Facebook rolls out Lookalike Audiences feature to replicate fan bases
Available to Power Editor accounts globally, advertisers upload a list of first-party data such as customer email addresses, phone numbers or user IDs (the sort of information they already have on their existing fanbase) and hit go. Facebook’s algorithms then analyze this custom audience and then, based on interests and demographics, produce a new, similar group of users.
Advertisers must choose which country they wish to target, and whether they want to optimize their searches for similarity or reach (similarity will yield more precise, but fewer, results. Reach, obviously, vice versa). These newly created audiences – which may take up to 48 hours to assemble – can be combined with other Facebook targeting options and used for any kind of advert.
It’s an exercise in blind faith, though, as the personal information of the new audience members is not shared with advertisers and can’t be exported to any site or software outside of Facebook.
Can advertisers expect results, then? According to InsideFacebook, Facebook has previously stated that while using the feature in limited beta, an online travel site saw 70% low costs per action using Lookalike Audiences, while an online shopping site saw 56% lower CPA and 94% lower costs per checkout. The official blog post also promises larger purchase size, and faster and increased return on investment.
It sounds promising. Let us know if you’re using the feature and how you get on.