How can you use lookalike audiences to maximize e-commerce sales?

Lookalike is like forecasting. It's highly dependent on the history of your orders and customers. The larger the customer base that you can leverage the better options you have to create lookalike audiences and leverage them through conversion.

To create lookalike audiences for eCommerce sales, start by using a platform like Facebook Ads or Google Ads. - First, upload your existing customer data, such as email lists or website visitor data, to the platform. These platforms will then analyze this data to identify common characteristics and behaviors of your current customers. - Once this analysis is complete, you can create a lookalike audience based on these traits. This new audience will consist of people who share similar demographics, interests, and online behavior with your existing customers. I think this can target potential customers who are more likely to be interested in your eCommerce products or services > increasing chances of driving sales and growing revenue.

Constructing lookalike audiences is a potent tool in e-commerce, and platforms like Facebook, Google, and LinkedIn facilitate this. Begin with a quality source audience that meets platform criteria. On Facebook, for instance, you use a custom audience—those familiar with your brand—as a foundation. From there, select the geographic scope and desired size. Whether targeting 1% or 10% of a population, it's essential to ensure that the lookalike audience closely mirrors the buying habits and interests of your source audience for maximum sales impact.

Your original audience should have good data behind it. First-time purchasers? Purchasers past a certain $ threshold? Repeat buyers? High LTV customers? Cart abandoners? Hopefully, you have good analytics in the back to bring over solid segments to FB and Google.

When it comes to lookalike audiences and targeting, I've found Amazon does it best. Since they implemented AMC (Amazon Marketing Cloud) they allow advertisers to link their 1st party data to Amazon's 1st party data. No cookies needed here. Then AMC allows you to study the overlap between your users, similar users on Amazon, and also users off of Amazon with their DSP functionality. The result is extremely powerful audience creation. Some of my favorites are people who searched or added to cart but didn't buy in my category, those who purchased during a promotion, those who are moving down the funnel, and more.

Leveraging lookalike audiences can be a game-changer in boosting your e-commerce sales on LinkedIn. These audiences enable you to connect with potential customers who are highly likely to be interested in your products and responsive to your advertising efforts.

Specifically for ecommerce, sharing my own experience with: * Facebook lookalike audiences: Algorithm matches user demographics, interests and behaviours. Audiences extrapolated to 1% of user population, although smallest in size, proved to be more effective. Compared to audiences based on 3% of population (never tried any higher percentage ranges). * Google similar audiences: Algorithm matches user demographics, search history and content of websites browsed. Audiences created from customer match lists based on actual sales transactions (eg on Shopify) were far more effective. Compared to audiences created from email subscriber lists, 'all website visitors' or even 'cart abandoners'. Google has removed similar audiences from Aug 2023.

You can use data from lookalike to massively personalize and merchandise your website, some ideas e.g. 1, A 'these items are trending in customers like you' list on home page 2, Promote items viewed by lookalike audiences for the customer on listing/search page 3, A 'People bought this also bought that' using lookalike audiences data

Experiment with varied sources, sizes, and ad formats – from website visitors to email subscribers. Optimize by segmenting sources and combining audiences for better results. Test dynamic ads to personalize product offerings. Achieve your goals efficiently with these strategies.

Consider the power of feedback loops in optimizing your lookalike audiences. Regularly analyze your campaign performance data, refine your audience criteria, and adapt to changing customer behavior. Leverage A/B testing to fine-tune ad creatives and messaging, keeping a keen eye on what resonates best with your audience. Lastly, don't underestimate the importance of staying up-to-date with platform updates and industry trends to ensure your lookalike audience strategy remains cutting-edge and effective.

For those interested, here is a historical timeline of when each platform introduced 'lookalike audiences' for advertisers to use: * Facebook: 2013. * Pinterest: 2016. * Google: 2017 (discontinued in 2023). * LinkedIn: 2019. * Tiktok: 2019.