46% Increase in ROAS through Product Feed Optimisation & Strategy

Andrew Martin has established itself as a prominent figure in the realm of global interior design, showcasing its imaginative, fusion-driven, and enjoyable interiors since 1978. Recognized for promoting top designers and shedding light on hidden cultures, Martin’s contributions have earned it the well-deserved title of the “bible of interior design” through their publication, “The Interior Design Review.” In early 2021, Realtime embarked on a partnership with Andrew Martin, aligning their goals with the brand’s focus on expanding their e-commerce endeavors and increasing their return on ad spend (ROAS). While our team possessed a robust e-commerce offering, our success with Andrew Martin was particularly notable in the realm of product feed optimization, which ultimately resulted in an impressive 46% incremental increase in ROAS. So, how did we achieve this remarkable outcome?

Challenge

Product feed ads on Facebook & Instagram are a common sight while scrolling through the platform. These ads, posted by brands, often feature a carousel of merchandise along with pricing information, coupons, and a ‘shop now’ call to action. Unlike regular static ads, product feed ads are dynamic and linked to the retailer’s real-time inventory through SKU numbers. This means that if Meta identifies a user’s potential interest in a couch, they can display ads featuring various couch options from the brand. But how does Facebook or Instagram determine a user’s product preferences? While the algorithm and pixel data play a significant role, we also have a part to play in guiding it. This is where product feeds become useful. Advertisers manually compile lists of products into these groups, which can be sorted based on various criteria such as color, material, type, or promotional aspects. By optimizing the characteristics of products within these groups, we can achieve maximum advertising efficiency. For example, if Instagram identifies users who have recently purchased a couch, we can instruct it to retarget those users with ads for complimentary products such as throw pillows from our ‘throw pillow’ product group. Similarly, if Instagram determines that users have purchased linen items, we can retarget them with other linen items from our ‘linen’ product group. The possibilities are endless, and conducting tests and gathering data on product groups is essential for a successful product feed strategy.

Strategy

In the realm of social media marketing, it is crucial to acknowledge the significance of optimizing individual channels to drive better performance and achieve a high return on investment (ROI). While our focus for this article revolves around Facebook and Instagram, it’s important to note that Realtime has also customized product groups for Pinterest and Google, tailoring strategies to suit each channel’s unique characteristics and user behaviors. Surprisingly, many advertisers underestimate the importance of channel-specific optimization, but it plays a pivotal role in driving success at scale.

To illustrate this point, let’s take a look at our experience with Andrew Martin’s inventory, comprising nearly four and a half thousand products. We delved into sorting and grouping these products to determine the most effective strategies. Eventually, we found success by categorizing them based on color, differentiating outdoor and indoor items, and highlighting promotional items. These groupings allowed us to tailor our approach, ensuring relevant and engaging content for each category.

Moreover, our success extended to leveraging Instagram’s dynamic overlays. These overlays, small PNG files, dynamically appeared on top of creative assets based on specific rules. For instance, in an effort to incentivize users to return to abandoned shopping carts, we retargeted individuals who had added items but not made a purchase. By overlaying a ‘10% off’ coupon on top of the ad, we were able to entice users back and increase conversions.

In addition to strategic optimization, personalization played a pivotal role in our overall approach. In the context of modern-day e-commerce, personalization is a fundamental component, especially in a digital landscape filled with sensory overload. We embraced personalization at every step of our strategy. One notable example was incorporating product characteristics into the ad’s headline. By highlighting key features such as material, color, or other relevant aspects, we observed a boost in return on ad spend (ROAS). Users were more likely to engage and click on ads that resonated with their preferences and showcased the unique qualities of the product.

Results

Regardless of your product feed approach, tailoring is the key. Through promotions, and messaging to the individual, brands can stand out from competitors and boost not only share of voice, but – ultimately – sales. For us, tinkering to find efficiencies in product groups, dynamic overlays, and ad headlines showed us a 46% increase in ROAS in just two weeks (compared to data two weeks prior to using product feed ads and steady spend/CPM [<10% difference]).