Marketing with ML – Part I

Marketing with ML – Part I

Part 1: Uplifts, Recurring Purchases and Starbucks

The evolving role of machine learning in marketing is transforming how businesses work – it is now an indispensable tool in remaining competitive.

In 2019, American bank JP Morgan Chasepartnered with AI-powered text generator Persado in a five-year deal to improve its marketing copies. In its pilot program, Persado reports, “Chase saw as high as a 450% lift in click-through rates on ads rendered by Persado, compared with others in the 50-200% range.”

Sample the following:

Human copy: “Go paperless and earn $5 Cash Back.”

Persado copy: “Limited Time Offer: We’ll reward you with $5 Cash Back when you go paperless.”

Results: AI copy generated nearly double the clicks.

The objective was to use optimised keywords to make the copy sound not only more human but also more click-friendly. This is an example of an uplift powered by machine learning. As summarised by VentureBeat, the online technology news platform:

“ML is also highly adept at gauging the incremental effect of a marketing campaign at the user level, as well as revenues, sales and other data, and then making predictions as to how this uplift will play out into the future.

“Algorithms can be used to simulate consumer reactions to special offers and other elements, which not only helps to guide them toward completed sales, but can lessen the cost of these efforts by more accurately targeting them to the right users, or discontinue the lowest performers altogether.”

How ML turns out to be beneficial to marketing

The role of machine learning in marketing has assumed greater importance today than ever before. The objective is to enable improved decision-making for marketing – especially digital marketing, in aspects such as personalisation, forecast targeting, lifetime value modelling, chatbots, smart bidding and customer segmentation, among others.

Machine learning can considerably increase the flexibility and speed of a bunch of marketing processes – but the trick lies in knowing what to use when, and where. There is no real one-size-fits-all approach that can be applied, but essentially it can fall under three categories – descriptive (based on past events), predictive, for forecasting and planning and prescriptive: for optimal courses of action.

Of the three, predictive and prescriptive are most commonly used to build machine learning algorithms while descriptive analytics apply mostly to dashboards and reports. Some businesses, in this regard, spend up to two years in data accumulation in order to adequately analyse consumer behaviour and in the personalisation of customer relationships. Yet, ML needs to be applied strategically to any marketing process.

Recurring purchases becomes the hallmark of a successful marketing campaign, especially for organisations experiencing dramatic scales. A well-trained model helps businesses not only maximise chances of purchase, but also the exact moment to engage existing customers as well as identify and recommend supplemental items based on previous consumer data.

VentureBeat writes:

“This requires careful analysis of multiple data points, however, such as the number of orders made in the past, the average order value, frequency of purchases or other factors.

“There is also often a narrow window in which a follow-up email will result in an additional purchase. Hitting this mark on a consistent basis has been shown to considerably boost click rates.”

Putting the tech in Starbucks

Starbucks, for example, obtains data from all its stores around the world to access purchase insights and by turning this information into marketing collateral using the mobile app and the loyalty card, they use predictive analytics to boost revenues and marketing. For example, “machine learning collects the drinks each customer buys, where they buy them, and when they buy them, and matches this with outside data such as weather and promotions to serve ultra-personalized ads to customers.” One instance even “includes identifying the customer through Starbucks’ point-of-sale system and providing the barista with their preferred order.”

Check out the following infographics to better understand the Starbucks marketing story:

For FMCG giants with millions of customers to give out ultra-personalised recommendations, it is crucial that they are able to sift through huge troughs of data with speed and efficiency – something they do consistently, and well.

Acknowledgement: Shikha Mehul Parikh at


Know more about the syllabus and placement record of our Top Ranked Data Science Course in KolkataData Science course in BangaloreData Science course in Hyderabad, and Data Science course in Chennai.

© 2023 Praxis. All rights reserved. | Privacy Policy
   Contact Us
Praxis Tech School
PGP in Data Science