X
Webinar
Quality Data to Fuel Your CPG Campaigns June 20, 2018 Register Now

Back to Basics: What is Look-alike Modeling?

May 6, 2018

What Is Look-alike Modeling?

Marketing professionals, sales teams, and business owners are constantly chasing new audiences to target. This is only natural, as the more you grow your business and the more customers you reach, the greater your success. It is the core element of all business: the need to achieve better and greater support.

But it’s not easy targeting new audiences, especially those beyond what you normally encounter or engage with. What tools can you use? For many, the answer is look-alike modeling.

Look-alike modeling is definitely something every marketer should be an expert on, especially if their goal is to reach greater audiences. So what is look-alike modeling? How does it apply to you?

What is Look-alike Modeling?

Every business needs to continue growing new customers to achieve success, but also to maintain it. But how do you identify those target groups of new people who will enjoy your products and services?

Look-alike modeling is a process that identifies people who look and act just like your target audiences. This tool analyzes your seed audience, identifies some key characteristics and finds users who are similar to your target.

Say you’re hoping to target people who are more likely to click on your ad or watch your video. Look-alike modeling uses machine learning to find more users who will take that action. This means your campaigns can scale to reach more people, with a higher engagement rate. In other words, more bang for your buck!

Why use Look-alike Modeling in your marketing efforts?

Every company has their “ideal” customer. Often, it’s someone who has purchased or shown interest in the company’s products and services. But sometimes it is difficult to find the right person or audience. When you do finally find them, you want to hold on to them. You want to nurture and maintain a relationship for as long as possible, to retain and upsell your customers and keep their loyalty.

Not only do you want to hold onto your audience, but you also want to find people who think or act like them. This grows your reach so that you have a much larger possible group of customers, which is a recipe for success.

Look-alike modeling does just that. It helps a marketer identify a larger pool of possible customers. You use the tool to seek audiences with behaviors that match up with your target audience, and so you have a greater chance to convert them. It’s essentially the same as building a robust profile of existing customers, only you’re doing it for the audiences you have yet to reach out and engage with.

Customers Who Will Increase Engagement

How does Look-alike Modeling work?

All this sounds great, but up to this point, the benefits are nothing more than empty promises or claims. How does look-alike modeling actually work? How do you know using one of these tools for your own business is going to return positive results?

How do you build a Look-alike Model?

Let’s say you are a shoe brand looking to boost online purchases for an upcoming sale. The first step would be to place a pixel — a small segment of tracking code — on your purchase confirmation page. This will allow you to track the behaviors of purchasers — during the current sale — as they move throughout the web.

The demographic and behavioral data points of anyone who completes a sale and makes it to that purchase confirmation page can then be ingested into a centralized platform like a DMP. Once the data is collected, it can be analyzed to identify which behaviors and patterns are most common among people who buy your shoes on sale. Machine learning and AI tools do this in an accurate, verifiable way using pre-configured algorithms.

Once the customer characteristics are identified, you can use them to engage with an even greater target audience. It’s a process that uncovers online users who act the same as your existing customers but may not know your shoe brand. In turn, you can target these new users in your future marketing campaigns!

Gain an Even Greater Following

Not only are you finding more people to target for your upcoming sale, but the machine learning tool has done the hard part for you – and identified exactly those people most likely to complete the purchase.

What do I need to get started?

The first step in building a lookalike model is understanding your goal: who are you trying to reach, and what are you trying to get them to do? You must start with your first-party data, using both offline and online stats to create an initial seed segment.

Initial Seed Segment

In the case of the shoe store mentioned above, the target audience was anyone who had previously purchased shoes. They used the data on those people who had completed the purchase to find more people who were most likely to complete a purchase as well.

Once this “seed” data is fed into the lookalike modeling tool, the machine learning will process it and start to look for new potential customers who share some characteristics with them. Finally, the modeled audience will be ready for your targeted campaigns.

Where Does a DMP Come Into Play?

Data Management Platforms are a unifying technology that aggregates your data from any source, including first, second and third-party data from online, offline, mobile, and beyond. By connecting all of these disparate data sources, you can get a clearer picture of your target audience overall, across screens and devices.

Once you have identified your seed audience, the DMP analyzes it closely, applies proprietary algorithms and techniques and finds more user profiles that match. Lotame’s Audience Optimizer is the perfect representation of a lookalike tool that a DMP uses to help build intelligent look-alike campaigns.

Leverage Data

Is Look-Alike Modeling a One Time Thing?

In short, no. Lookalike modeling can be used continuously. It’s not something you do once and walk away with, using as a reference point forever. The information and insights you collect must be maintained over time, which means refreshing your models regularly.

In fact, the longer the models run, the more they “learn” about your audience, so they can continue to adjust the new lookalike audience for accuracy.

To summarize, look-alike modeling takes your existing audiences and their behaviors and expands it based on their behavioral and demographic characteristics. It offers marketers or publishers a valuable approach to reaching new or current consumers in a cost-effective way, to help grow the scope and reach of your business.

I’m All-In for Look-Alike Modeling. Where Do I Go Next?

All that sounds fantastic, and you’re ready to proceed. What next?

If you haven’t already, start with a reliable, effective DMP provider. That’s your initial gateway to all of the additional data systems and tools. Before choosing a provider, however, consider your future relationship. What alternative tools and platforms do they offer their customers? Do they have the resources to help you grow?

Lotame’s Audience Optimizer is a great example of additional features and tools that you should look for when choosing a provider.

Request A Demo