During my training sessions at Cibirix, I stumbled upon a very relevant question from one of my marketing interns. His question was straightforward and very relevant to any small or medium-sized business that has no experience with Facebook ads.
“How Do I Build an Audience When I am Just Starting on Facebook?”
Facebook Ads targeting is one of the main reasons why ads may fail. You may have the perfectly crafted video or the nearly optimized canvas ad that you are ready to show to the world. However, you too know that the ad may fail if you do not know who you’re targeting.
During that conversation, I told the intern about the power of Lookalike Audience. Lookalike Audience is Facebook’s magical way to multiply your seed audience by matching their attributes against the people on Facebook and finding the best audience for your purpose.
Small Initial Audience
If you are a company just starting with Facebook advertising and wondering about the small sample size that you have gotten from your website or your offline marketing efforts, worry no more. You can use these seed set of audiences to create a profile of your ideal users who will buy from you. Unlike, targeting people on Facebook based on interests and then narrowing them on geography, you can now target people who closely match your existing customers and existing prospective buyers.
Lookalike Audiences can be a great tool for your business and you should look at them to target cold leads at the top of the sales funnel. How cold? Well, that depends on the % similarity that you choose for building these audiences. As a Facebook advertiser, you could ask Facebook to be very specific (1% match) or very broad (10%) or anywhere in between. This % match in Lookalike Audiences tells Facebook to limit to just 1% of the top match results.
Lookalike Audiences – Size Does Matter, but Relevancy is the Key
We have typically observed a seed audience set of 1000 to 25000 performs very well to create a Lookalike of between 1-10 million Lookalike users. Please remember that while a larger number sounds very enticing, it is always good to mirror Interests or use the famous “And” operator of Facebook Audience targeting when running ads. We say so because we do not want your ads to perform poorly. Think about it, these people have most probably never heard about your company and your products. While they may be like your prospective buyers, they need to first warm up. A reach and frequency buying type campaign targeted for brand awareness or the reach objective works the best for a Lookalike Audience.
How to create a Lookalike Audience
Lookalike Audiences can be created from any other Custom Audiences. Cibirix has been creating Lookalike Audiences from our clients’ customers list for them to efficiently create audiences that match the attributes of this list. To get the highest match rate possible from this data, we include as many data types as we can. While using a single-column file to create your audience can technically work, but we may not see as many matches as we do when we use additional data types. These data types include both First and Last Name, email address, phone number, city, state, zip, country, and age.
We also create Lookalike Audiences based on website visitors and the client’s Facebook pages. It is possible to achieve a great Lookalike Audiences list with these methods and we refine these audiences based on the category of the products or sometimes even a single product. Yes, when you create a custom audience, you can include (and exclude) website pages. When you create audiences based on these page visits, you are telling Facebook to find people on Facebook who are likely to be interested in that product or product category.
A great example of this industry is for the real estate businesses. When we advertise for our local realty clients, it makes sense to target people based on the attributes of the people who visit a range of homes between a $$ amount. We then narrow down these audiences to people who are:
- Living in that area (city/suburb/county
- Moving to that area
This helps us target users like those who visit our website and might be interested in finding a home in interest to our client.
Lookalikes of People Who Buy from Your Website
We always look for eCommerce conversion data through Pixel based conversion tracking. When we get conversions for our eCommerce clients, we build upon that advantage. We have developed in-house tools that fully sync your custom audience with your CRM. When a new customer is created, we not only upsell and cross-sell to that customer using their Lifetime Value, but also create an always updated Lookalike Audience based on this custom audience. This helps us find more prospective customers who match the profile of our clients’ customers.
Use Lookalike Audiences to Sell Anywhere in the World
Let us consider the case of an eCommerce startup company that is selling merchandise in US and Canada. Through the sales and marketing efforts, this company has been catching the eye of US audiences. However, they are still relatively unknown in Canada with very fewer sales. Lookalike Audience on Facebook can be used for this company to reach new people in Canada who are like its American customers.
Lookalike Audience will only have people from a country or region that you select while creating this audience. You can create multiple Lookalikes of a seed audience targeting different countries. You don’t have to have people from the country/countries you want to target in your source audience.
Value-Based Lookalike Audience
Lifetime Value of a customer refers to the net profit you will earn from a customer over the entire lifetime of your relationship with her. Some of your customers will buy once, the others will always be ready to buy multiple times with varying frequencies of purchase. When you’re creating a Custom Audience in Facebook, you can add Lifetime Value of the customer as an additional field to your source CSV (or excel) file to tell Facebook about the best customers, how much they buy, and how frequently they buy. All this is represented by a numerical value that you put against the customers. If you normalize the data, make sure the scale is correct. Do not rate or rank your customers because the scaling usually doesn’t work. View this excellent help article from Facebook on Customer Lifetime Value.
Once you have your customer lifetime value added to the data file, you can create the custom audience. From this custom audience, you can create a Value-based Lookalike Audience. You’ll need to select the country for each audience you create.
What Problems Does LTV Solve?
When you add Lifetime Value of customers to your Custom Audiences on Facebook, you can create a very closely matched Lookalike Audience. In this article, we discussed about how the 1% to 10% match leads to cold audience with top of the funnel marketing opportunity, however, if you carefully build your LTV column and tell Facebook about what which customers are higher LTV to you and which ones aren’t, that data can create lower funnel audience for you. You can sell your product or service to them rather than just targeting subscribing to a newsletter or reading a blog post.
In this article, I showed you how you can create a top of the funnel and bottom of the funnel audiences using Facebook’s Lookalike Audiences. This comes handy for sales managers looking for new selling opportunities for companies that are new to Facebook Marketing. Will you be using any of these strategies? Do you have any other insights that you use for better Facebook targeting? Feel free to share with us here.