How PMG built a real-time marketing analytics platform to increase return on ad spend.
Name of Organization: PMG Worldwide, Inc.
Industry: Marketing, Retail
Location: Ft. Worth, TX USA
Opportunity or Challenge Encountered:
Whether you’re into cycling, baking, or home renovation, then you likely are open to receiving advertising from appropriate equipment and apparel providers. This is called affinity marketing, because you have likely been identified as enthusiastic about your pastime. However, the challenge for marketers is to see and understand, in real time, what kinds of searches or transactions you are conducting.
In this mobile and web-intensive world, marketers and advertisers must be able to use real-time analytics to target consumers in a more finely tuned way by offering unique, personalized messages that are seen in the right environment at the appropriate time. While the technology is in place to deliver the right data, marketers struggle to connect with their customers in the optimal context because of audience pools that are incomplete, inaccurate, and siloed from activation.
Such was the challenge for PMG, a digital agency and leader in audience marketing, which sought to improve marketing personalization for its client, a global beauty brand. The client was grappling with a limited ability to target category and product pages due to pixel constraints on its website, as recounted in a recent case study. The company’s site did not directly allow for more granular targeting with specific messaging aligned with pages that users were frequently visiting.
PMG sought to help its client deliver its ads to audiences in ways that were optimized and personalized. To achieve this, PMG knew it needed alternative audience-based solutions that would allow the agency to reach additional users and pages within the brand’s site to support product consideration, and ultimately drive demand on third-party sites.
How This Opportunity or Challenge Was Met:
To meet the needs of its beauty products clients, PMG partnered with MediaMath to develop a strategy to construct audience profiles in a way that would allow PMG to target new and dynamic users in real time. The team employed data layer segmentation to develop custom audiences in order to reach users based on their onsite content consumption and throughout all stages of the purchase path, the case study relates.
PMG worked with MediaMath to collect data into a centralized platform and turn the raw data into usable marketing signals. As data was ingested, PMG was able to use a series of four key “adaptive segments” to create and build precise consumer profiles that populated retroactively, and update those profiles in real-time. This provided PMG and its client the ability to quickly identify customers at various stages of search and transactions.
For campaigns to “co-op-based segments,” PMG customized audience segments based on users that browsed the applicable co-op brand and products within the retailer’s site, and used segment sizing to understand how many people they would actually target for a specific brand.
For what MediaMath brands as “high-value cart abandoners,” PMG and its client can identify individuals with basket values that were higher than the average order value. Then, there are “high-frequency browsers,” or frequent visitors who are highly engaged based on product page activity, but have yet to purchase. Finally, there are “high-value converters,” or purchasers with order values that are higher than the average order value.
Benefits From This Initiative:
The campaign delivered impressive results, PMG reported in the case study. During retail’s crucial fourth quarter, the agency’s client saw audiences scale throughout the time period, with conversions peaking during key shopping holidays.
Some stand-out outcomes included improved return on ad spend (ROAS), the case study reports. The creation of the audience profiles was validated by ROAS that reached as high as $149.59 per customer.
Real-time engagement also resulted in greater efficiency, with 20 different variables passed back. Audience targeting was more granular as well — instead of having to deliver generic messaging and generate strong return at a high level, PMG is able to serve more personalized and custom messaging to a finite group of users, depending on their browsing behavior. PMG is working with its client to access supply markets and work on an even more extensive data strategy.
(Source: MediaMath)
