Amazon Marketing Mix Modelling

Entropy is one of a handful of companies in the UK with access to the Amazon MMM data feed to help you measure and get deeper insights into your investments on Amazon.

Ecommerce continues its rapid evolution. Your own website is now just one of your online sales options. The rise of third-party ecommerce platforms has been followed by new marketing opportunities to support your sales effort, meaning retail media is now the fastest growing marketing channel.

So how do you decide how much to invest in retail media driving sales on Amazon vs media driving direct to your website? Which is more cost-effective, Google PPC or Amazon Search, Facebook, or Amazon display?

Entropy DMMM (Dynamic Marketing Mix Modelling) offers a like for like, privacy safe, performance attribution solution. DMMM allows you to make the right investment decisions across your sales portfolio. Optimise your performance across both your website and Amazon based on the marketing gains you can achieve on each.

DMMM model

Amazon Ads: Spearheading the rise of retail media

Amazon is clearly historically well established as the largest international online retailer, but the scale of its advertising business is perhaps less well known. In terms of ad spend, Amazon Ads is actually a larger channel than YouTube or Google Display, as well as the entirety of the global newspaper ads business (source: Global Advertising Revenues, WARC 2023). 

As well as the perhaps more familiar product discovery/search ads on Amazon commerce websites (Sponsored Products, Brands and Display), the Amazon Ads ecosystem also comprises Display & Video (e.g. DSP Video, DSP Display, Fire TV) as well as Streaming (including Prime Video, Freevee and Twitch).

Such diversity means that Amazon Ads can accompany a typical customer journey at all stages of the funnel – but the wide range of accompanying campaign metrics can be confusing.  With the ongoing need to prove true ROI from marketing investment, this can be tough with standard Amazon Ads attribution, which is based on last touch with a 14 day lookback window.

This means that Amazon Ads won’t tell you the true impact of all activity within the Amazon ecosystem itself as well as from other media (Facebook, Google, brand activity etc) or external factors such as seasonality, competitors or pricing and promotional activity.

How Entropy can help

As the reach of Amazon Ads grows and it becomes ever more sophisticated, Amazon is increasingly leaning into marketing mix modelling (MMM) for holistic evaluation.

Entropy are official MMM partners with Amazon ads, which gives us speedy access to highly granular data covering Amazon Sales and Amazon Ads. You can find out more about our overall MMM services here – or get in touch to find out more about our Amazon specific MMM services.

Our process for Amazon MMM typically follows 4 main stages:

  1. Scope: Understand brand and market background, identify brand modelling objectives, align with stakeholders on key objectives and ways of working to refine scope and outputs, develop project plan.
  2. Discover: Hold data brainstorm, data provision and preparation of modelling dataframe, feasibility study with ‘go/no go’ decision, review and harmonise all existing data sources and models.
  3. Quantify: Full model build & quantification of incremental impact of marketing/media drivers and ROIs, run additional analysis to address all key questions, transparent partnership including access to model outputs.
  4. Action: Translate model findings into key insights, workshop learnings among functional stakeholders and agency partners, embed and activate insights and recommendations.


Sign up to our Newsletter

We’ll send you the latest digital commerce related brainfood. We’ll also let you know about anything interesting we are up to.

This website uses cookies to ensure you get the best experience on our website.