The approach to audience targeting is constantly evolving. In the 2010s, the best practice was to use incredibly niche targeting based on specific interests, traits, and demographics. We had more targeting parameters available to us, and adding inclusions & exclusions were the norm. However, in the 2020s, a new trend has emerged due to changes in privacy laws and a decrease in the accuracy of interest-based targeting. The current best practice is to embrace a broader approach, believing that “the bigger the audience, the better”.
Don’t fall into the trap of using a single campaign and ad set, as this limits you to one optimisation and one audience. We strongly advise a detailed testing strategy that makes use of Meta’s different audience groups and funnelling. We’ve laid out the specifics of this approach below.
Understanding the simple targeting funnel
When strategising a Meta campaign, it’s helpful to structure your audience approach around a simple funnel. This funnel breaks down your potential audience into three core groups based on their relationship with your business.
| Audience type | Description | Typical budget allocation* |
|---|---|---|
| Prospecting | Finding new customers who are not yet aware of your brand, have not purchased, or have not converted before. Requires testing different audience options. | 70-80% of monthly budget |
| Re-targeting | Building a deeper connection with users who have visited your website or interacted with an ad but haven’t converted. The goal is to convert them. | 20-30% of monthly budget |
| Retention | Focusing on existing customers to retain, upsell, or get them to convert on other services. Used to lower churn. | 0-10% of monthly budget |
*Allocation may change depending on your actual campaigns. We also recommend being fluid with how you use your budget – make it work for you and your needs!
Prospecting audiences
Prospecting is the top of your marketing funnel and involves finding new customers who aren’t yet aware of your brand, haven’t purchased, or haven’t converted. In Prospecting, you must investigate and test the various audience options available on Meta. The three primary types of prospecting audiences are Interest, Broad, and Lookalike (sometimes referred to as “LAL” or “LLA”).
| Audience type | Description | Key features & best practices |
|---|---|---|
| Interest Audiences | Built using Meta’s knowledge of what a user likes or is in the market for. |
|
| Broad Audiences | Allows social media algorithms to handle the targeting. |
|
| Look-a-like Audiences (LAL/LLA) | Finds new users who are similar to your existing customers. |
|
Ultimately, the key rule for prospecting is to test all three types (interest, lookalike, and broad) against each other in your campaigns. Remember, we never make assumptions as to what will work when it comes to paid ads – we test and check the real data!
Interest audiences
Interest audiences are built on Meta’s understanding of a user’s likes and market behaviours. When setting up a campaign, you can target people based on specific interests, such as “Small business owner” or “florist”. For these audiences to be effective, a recommended size is generally between 300,000 and 500,000 users. It is advisable to use multiple interests within a single persona to achieve this size and avoid making the audience too specific. Testing multiple interest audiences against each other is crucial to finding what works best.
Broad audiences
Broad audiences represent a significant targeting trend in the 2020s. This method gives the platform’s algorithms the power to do the targeting for you. With broad targeting, you only specify basic demographics like age, gender, and location, and the algorithm handles the rest. Meta’s algorithms are highly effective because they have access to real-time data that advertisers do not. For this strategy, the rule is simple: “the bigger the better”. A larger audience allows the algorithm more flexibility to find the users most likely to respond to your ads.
Lookalike audiences (LAL/LLA)
Lookalike audiences are a powerful tool for finding new users who share similarities with your existing customer base. You create them using a “seed audience” (a source of data like a customer list, mailing list, or website pixel). Meta then finds new users who have similar demographic or behavioural traits to those in your seed audience.
The percentage you choose for a Lookalike audience is a key factor in its size and similarity to your seed audience. A 0% lookalike audience is the smallest and most similar, while a 10% audience is the largest and least similar. It is often a good strategy to test different percentages, such as 3%, 6%, and 8%, to find the optimal balance between audience similarity and size, which can lead to the best cost per acquisition (CPA).
Setting up retargeting and retention campaigns
To set up retargeting and retention audiences, you must have a working Meta pixel and be familiar with Audience Manager/Events Manager. This is where you can create audiences based on specific visitor data or event data. To create these audiences you will need to go to Audience Manager and set up a Custom Audience.
| Audience type | Description | Source data |
|---|---|---|
| Retargeting | Targets people who have previously engaged with your brand but have not converted. | People who have visited your website as a whole during a set period or those who have visited particular pages.
People who have taken action on your ad. People who have engaged with you in any way on socials. People who have taken action on your website, e.g., added an item to a cart but not purchased. |
| Retention | Targets existing customers to encourage repeat business or engagement. | People who have visited customer-centric areas of your website in a specific number of days.
People who have completed conversion actions on your website in the past. A list of customers you have access to (uploaded to Meta). |
While a 20-30% budget is a good starting point for retargeting, it’s important to monitor your campaign’s frequency. Frequency is the average number of times a person will see your ad. A high frequency, especially above 3-4 over a 7-day period, means you might be annoying your audience and wasting money on people who aren’t interested. You can monitor this metric by customising your columns in Meta’s ad reports.
