AI for Marketing Today: Overcoming New Obstacles

This is part seven in our series designed to help businesses gain efficiencies, cut unnecessary costs, and protect their bottom line in today’s uncertain and complex landscape. The Kepler platform addresses a vast array of Machine Learning (ML) use cases for Marketing, Media and Advertising – and its easy-to-use Automated Data Science Workflows ensure all leaders in these spaces can leverage ML-powered insights quickly and effectively. Below, we outline how to leverage AI for marketing to make the most out of every single campaign.

The new marketer’s landscape: changing media habits and fluctuating consumer behaviors

While COVID-19 has torn through countries across the globe, people everywhere have been faced with new realities. From conducting typical day-to-day tasks like working, socializing and exercising from screens, to dealing with the new normals of shopping and homeschooling, the impacts of the pandemic have been felt in virtually every aspect of daily life.

It’s not surprising, then, that consumer sentiment has changed considerably. McKinsey has created consumer sentiment reports for several global regions, that track how consumers are currently spending, and how they envision the future. In Canada, McKinsey found that 45% of consumers are finding it hard to make financial ends meet, while 48% say that economic uncertainty is curbing spending. And most Canadians, a whopping 93%, say that they expect financial difficulties to last over two months. Beyond the border in the United States, a more optimistic outlook abounds: 36% expect the economy to bounce back quickly. Both countries, however, are aligned on one fact: the expectation to increase online shopping for the foreseeable future.

Networks like Facebook are seeing record usage times, but this is counteracted by declining revenues from slashed advertising budgets. The New York Times published a piece, The Virus Changed The Way We Internet, that highlights where eyeballs are going, and how that change has trended: Netflix and Youtube websites are up 16% and 15% respectively, while news sites like CNBC and the NYT have seen a 50% rise from March to April. In simpler times, accelerated traffic means great things for marketers. But in a highly sensitive market with fragile, scared consumers, designing the right campaign, and delivering the right product at the right time to the right people, requires rethinking strategies and typical formulas for success altogether.

Responding to shifting realities: decreased budgets and curbed spending 

An article in MarketingProfs urges marketers to rethink the “4 Ps”: product, place, price, and promotion, as they help their companies get back on track and stay top-of-mind amongst consumers. Specifically, the piece speaks to the need for marketers and advertisers to think critically about how they’ll overcome the “place” interaction and human element, when physical purchasing spaces no longer exist – which is not only key for converting customers, but also for driving brand loyalty. Furthermore, from a pricing and promotion standpoint, marketers and advertisers should think about savvy bundling options to entice consumers, while ensuring the promotions they are offering are thoughtfully adjusted to align with new spending realities.

As certain marketing, media and advertising channels have been eliminated, it’s crucial for leaders to shift their focus to the right channels, while keeping a razor-sharp focus on the realities individuals are confronting on the other side of the screen. While this industry has been one of the quickest to adopt machine learning, in today’s time, AI for marketing can add significant benefits to your adjusted strategies, both today and in the future.

1. Personalize product bundles with savvy insights.

Kepler’s ADSW: Clustering

In Machine Learning, Clustering is a powerful method that allows users to group data into various categories, according to patterns and similarities. With clustering, marketers can create product bundles based off of specific consumer groups by gathering a wealth of consumer data from across the entire customer journey. From historical sales data and past behaviors to location and demographic information, clustering can be an effective way to gather and combine various data sources to generate insights that lead to more personalized offers, and better sales.

2. Give consumers attention when they need it.

Kepler’s ADSW: Tabular Classification and Text Classification

Tabular Classification is an effective Machine Learning method for predicting categories based on key information. For predicting churn, Tabular Classification can be combined with Text Classification to classify customer reviews and emails, for example, in three classes: ‘positive’, ‘negative’ and ‘neutral.’

This classification can then be added to your tabular data (through a column with your “sentiment” classified by Text Classification) and then you can apply your Tabular Classification ADSW to enrich your churn prediction results. Marketers can leverage these churn prediction insights to build and distribute personalized communications and promotions, and run campaigns targeting these customers that speak directly to their pain points, and, most importantly, how they can be solved.

For more information on customer churn, watch this demo video.

3. Accurately predict global trends.

Kepler’s ADSW: Time Series Forecasting

With Time Series Forecasting, marketers can predict future outcomes of specified targets. This is especially useful when uncovering market trends, which marketers can do by taking a look at historical data from past economic events, or by examining how markets in slightly recovered areas may reflect what they can expect in the future. Accurate forecasting can help better understand budget allocation and strategies.

With contributions from Rodrigo Araujo. 

For more on how Kepler can solve your current marketing needs, download our Infosheet.

This site uses cookies to provide the best experience. By continuing to browse the site, you are agreeing to our use of cookies.