What’s a lookalike audience and why’s it important to my ad strategy in the lead up to the holiday season?

Before modern-day digital ad campaigns…

Brands were subject to blanket advertising strategies that targeted a huge swath of potential customers, but not necessarily the right customers. It’s like paying for a billboard on the freeway or in Times Square—you capture a ton of eyes, but not all are from your target audience. 

Now imagine if you could curate a list of your top customers pulled from your target audience, put them into a magician’s hat, tap it with a magic wand, and have the list grow ten times larger into a list of customers that precisely match the profiles of your top customers—after which you could hyper-target them with your ads. You would think that’s the hard part. It’s actually the easiest.  

Facebook is the magic hat and wand. It’s a treasure trove of microdata from billions of potential customers all over the world. 

On a macro level, the microdata of your curated list of top customers is dissected by matching the email address with a Facebook profile (because who doesn’t have Facebook?). 

Then, Facebook compares the aggregate data points against millions of other Facebook users to formulate a larger, hyper-targeted group of customers that match closest to the likes, preferences, interests, locations, and gender of your original list (amongst many other data points).  

This list is your treasure map to find the seeds to plant your cash crop and all you need to do is start digging.  

So, if that’s the easy part, what’s the hard part?

Curating the list of your top customers.  

Focusing on your top-spending customers is part of the strategy to attract more of your best customers, but it’s not everything. Average Order Value (AOV), Frequency of Purchase, Customer Lifetime Value (CLTV), and other factors play into the whole analysis that should be done in order to curate the best segments for use in a Lookalike Audience.

For example, if you’re marketing for a gourmet coffee brand and a particular customer spends once and buys your highest price product, they may be added to your list of top spenders. 

But if they never buy again from you, is that a customer that you want to replicate? Perhaps. But it’s better to dig deeper and find customers that not only spend the most but also have the highest Customer Lifetime Value (CLTV). 

What’s CLTV and why is it important to my business metrics?

If you want your money tree (they do exist) to grow strong with as many leaves as possible, you need to start by planting the right seeds.  

Using Facebook’s Lookalike audience is akin to adding water, fertilizer, and love to help them grow. The leaves that sprout are your larger, hyper-targeted group of customers that have grown from your initial curated list of top customers.  

More leaves mean more surface area to reach the sun (which is your advertising campaign) which eventually translates to more of the right traffic coming to your store.  

So, what are the right seeds?

The right seeds are essentially composed of your customers that have the highest CLTV.  CLTV for a customer or group of customers is roughly calculated as the average order value multiplied by the average purchase frequency rate multiplied by the average customer lifespan.  

As mentioned earlier, the customer who purchased your highest priced product once and never purchased from you again may not actually be a part of your highest CLTV segment according to the formula discussed above.  

But imagine a customer who initially purchased a mid-priced product and continued with a high frequency of purchase along with a long customer lifespan. That customer far outperforms the customer who bought once and is far more valuable. That customer is the right seed.  

Now you know what the right seeds look like, so let’s go get them. 

How do I find my highest CLTV customers?

Using Klaviyo’s advanced segmentation capabilities, create a segment that includes customers who:

• Have made a higher than average number of purchases, basing “average” on your brand’s historic average number of orders per customer  
• Have generated X amount of revenue over all time
• Have returned less than X amount over all time—this removes customers who may be more likely to return products whether it’s because your products were not right for them or because they habitually buy and return 

You can also bring in Klaviyo’s predictive analytics to determine high CLTV customers.  Klaviyo automatically builds a CLTV model using your company’s data and retrains the model at least once a week for ever-increasing accuracy. 

By playing around with these metrics, you can create a segment that is at least 1,000 people, which is the number needed by Facebook to create accurate lookalike audiences. 

When’s the right time to create more of those seeds and how do I do it?

When? NOW!

If you have a list of 1,000 customers with the highest CLTV on your site, wouldn’t it be better to have 10,000 or even 50,000? 

Your initial list of the highest CLTV customers drawn from your store may not be the most accurate sample. And just like a plant, you need to groom it and prune your list and create a larger and more accurate sample.  

It’s advisable to have a minimum of 1,000 emails to utilize for your initial Lookalike audience. But as my parents always said, doing the bare minimum gets you the bare minimum. 

Take your initial 1,000 customers that constitute your current highest CLTV customers, add them into a Facebook Lookalike audience campaign, water, prune, and repeat.  

As customers click on your call to action (CTA) ads from the Facebook Lookalike campaign, the iterative process of conducting the campaign will yield larger and larger versions of the highest CLTV customers that you have. 

As the list becomes larger, it also becomes more precise. It’s similar to watching a show on 4K TV in HD rather than a TV from the 90s in regular definition. Why? More pixels give you a more precise picture. 

The same happens with your highest CLTV audience using this process. Attracting more customers based on your best customers is a great way to grow especially now, during this time of market volatility. 

By creating a lookalike audience based on your highest CLTV audience, you are growing with capital efficiency, not wasting advertising spend on an audience that’s unlikely to buy or create high returns. 

What are my action steps?

Here are the steps you can take now:

• Go to your Klaviyo account, segment and find your initial highest CLTV audience (minimum 1,000 people)
• Add them to a Facebook Lookalike audience ad campaign and drive them to your highest LTV products
• As the ad campaigns drive traffic to your site, you’ll see your highest CLTV audience list grow and it’ll becomes more precise
• Start now so that when the holiday period approaches, you have the largest and most precise lookalike audience possible to drive to all of your fantastic deals

In the lead-up to the holidays, use your lookalike audience to generate new, valuable subscribers who are predicted to act just like your best customers do now.  

You can ask them for their email address directly in Facebook, lead them to an ideal landing page that introduces your brand, or, even better, run enticing sweepstakes guaranteed to generate even more leads.  

Subscribers from sweeps and contests are notoriously less valuable than other subscribers, but by using lookalike audiences, you can ensure a higher quality of participants and a far better-performing list.

Happy planting!  

How do you make the most of new customers coming to your site?  Check out how to create lookalike audiences with Klaviyo.