Data Modeling as Explained Through Cookie Sales

Data modeling is a hot commodity for marketers – and the emergence of AI only further exemplifies how much value companies place on their audience targeting data.

But with many popular marketing buzzwords, not every person outside the industry truly understands what the process may look like. 

To best demonstrate this methodology, let’s consider a situation many Americans are familiar with – Door-to-door public school fundraisers.

Picture this scenario: Student Henry Jones is helping his school during its annual fundraising cookie drive. This fundraiser offers prizes for the top students who sell the most cookies by the end of the week. 

Henry, motivated to contribute to a good cause and compete against his peers, sets out to hit record sales within his neighborhood. 

With so many other students also competing for the top prize, Henry knew he had to strategize. He could only walk around his neighborhood in the afternoon before it got too dark – so he needed to make the most of his limited daylight hours. 

 After some brainstorming, he devised two potential tactics to sell the most cookies within the week for his fundraiser:

Related: What is Data Modeling in Marketing Campaigns?

PLAN A: Visit Every House Within The Neighborhood

Henry’s first strategy would target every house within his neighborhood’s vicinity. In theory, this means Henry would win the school fundraiser, since he combed through his neighborhood to find every potential customer. 

Upon further speculation, Henry quickly realized that reaching every house within such a short turnaround time was extremely unlikely. For one, his neighborhood was very large, so he would only be able to visit a couple of streets every day before nighttime. 

Not only that, but if Henry wasted too much time visiting houses that were not interested in his fundraiser, he would overlook other houses that were more likely to contribute – giving other students within the neighborhood the chance to reach them first. 

Related: Trimming the Fat | The TargetList Direct Mail Advantage

PLAN B: Using Data Modeling For Henry’s Cookie Campaign

Growing up, Henry became well acquainted with the people that resided within his neighborhood. He knew a fair bit about the families, the children, and each household’s typical lifestyle. The boy wondered if he could use this knowledge to predict which neighbors would be most likely to contribute to his fundraiser.  

Henry understood that his ideal audience were people that liked cookies and would be motivated to buy them from his fundraiser. He also acknowledged that they needed to be available during the afternoon when Henry walked around his neighborhood. 

By understanding who his target audience is, Henry thought about his neighbors and considered what made them unique from one another. To keep his strategy simple, he narrowed down these descriptions into three distinct categories: Personal tastes, responsiveness, and household characteristics. 

1.) Behavior & Personal Tastes

Henry knew that some of his neighbors are very health conscious. He remembered riding bikes around the neighborhood with some of their kids and he knew which families liked to walk outside during nice weather.

These households, he theorized, would be less inclined to buy cookies based on their active lifestyle. 

Furthermore, he was aware that several students from his school also lived within his neighborhood. While Henry theorized that those households may enjoy cookies, it was extremely unlikely these families would buy cookies from anyone besides their own children. 

Based off this information, Henry chose not to visit these households during his door-to-door sales route. 

2.) Responsiveness

Even if a neighbor wanted to buy cookies, there are still many variables that could determine if they would become an actual customer.

Since Henry’s campaign route would be limited to only a couple of daylight hours, he would be forced to prioritize the houses that are most likely and most eager to buy cookies when he knocks on their door. 

Due to later work schedules, some neighbors may not be home at the time Henry visited their house. Even if these households wanted to buy cookies, their lack of availability means they would never convert if they are unavailable when Henry visits. 

Because he had to prioritize the ideal houses that like cookies AND were more likely to respond at their door, Henry removed these addresses from his consideration.  

3.) Household Characteristics

Once Henry removed the prior households from his list of potential customers, he was then left with a list of families most likely to buy from his fundraiser.

However, if Henry wanted to sell the MOST cookies, he needed to consider which houses were more likely to buy more than one box – before his competition had a chance to get to them first. 

Henry knew that the households with larger families were likely to buy more cookies than single households or smaller families. He also suspected that previous student alumni may have a stronger incentive to contribute to their old school. 

After categorizing the types of households within his neighborhood, Henry then created a ranked list of houses that liked cookies, were likely to engage with his campaign, and were likely to buy a lot of cookies.  

Data Modeling in Practice

Henry understood that his best customers enjoyed cookies, were available to purchase cookies, and would be motivated to buy a lot of boxes. 

As he made his way through his neighborhood, Henry prioritized his ideal audiences first – before other students had a chance to sell to them. After visiting these houses, Henry decided to change his approach to appeal to other types of audiences. 

Health-conscious households were less likely to buy cookies for themselves, so Henry chose to highlight the benefits of contributing to his school fundraiser instead.

He also explained that the cookies were made by a premium brand, making them an excellent gift for others while also supporting his school.

This personalized messaging helped Henry appeal to these households that may have initially dismissed his fundraising efforts. 

Finally, because Henry prioritized his ideal customers first, he had extra time at the end of the week to explore other neighborhoods within his area to search for additional sales. 

Data Modeling For Marketing

Data modeling is about understanding your best customer and using their unique audience profile to target similar households within a specified location. 

This audience profile can be made up of an audiences’ characteristics, buying preferences, personal tastes, responsiveness, as well as anything else that differentiates them as customers from the general public.

Just as Henry used each households’ unique set of characteristics to determine who to target, data modeling can help reveal which company prospects would make ideal customers. 

Instead of targeting an entire haystack in the hopes of finding a few needles, what if you could target only the needles? 

Data modeling is not about finding every potential prospect within a location, but rather prioritizing ideal prospects most likely to convert.

This means maximizing profit margins by leveraging audience targeting metrics to find customers without burning through the entire marketing budget to find them. 

That way, your campaigns improve ROI with smaller, targeted campaigns WITHOUT sacrificing sales goals.