But before we get to look at the audiences, which types of analysis did we use to get audience insights?
Well, that’s a good question.
In short, a mix of quant and qual. You want to be able to gather the vastest quantity of data available concerning your business/market/industry, make them manageable through segmentation and clustering, and then inquire about specific business problems.
This could mean using automation and machine learning to access large datasets and find patterns in them, spending hours and hours in the practice of netnography and content analysis and, when needed, dig deeper into some samples using interviews and surveys.
Generally speaking, the aim of this kind of research is
This type of analysis can serve many business goals and produces actionable insights (we talked about actionable consumer insights here), but perhaps it’s easier to have a look at the most common problems we faced to understand its usage in marketing.
Most of the audience analysis we’ve run in the past started with one of the following requests, from an executive marketing or a CEO of a customer-centric business:
So, here’s the list of audiences: all the places where you can run a solid audience analysis.
Understanding the why of your product and how the marketing messages resonate with its consumers is easier if you start asking your best customers.
While we shared a sample of questionnaire for your best customers in this article, I want to stress here the importance of learning about your products and message from the very people that consume them.
Simply said, your product exists because of them.
Asking them about their journey from being aware of the problem you’re trying to solve, to what could be a valid alternative to your product, means drawing down a map for success.
Most of the clients we worked with were already doing surveys and focus groups with samples of their customers. What to add on top of that?
A good understanding of the ecosystem of your brand. Meaning, instead of just focusing on demographic data about your customers and insights about your product, going out there in the real world of your customers by asking questions like:
These kinds of questions pave the path for your marketing strategy success, targeting your existing customers.
Respectively, they give the marketing team tangible directions on:
1 to 8 — SEO, content marketing (content marketing serves the narrative of all social media and digital channels), targeted advertising, marketing campaigns
9 — influencer marketing, PR, media placement, sponsorships
All of them — help you take better and safer decisions about brand positioning in the market and brand voice across channels, from a consumer’s point of view.
There are of course many other questions, and every business case is different, but this is a start in the right direction.
Audience insights pulled out of the best customers’ interviews can serve your further exploration as well. It gives you very precise directions on where to look for further insights concerning your existing customers, but you should always keep in mind the multitude of audiences out there. Don’t stop at this sample, search further.
Web analytics (such as Google Analytics or Google Analytics 360) gives you information about the visitors to your website, which constitute a precious segment of your audiences.
The audience that you can find in your website analytics is mostly aware of your brand and active in establishing a form of relationship (from a first approach out of curiosity and need of knowledge, to shopping in your website if you sell online). These people are generally customers or close to becoming customers of yours.
Web analytics gives you rich insights if you are trying to understand user behaviour that generates conversions, or if you want to quantify and proof specific marketing efforts.
Among the website analytics metrics, the most relevant for your audience analysis are:
More qualitative techniques such as session-recording tools (to see videos of visitors’ screens) and using live chat (to let your visitors tell you what’s wrong with your pages) can be implemented and raise the quality of the audience analysis. You can find tools and specific usage of those in this article by the Conversion Rate Experts.
Social media analytics is the analytics gathering and reporting on data related to social media accounts.
You can find them in all the inbuilt social media analytics such as Facebook Insights, Instagram Insights, Twitter Analytics, YouTube Analytics, plus in the cross-platform social media analytics software, like Hootsuite, Sprout Social, Falcon, Talkwalker…
Mid-size to large size companies tend to use the latest, which I must say made many improvements in the last few years in terms of usability. Cross-platform social media analytics software improved dashboards UX and closing the cycle of engagement with the users online.
So, if what you’re looking for in terms of audience insights is understanding how many people follow your brand and engage with it in your official channels, both inbuilt social media analytics and cross-platform social media analytics tools are great. In the cross-platform social media analytics, it’s easier to cross-check across social channels and interact with the users.
Having said that, the data they give to you, and audience insights you can generate from it, are still partial, mainly for two reasons:
In the past, we had the chance to try most of the software in the market, worked with some of them, and we got so frustrated by the limit of the data that we built our own tools.
So, no, if you asked me if social media analytics software are good tools to run audience analysis, I would say that they are just good for monitoring your own channels. They don’t add much to the inbuilt analytics’ features, they just put all your channels analytics in one platform, easy to check.
The most valuable insights you can get there in terms of audience analysis refer to engagement, and specifically:
Which content, out of your content marketing efforts, is performing best?
A good analysis of the top posts you’ve published in the last six months to one year can say a lot about your audience attitude towards the topics and narrative adopted so far by your brand.
Advice: make sure the values make sense. The question you have to make yourself (and to your suppliers) is:
How does [social media analytics tool name] measure engagement?
A like, a comment and a share of a post don’t have the same impact in your marketing ROI, so make sure that the value the analytics assigned to each interaction feels right.
Just to give you a hint, when we designed our analytical system, we assigned a value percentage to shares/repost/RT higher than likes/hearts and comments, and added content analysis to understand whether comments were positive or negative.
Some of the social media analytics software in the market, and many of the social media monitoring software, claim to offer another service that is essential to audience analysis and to gathering actionable audience insights: social media listening.
Social media listening is the action of tracking all social media channels where your brand is, and gathering from there the actionable insights for your marketing strategy.
At the time I’m writing this article, social media analytics and monitoring software can offer a good understanding of:
Though, they are not able yet to give you:
In fact, when they say they take data from different sources in the social media and web scenario, they actually mean Twitter.
The solution for this is to create a custom dataset across social media channels (including unofficial channels that influence your audience, for instance) and gather insights with the help of your data experts/researchers.
We’ll get further into this in the online conversations and user-generated content chapter.
Out of all the inbuilt social media analytics, Facebook audience insights stand out for quality and quantity of data and so I think it might deserve a special mention.
It’s no secret that Facebook has a huge amount of data about its users.
But how can we use Facebook audience insights for audience analysis?
The two best actions you can take — at the time I’m writing this article — are:
Keyword research is a quant qual analysis on what your audience is searching on their browsers, that somehow has a connection with your brand.
Now, here is when you abandon your own channels and go hunting for your audience out there. You are already in the Hall of fame of the Bravest Marketers and Adlanders, and yes, entering the 5% mentioned in the title.
If the audience analysed in your website analytics and social media analytics is constituted by people that know you and often buy your product, here in the keyword research the audiences might not even have heard about your brand, but in what they search online you can turn up and make their life better, making them your new customers.
Thorough and specific Keyword research can make you understand what it is that people look for
These kind of audience insights are the base of every successful long term marketing strategy. Why?
Because with these insights you can
Keeping in mind that audiences and potential audiences are located at different stages of the level of awareness about your brand (check the Consumer Insights Roadmap if you haven’t yet or download it here), you can generate solid growth in unaware and problem/solution aware audiences.
Who are they? People that don’t know they want you, yet. That’s the bad news.
Good news is, though, your brand identity covers semantic areas (areas of meaning) that the unaware audiences love already. So you have a connection, even if they don’t know it yet.
Your connection is what we call topic affinity, which means you can become an authoritative, influential resource in their favourite topics and start a relationship with them.
It gets even easier with the problem/solution aware audiences that are looking up into solving some problems you’re already solving. The connection here is stronger: it contains not only a topic affinity but also a problem intent and sometimes a shopping intent.
The intent should always be a driving factor of your keyword research, even though it’s not the type of data you can get out of tools like Google Adwords, Ahrefs or other keyword research tools — those tools though give you the other fundamental data for audience analysis: volume of searches, competition, correlated keywords.
Intent, instead, requests qualitative analysis, but believe me, it’s worth it.
Intent is what makes you understand not only what the audiences are searching for, but also why, so that you can refine your list with only the keywords that matter for your brand and marketing strategy.
You just learnt how you can use keyword research to uncover new target audiences while staying true to your brand identity. Most companies leave the keyword research to SEO experts and techniques, while you can now point out marketing opportunities generated by keyword research.
For the explorers that came all the way here, in the unknown territories of “what people want to talk about”, the step to “what people are talking about” is not that big. It’s actually pretty smooth, as it makes sense.
You left the other 95% of marketing executives in their own land, wondering only “how are people reacting to my message?” and forever trying new messages until they find the right ones.
This is the open land of online conversations and user-generated content alias “what people are talking about”.
It is studied by the so-called “netnography”, ethnography on the internet, a qualitative research method devised specifically to investigate the consumer behaviour of cultures and communities present on the Internet.
Why is it extremely important for marketing?
We’ve already mentioned how the audiences’ role has changed in the last few years, how the vertical paradigm of “one-to-many” message turned into a “many-to-many” horizontal flow, and how its proactive participation is both a challenge and an opportunity for the marketing world.
If the challenge sits into the mostly overwhelming crowd of everywhere and anywhere communications and billion of messages, and that’s clear to most of you, where is the opportunity?
The opportunity is that the audience is telling us exactly what they want (to talk about, do, buy…). They also tell us how they want it and why they want it.
Moreover, research says there’s evidence that user-generated content increases purchase behaviours.
But how to start, giving the enormous amount of conversations and user-generated content online? What is good and what isn’t out of the overwhelming crowd of everywhere and anywhere communications and billion of messages?
You might feel like an explorer when he faces the unknown. Lost.
The answer is: you follow the directions taken in the keyword research which is, with the customer journey, your compass in times of exploration.
The semantic areas, topics, and keywords identified as characterizing your audiences (“what people want to talk about”) can now be used to map conversations and user-generated content that can resonate with them.
This progressive process of understanding is powered by the audiences, for the audiences, and that’s why it works.
The following chapters define all the conversational territories, alias where the audiences are hanging out to talk about the semantic areas, topics, and keywords previously identified.
Being the explorer in unknown, rich territories, this is when you find regions, cities and towns and get to meet their tribes.
According to the Dictionary of Media and Communication Studies, a good deal of communication takes place within groups of one type or another.
Now that you’ve learned what topics resonate with your audiences, you can find online groups that are born to talk about those topics.
The members constitute informed and passionate audiences of such topics, and the closer the topics are to your brand identity, the more targeted the topics groups are for your brand.
They certainly contain high-quality user-generated content and powerful influencers for your marketing strategy.
An interactions-rich form of groups in the digital world is the one of the online communities and forums.
They are a sort of subset of the topic groups. They are powered by the interest on a common topic, like the topic groups, but they take to an extent the mutual support.
The members find in the online communities and forums help in their everyday life and share their experiences related to the common topic and their journey into it/with it.
The kind of audience insights you can generate from applying audience analysis to online communities and forums is the same mentioned before for groups in general, and it applies also to Facebook and Linkedin groups: Community insights, Behavioural insights, Content insights, Brand insights, Market insights, Influence insights, Network insights.
Facebook groups are now what communities and forums used to be, back in the time of web 1.0: the biggest pools of like-minded people one click away from us.
This is because Facebook brings in its huge adoption by people that are not necessarily web or tech savvy, so it opens up the gatherings to the masses.
At the time I’m writing this, Facebook groups count 1.4 billion people — more than half of Facebook’s user base. Of those, 200 million people belong to so-called “meaningful Groups,” considered a vital part of users’ daily lives.
so they work well in the granular yet massive amount of audiences on Facebook.
In one of the most fun times of my career, I visioned around 6 Facebook groups of people materializing in person to see an exhibition of them: Facebook’s ‘Gallery of Groups’.
There were the sausage dog owners and the elder skateboarders, the panini sticker lovers and the Rock Art purveyors. It was just amazing to see them meeting up with their virtual friends and sharing common stories. You could see, in 3D, how people are connecting and building relationships with others over their shared interests.
Having started audience analysis on Facebook groups in the early days, they were a sort of trigger reason why we’ve started building our tools on the first place: nobody was offering Facebook groups insights and we couldn’t analyse them at scale without a custom technology. Since then, they’ve been an amazing resource for consumer insights and targeting. This is just to show you that analysing Facebook groups would definitely give your brand an advantage in the market, as most of your competitors are ignoring them.
Linkedin groups constitute a rich, broad database as everybody is, professionally speaking, on Linkedin, and people accept in large scale invitations to groups.
At the moment I’m writing this though, interaction is very low and spam makes activity noisy so they are not the ideal place to analyse qualitatively.
But we’re holding onto the hope that Linkedin’s investment in the New Groups Experience will bring a new spark to their contacts’ pools.
A hashtag is a type of metadata tag used on social networks such as Twitter and Instagram, allowing users to apply dynamic, user-generated tagging which makes it possible for others to easily find messages with a specific topic.
Hashtags create streams of user-generated content and conversations available to research, marketing and advertising experts for audience analysis.
They are the ideal means to monitor and analyse:
They are particularly representative of engagement around shared experiences which use Twitter as a backchannel, such as television, online streaming content (Netflix, Amazon etc) and live events.
Audience analysis on hashtag datasets could produce the same kind of insights that we’ve mentioned above for groups, but there is a fundamental difference with groups and communities: people gather around hashtags in an occasional way, they don’t feel they belong to a structured network and they don’t follow ongoing rituals (there are few, rare, exceptions). For this reason, audience analysis on hashtags reveals weaker connections and interactions and users’ interest on a certain topic could also be superficial and flowing. It is recommended to keep in mind the temporary nature of these kinds of groups while analysing them.
Events form part of all our lives and they have been used to signify important aspects of our culture since 60,000 years ago.
From an audience analysis point of view, they are extremely valuable pools of target people, assuming that their topics and protagonists are close to your brand identity.
Their strength is highlighting a segment of people whose connection with the topic and protagonists is so strong to make them take an action that requests undoubted belief in the topic and protagonists: leave their place and attend physically the event in order to meet them.
Even though audience analysis for events requests a specific, in-depth education, here are the basics of how to analyse events’ audiences to obtain audiences insights (we are designing a specific course for this, so drop me a line if you want to dig deeper):
Lookalike Audiences is a method hugely used on Facebook to reach new people who are likely to be interested in your business because they’re similar to your best existing customers.
Now, what 95% of marketers and advertising agencies we’ve met were missing was:
Your competitors’ audiences, or some segments of your competitors’ audiences, are insight-rich pools of people and constitute a great opportunity to understand how to stand out from the competition’s noise.
The most important factor in conducting audience analysis on your competitors’ audiences is understanding what’s working with your competitors’ audiences, and what’s relevant among them.
This gives you a great marketing advantage and a way to tackle a customer-centric competitor analysis.
In the video content world, Youtube gathers a great majority of conversations and user-generated content.
Any movie, series, documentary, or product advertising (what we call “commercials”) has active audiences in Youtube that leave comments and feedback, and therefore you can run an audience analysis.
Also video content online databases such as IMDb are giving a voice to audiences, and just like all review sites, become most authoritative players in the market of the times.
This is because audiences tend to be influenced by the opinions of other viewers more than advertisers and media, and understanding them is the first and most important tool for the success of video content.
Original video content producers and distributors often base their future products on successful formulas, and finding the elements of such formulas is pretty straight forward through audience analysis on Youtube and video content online databases.
In the customer journey, reviews are a determining factor of evaluation and often perception of a brand.
What 95% of brands and marketing professionals usually miss in their strategies is seeing their products as seen by their customers.
The ugly truth is sometimes scary and complex to embrace in a company. But it is, after all, the truth. And marketers/advertising agencies have the responsibility to represent this truth, the customers’ truth.
Reviews represent a segment of customers that are willing to share their journey with others, aware of the fact that they’re going to have an impact, even only for the influence they have on others’ decisions.
Brands that clearly acknowledge this and give the customers a seat at the table have always a huge outcome.
Some crucial insights you can obtain from an audience analysis on reviews are:
Reddit is a highly ranked platform featuring online conversations and user-generated content (ranked #5 most visited site in the US according to Alexa).
It is organized into subreddits: communities and forums where Redditters help each other and share content about a certain topic.
The benefits of using Reddit for audience analysis are in the quality of the content and level of expertise of the communities on the topics. Researchers can sometimes start a broader audience analysis from Reddit users’ indications, or even sample super fans there.
On the other hand, though, the cluster of Redditters might not be representative of every business’ audience, so it really highly depends on the business case.
Some excellent pieces of content have been written about your industry/market/product by those that are embedded in your segment and know how to tell its story.
Audiences that have engaged with this sort of content have a connection with your brand based on topic affinity and it’s worth including blogs and article comments in your audience analysis, if the topic is strongly targeted.
Another practice that we’ve used in the past for audience analysis (aimed in this case to content marketing), was to cluster sharers of targeted articles to propose them similar content.
Some brands use apps to foster relationships with their customers and serve them in the best way.
In those apps, just like we previously mentioned for the websites, users leave tracks of their experience that are precious to your understanding of consumer behaviour.
Connecting the app analytics to your audience analysis enriches your thorough vision on the audience with a crucial segment: some of your most engaged and motivated customers.
Kudos to you for keeping up with this huge list of audiences!
I hope this can inspire you to go beyond what’s known and familiar, and aim higher.
A thorough understanding of the audiences will give you a point of advantage on your competitors (you’re in the top 5% of marketers and advertising agencies now!) and will help you take brave decisions for the brand by ensuring that your message is going to stick.
Please let me know if there’s anything that you want to examine in depth or if I’m missing something here that you would like to include.
In the ever-changing world of audiences, we constantly develop tools and techniques to meet and understand people where they are and take action.
Only a few weeks ago, I started looking into Instagram pages as conversational territories for students, inspired by the great article by Taylor Lorenz in The Atlantic. We now call them Instagram Boards.
This is just to say that in terms of audiences, there are new opportunities every day; possibilities are endless. It highly depends on the brand/project where the target hangs out and how to reach them.
I would love to keep this conversation going: what are your best audience analysis techniques and datasets?
(If you wish to get the article in your inbox, get it here.)
Don't know where to start or looking for clarity in your audience analysis? "Chat with a researcher" is a custom, lightweight engagement designed to give you data-backed actionables for your most pressing audience analysis concerns. For as little as 300 £.