How to Use Predictive Analytics in Advertising

By IBM Watson Advertising

Predictive analytics is a form of advanced analytics that utilizes AI, machine learning and historical data to make predictions about future outcomes.

While the concept and practice of predictive analytics has been around for some time now, technology has become increasingly advanced in recent years and organizations have been wise to reap its benefits in regard to improving business performance. Advertisers, in particular, have an opportunity to use the power of predictive analytics to drive performance. By utilizing complex data sets backed up by machine learning, advertisers can create highly personalized advertising campaigns based on the probability of a user taking a particular action.

Types of predictive models and techniques for advertising

There are many different predictive models and techniques that advertisers can leverage for their campaigns. Here are some of the most used.

Clustering models

Clustering models categorize people or items based on specific characteristics or attributes. For advertisers, this can mean looking at geo location or interests to target specific groups.

Forecast models

Forecast models can combine past trends and other data sources to make future predictions. For example, advertisers may be able to leverage weather data, location targeting and past foot traffic to better advertise before storms.

Time-series models

These models use specific periods of time to make predictions. This can be useful for advertisers when past trends may not influence future outcomes. For example, many advertisers needed to change their strategy during the pandemic. Many of these choices were not based on past trends since marketers were dealing with a completely new landscape. 

Neural networks

Neural networks are a type of model that seek to mimic human intelligence to find relationships between data sets. They are often composed of a series of complex algorithms that are inspired by the biological structure of the human brain.

What are the benefits of predictive analytics in advertising?

There are many benefits to predictive analytics that advertisers would be wise to use when developing their marketing strategy. They include the following:

Optimize advertising campaigns

The predictive analytics solution allows users to attract, retain and nurture customers based on historical data and their likelihood of performing a given action. For example, an advertiser can determine the likeliness of a consumer buying rain boots when there’s a 70% chance of rain. Alternatively, advertisers can identify trends in how consumers prepare for severe weather, allergy seasons or a warm day. This information can be combined with contextual and location data to serve the right ad to the right person at the right time without the use of cookies.

Improve team efficiency

Advertisers can streamline operations through predictive analytics by better forecasting resources and costs. They do this by identifying the advertising channels in which increased marketing spend and resources are warranted. The predictive analytics solution makes it easier for advertisers to get their targeting and messaging right the first time, opening up resources for other strategic projects. 

Boost audience engagement

You can gain better insights about your customers regarding their interests and behaviors by using predictive analytics. Advertisers can better understand their audience and what they want so that you can put information in front of them that they are more likely to want to see. By using algorithms, ad placements can be delivered to more-targeted audiences to help make the most of your budget.

Create relevant messaging

By taking a look at data and trends, users can get a sense of what messaging is resonating with audiences and what’s not. Analytics allow you to create a truly personalized experience for your target audience that’s timely, relevant and drives conversions.

Target the right people

The predictive analytics solution provides advertisers with insights regarding the type of audiences, demographics and niche groups you should be targeting in your ad campaigns. In fact, some tools, like the IBM Watson Advertising Predictive Audiences solution analyzes relevant data and scores users on the probability of them taking a particular action. By targeting the right audience the first time, your team can save on resources that would be used in testing the effectiveness of a particular strategy. 

Industries that are using predictive analytics in their campaigns

Many companies that span industries are realizing the benefits of using advanced data to optimize their campaigns. They include the following.


The retail industry can use advanced data to help them with merchandise planning, price optimization and planning events. They can also analyze buying behavior and better determine their ROI. Retailers can promote specific products based on insights from weather forecasting technology and AI. They can also use weather and location data to drive foot-traffic to their stores or encourage users to shop online. 


Predictive analytics can help the healthcare and pharmaceutical industry by improving quality, cost and patient satisfaction. How? By gaining insights regarding who to target, what to offer and how to offer it. There is an opportunity to use weather analytics and target niche groups based on symptoms caused by weather, such as dry eyes or allergies.


The auto industry can use predictive analytics to determine who is most likely to purchase a given vehicle. Advertisers can leverage advanced consumer insights like time since last auto purchase, current mileage, number of repairs on a vehicle, and model of care. All of these details can help advertisers identify the right audience to put in front of their campaigns.  

Marketing and sales

Predictive analytics can be used in any marketing or advertising campaign to improve audience engagement and increase ROI. By uncovering the behavior and patterns of each unique audience, advertisers can be smarter about what they put in their campaigns and help ensure that what they’re putting forth is information the audience truly wants to see. 

Predictive analytics in action

The predictive analytics solution uses a combination of machine learning, statistical modeling, and data mining techniques to make predictions about future outcomes. Organizations can use these models to search through current and historical data and detect patterns, forecasts, trends, events and conditions that might occur. 

Getting started with predictive analytics should involve utilizing predictive analytics tools. At IBM Watson Advertising, we offer the following:

  • Predictive targeting: Through AI, advertisers can better predict who is more receptive to specific messaging to deliver the right ad to the right client at the right time. The ads that get in front of consumers contain more-targeted information, which may lead to more clicks and conversions. 
  • Weather targeting: IBM Watson Advertising Weather Targeting combines the power of the ability of weather to drive emotion and action with IBM’s AI capabilities to model and train algorithms.  For example, a forecast of 50 degrees in one city may not cause the same behavior in another. Rather than relying solely on temperature or other basic factors, each Weather Targeting trigger uses machine learning to improve resonance by recognizing what the weather “feels like” and how consumers in that specific area are likely to react.
  • Accelerator: IBM Watson Advertising Accelerator uses dynamic creative optimization (DCO) for digital display, video and OTT to help exceed your campaign goals and reveal robust creative insights. The tool uses machine learning to understand real-time consumer engagements to predict the best creative for each user. 

Predictive analytics example: Mastercard

Mastercard came to IBM Watson Advertising with the goal to educate consumers about their partnership with “Stand Up to Cancer” and their mission to donate up to USD 4,000,000 to help fund cancer research. 

Mastercard uses IBM Watson Advertising Accelerator to continuously learn which creative elements will resonate with each audience based on not only how consumers react but also on many key elements like DMA, device type, and time of data. They were able to use Accelerator to predict and deliver ads with creative elements most likely to be relevant, engaging and translate into action. With IBM Watson’s help, Mastercard was able to achieve the following results:

  • 81 creative variations
  • 144% lift in CTR from start of campaign 
  • +54 campaign CTR above their benchmark 

Final thoughts

At IBM Watson Advertising we are dedicated to designing cutting-edge solutions that enable users to make the most accurate predictions about their target audience. By utilizing advanced algorithms, historical data, and AI, advertisers can start optimizing their campaigns and achieve their goals the first time around. 

Ready to learn more about the benefits of predictive analytics? Contact us today.

The performance data and client examples cited are presented for illustrative purposes only. Actual performance results may vary depending on specific configurations and operating conditions.