When it comes to the applications of AI in marketing the only limit is your imagination.  The possibilities are really endless and that is why the most important starting point is not the technology but the problem you are trying to solve.  In most cases a problem that has been draining resources, time and motivation can be addressed with AI and ML.  

We are hard wired to want to go for the big flashy AI executions that we can write about in campaign magazine.  However, more often than not the executions that automate marketing activities are the ones that have the biggest impact over the long term.  

If we break it down into three categories of use cases for simplicity’s sake they would be AI for personalization & creatives, AI for media buying, and AI for analytics.  

AI for personalization and creatives

Personalization is becoming more and more important as recent studies have found that over 70% of consumers expect a personalised experience from brands.  (Mckinsey & Company, 2021)  They also found that personalization increased their revenue from marketing activities by 40%. (Ibid). So how can AI help you improve your personalization and customer experience?  The first way is to ensure you leverage AI to gain an understanding of your customers based on their purchase behaviour, browsing habits, interests and intent to be able to segment them more effectively and reach them with ads, landing pages, offers and product suggestions that meet their needs and preferences more closely.  You can go a step further by utilising an AI powered chat bot that has been trained with the same data and can carry on a conversation with your customers that feels tailored to who they are.  

AI can also be used to improve the quality and effectiveness of creatives by analysing enough data on ads to ascertain what works and what doesn’t for different industries and types of products.  It can do so on banner ads but also on video ads.  In some cases they’ve even used AI to automate video ad creation.  

AI for Media Buying

Media buying is an area where AI has been in place for a very long time and perhaps marketers haven’t put much thought into it.  When you buy ads on Google, Meta, Snapchat or TikTok you are relying on AI to ensure you are placing your ads in the right places, reaching the right audiences and doing so at the most cost effective prices.  The algorithms that absorb audience, location, quality of assets, time, date, context and more do so in split seconds and have learnt, thanks to an astronomical amount of data, how to make the best decisions.  These split seconds would translate into days and weeks for us humans.  Sometimes it means you need to make sure you are leveraging whatever bid optimization capability is offered in the platform.  You’d be surprised that it’s not always done when it takes a minute or less to activate in many cases.  For more sophisticated executions you can even train the bidder with specific data (occupancy rates, route profitability, inventory data etc.) to go a step further (custom bidding).  

AI for Analytics

Predictive analytics is another area where AI can play a big part.  We are no longer at a stage where it’s enough to understand our audience, we need to know what their decisions will be in the future.  Training an ML model with a good amount of data helps do just that.  Privacy safe data that will be processed in a clean room of course.  For example, if you have enough data about purchases on your site or app you can determine the lifetime value of your customers.  Not just the ones who made a big purchase last week but the ones who are likely to continue to make large purchases into the foreseeable future. You can also use AI to determine the likelihood of churn amongst your client base and activate retention strategies to protect your market share.  Some analytics providers have AI built into their platforms directly, for example, Google Analytics has a feature called analytics intelligence that conducts a good deal of insight gathering for you directly.  It can even automatically generate smart audience lists, which are lists of users who are most likely to convert.  This takes a lot of work directly off your plate and possibly allows you to surface insights you might have missed.

It’s impossible to list all the different ways AI can make marketing more effective but these three areas are a good place to start.  Always start, as mentioned above, with the problems you are trying to solve so that you aren’t just using AI for the sake of AI.  Then go after the quick wins: are you leveraging AI/ML in your media buys?  Switching this on is incredibly simple.  If you’re using Google Analytics for example, make sure to use analytics intelligence.  Next are you collecting data in a clean and structured way so that you can make use of it?  Where is your data, do you own that data and how can you use it to train AI models?  What technology do you already have at your disposal and are you making full use of the AI and ML capabilities?  Answering these questions should allow you to get started on multiple AI projects that will result in significant gains for your business.