VIETNAM - Room 801 No.41, Street 41 Thao Dien Ward,


Machine Learning Use Cases in Digital Marketing

June 12, 2019

With the rise of technologies, digital marketers have a wide variety of tools to make the best of their campaigns. Among those, machine learning is becoming more and more popular in the digital marketing landscape.


What is Machine Learning (ML)?

In the most simple terms of computer science, ML is an algorithm that improves on some task with experience. Basically, you can feed a set of data as input for the machine to learn, so that it can predict the output from new data that it has never seen before.


Application in Digital Marketing

But, how can machine learning help marketers?

  • ML can analyse over an enormous amount of data, to collect useful information and reliably predict the outcome, such as if this segment of the audience would be interested in / attracted to what kinds of campaign, or more importantly what kinds of product / price range / service / etc.;

  • ML can also be trained for ‘Programmatic Advertising’ - in the day and age of digital ads, bidding is an important step to gain access to your audience, and ML can optimize that by adjusting bid in real-time based on whether or not users are clicking on their ads;

  • And lastly, for marketers, ML’s ultimate goal is real-time personalized advertising distribution across all platforms. ML can learn the user’s preference over time, about which ads are the most effective, and deliver optimized messages to individual customer.

Below are some in-real-life use cases of companies working alongside with machine learning algorithms.


Use Cases


1. LuBan - Visual Design AI

LuBan is Alobaba’s group AI that is built to generate banners and other user-interface products on their website, personalized for each customer and product range. With LuBan, companies would not need to pre-designed their display advertising, but instead leave it for the powerful AI to automatically generate contents in seconds with a personalized touch for each kind of customer and their interested product.

LuBan is powered by machine learning, and is trained on millions of sets of design data.

 LuBan is generating new banners based on selected design style


LuBan was an immense help on major shopping festivals, such as China Single’s Day Event, as it helped filling the Alibaba’s shopping website with all visual content, without the need for the human designers to spend time on it.


2. Google Ads - Smart Biddings


Marketers don’t need to be coder or data scientist to use ML algorithms tools, especially when Google Ads has all kinds of tools that just need to be set up and is ready to use.


Smart Bidding is defined as ‘a subset of automated bid strategies that use machine learning to optimize for conversions or conversion value in each and every auction—a feature known as “auction-time bidding”’ (Google Ads Help, 2019). Within Smart Bidding, there are a lot of smaller strategies for marketers to choose.

 Strategies correspondent to each stage of a conversion funnel


Smart Bidding is helpful for large or small businesses to attract customers with the benefits of maximizing the most Return on Investment. Marketers can set up parameters on when to maximize bid for the best performance, or minimize bid to conserve budget on dead end leads.

 Smart Bidding examines the pre-set parameters to adjust its bid


3. Custom Machine Learning Projects - Kaggle


Kaggle is a treasure trove for more technical marketers, where there are numerous projects on ML algorithms building that would open up to the idea of a custom model for your own company, just like LuBan is Alibaba’s in-house model.  


 A Machine Learning model construction - Classification model


Or, you can post your own machine learning driven project on Kaggle to attract millions of eager data scientists out there who are more than willing to help build your solution.


You can encourage internally with your marketing campaigns to be more data-driven with these few steps:

  1. Based on business objectives, define relevant KPIs to monitor your campaigns;

  2. Setup tracking for those KPIs and ensure that the quality of the data collected is up to standard;

  3. Visualized the collected data so that employees can be more aware about what they are doing is making impacts on the business goals & objectives;

  4. Analyse on a regular basis on your KPIs and find insights you can leverage;

  5. Optimize your businesses with AB tests.


Here are some marketing-centric ML projects on Kaggle that I found interesting:



Created in 2017, DRIVE is a data & analytics consulting company providing clients with a data-driven business strategy, powered by skilled talents, reliable data, and scalable technology system. Certified on major analytics and conversion solutions such as Google Analytics, AT Internet and Adobe Analytics, we combine our expertise to meet your business expectations. With experience in both brand and agency matters and working within international environments for more than ten years, we can accompany you in facing your challenges.

Please reload

Our Recent Posts

Metabase vs Tableau

October 16, 2019

5 Marketing trends applied to dominate e-commerce market

September 12, 2019

Build vs. Buy: What is best for your project?

September 6, 2019

Please reload