Introducing iROAS...
In this DEN Talks installment, Amazon marketing expert Jeremy Dodge provides a deep-dive into the intricate world of measuring advertising effectiveness, with a particular focus on navigating the complexities of Amazon's platform.
One of the central points of the discussion is that traditional ROAS is an unsuitable north star metric on Amazon. The challenge of accurately measuring incremental return on ad spend brings about iROAS--a metric that measures return on advertising spend with a weighted coefficient (i) that takes into account that no advertising effort is 100% incremental when it comes to impact in driving conversion. This is especially true on Amazon, which has among the highest user purchase intent of any ecommerce website out there.
Jeremy makes a point to call out that any model or vendor claiming to provide a completely accurate measurement of iROAS is inherently flawed. There is no way to perfectly measure incremental impact, especially as the quantity of impactful touch points across an increasing array of marketing channels across the funnel continues to rise. While iROAS metrics offer directional guidance, they often fail to capture the full spectrum of advertising effectiveness.
What are the top strategies for testing and learning? For one, Geo Holdout Tests. Identify similar DMAs, run ads in only one, and measure the impact on sales. Another is Media Mix Modeling, which, while expensive, is one of the best data-backed methods of individualizing the impact of each small piece of your overall marketing pie. Finally, don't forget about Customer Surveys--QR codes in your product packaging can do the trick--and you should always obsess over your products' Organic Rank. As Jeremy says, a sale from an organic listing has infinite iROAS!
Metrics like iROAS provide valuable insights, but they should not serve as the sole barometer of success. Building scalable tactics for how to avoid serving duplicative or unnecessary Amazon ads does not detract from the importance of optimizing organic content and strategically targeting high-value keywords. Of the utmost importance is relying on your own empirical methods of ongoing ‘testing and learning’.
In short, always embrace experimentation, continuous learning, and a healthy dose of skepticism in the evaluation of incrementality-weighted metrics. Measuring iROAS is inherently imperfect, but a better understanding of how and what drives customers to make a purchase on Amazon is integral to marketing efficiency. Freeing up low/no impact direct response spend allows for more investment in upper funnel activities that engender a stronger connection to your brand, especially on a channel like Amazon where brand storytelling and stickiness are inherently lacking.
For any questions on the topic covered in this session, or to connect with Jeremy directly, reach out to hello@theden.co.
