We know Facebook has a huge amount of data on people. For the last 18 months, they've been sharing more of this information than ever before through their platform, Audience Insights. As a result, we can begin to pull together audience personas for very little cost other than time, effort, and a Facebook account.

This post is going give a whirlwind tour of how we can begin to use Audience Insights to build personas for our business that will allow us to target content better and keep people in mind rather users & sessions.

What is a persona and why should we build them?

A persona is the summary of research or observations based on a key group of users who show similar behaviours and lifestyle choices.It allows us to collectively group users into buckets, rather than having to focus on thousands of individual needs and wants.

This is then distilled into a fictitious person that can be referenced to guide business decisions, whether they be the type of design we use, the content of our email marketing, the tone of voice we use for our brand, or even the types of products we may look to be selling.

We may end up generating multiple personas to connect to various users we are looking to share our brand with. This will help guide business decisions, rather than taking a one-size-fits-everyone approach.

Three things to be aware of

1. Facebook offers two audience options — “All of Facebook” or “People connected to your Page.” In the past, if we have paid for Facebook followers or used extremely broad advertising options, the “People connected to your Page” data could potentially be an inaccurate representation of our target audience.
2. If our audience segment is less than 1,000 people, Audience Insights won’t display any information.
3. If you select more than one option in the faceted navigation, this uses an “or” functionality. This can make data hard to dissect, particularly if you input multiple interests.

How to build a persona using Facebook Audience Insights

Let’s assume here that we are building a persona using all of Facebook’s data, either because we are a new brand, doing some client research and we don’t have access to their Facebook account, or as mentioned earlier, our existing Facebook followers have been dirtied by either buying followers or previous advertising campaigns being too broad.

*If we decide to build personas based on existing Facebook followers, the process is extremely similar; however, if the number of followers is low, we may not be able to segment our audience interests as much as in the steps that follow.

1. First thing, head over to Facebook Audience Insights. You don’t need an advertising account — all you need is a Facebook profile.

2. Let’s assume we have a fictitious sports clothing brand who are trying to appeal more to runners. Enter an interest closely aligned to your brand or products; in this instance, it's “running.”

3. The initial search gives us some pretty broad options that probably aren’t that useful. However, they do indicate that of those interested in running, 60% of them are women, so let’s narrow by gender for our first audience persona. Remember, we will end up creating multiple personas for our brand — this is just one demographic we are targeting.

4. The results show that a large portion of our audience sits between the ages of 18 and 44 — however, that is once again a quite broad segment of our audience. Let’s focus on where the bulk of the market appears to be by also filtering by age, 25–34.

5. From the Demographics screen we can start to dig into the type of people who might be interested in our product and start building their persona.

Audience Insights categorizes our audience into Lifestyle demographics and provides us with a brief description about the type of lifestyle they may lead from which we can extract relevant information to filter our audience further. The table below indicates those lifestyles that fit best with our running brand for the demographics defined so far.

6. From here we can build a better picture of the type of person that may be interested in our products. Using the above information, we now know the following information about one group of potential customers.

They are:

• Females aged between 25-34
• Mainly homeowners
• Both singles and couples
• Working
• Mixture of childless or young children families.

Adding all of this information as facets and we have cut our audience down to between 300k and 350k monthly active people.

7. From here, we can drill into each of the individual tabs to extract relevant information about our target persona, such as:

• Demographics: Age, gender, job title, relationship status, education level
• Interests: Categories and page likes
• Location: Where they live
• Activity: Frequency of online activity and device usage
• Household: Income, home ownership, home market value, spending methods
• Purchase Behavior: Likelihood of online purchases, purchase behavior

Example of a persona that can be built using Audience Insights

We now have a decent amount of information that we can pull together to start to build the profile of the type of person we want sell our products to. We can keep this person in mind when pulling together any content or carrying out any type of marketing activities.

Below is just one audience persona that we might look to target for our fictitious running brand.

So, that’s all we have to do?

We would need to rinse and repeat this process for different demographics that we believe may be interested in our brand. But yes, these are some initial steps we can take to building audience personas that we can target our products and brand towards.

However, I’m not saying that this is the only task we have to complete to build audience personas.

Yes, Facebook has a lot of data, but they're still piecing together all the parts. In the same way we know we shouldn’t completely trust the numbers that Google Keyword Planner gives us, they just give us a ballpark to play in. As such, the data we are given by Audience Insights should be just one of multiple research methods we should be looking to use to build audience personas.

Big brands spend huge volumes of their budget trying to understand their customers. Many of us can’t compete on that level. We need to look to the tools and data we have available to us, and build the best personas we can for our budget.

Here are a handful of resources that can be used to help develop personas that won’t cost a lot, other than perhaps in time:

• Dig around Google (other search engines are available) for other surveys that have previously been carried out within your niche. Everett Sizemore discussed this in a post on inbound.org on how to carry out audience persona research on a budget.

• My colleague Craig Bradford took a look at how to use the the Alchemy API to build personas from users Twitter profiles
• Google Analytics has been giving us demographic data for a while now (if you’ve turned it on). A quick insight on how to do this can be found from Steven Sparber here.
• Actually speaking to people. Zachary Cohn has a great slide deck discussing 18 tips on running customer interviews.

What other processes do you use to help build audience personas? Have you used Audience Insights to build personas, and if so, do you have any further tips? Please share in the comments below or reach out to me on Twitter @the_timallen.