How to Build a Custom Marketing Attribution Model

by The MNTN Team

6 Min Read

Marketing attribution models play a vital role in planning your marketing strategy. Using these models helps you determine which of your digital marketing channels are most responsible for each sale. There are plenty of ways to approach attribution, including first-touch, last-touch, or multi-touch models, and each has its own benefits and drawbacks. But you’ll see the most accurate picture for your business with a custom attribution model.

What Is Custom Attribution?

A custom attribution model is tailored for your business instead of relying on a standard attribution model, such as first-touch attribution, which assigns all credit for a conversion to the first time a customer engages with your ads; last-touch, which gives all the credit to whichever ad directly led to a conversion; or multi-touch attribution strategies, which consider the whole buyer journey (but the weight given to each touchpoint might not be appropriate for your business).

A custom model considers your customers and their specific journey(s). With a custom attribution model, you could assign credit for sales to various touchpoints on the customer journey based on how your specific customers interact with your brand.

Advantages of the Custom Attribution Model

Perhaps the biggest advantage of a custom attribution model is that it’s specific to your business. Instead of relying on a generalized model and using the data to allocate your marketing time and budget, you can use customer-specific insights instead.

This type of attribution model helps you understand how customers at various points of the buyer journey interact with your brand. That way, you can focus your marketing efforts on the digital marketing channels that offer you the most return on investment, or you can develop strategies for improving your least successful channels.

For example, you may find that most of your customers respond well to your conversion campaigns, but your brand awareness efforts aren’t generating as much traffic. With this data, you could look into different ways to build brand awareness.

Disadvantages of the Custom Attribution Model

Because a custom attribution model requires a lot of data to get it right, it can be complex and time-consuming. You need access to historical customer data, and you need to be able to map a customer’s journey with your brand.

If you’re using Connected TV to build awareness, for example, you would have to match the email associated with a user’s streaming platform account to a new signup for your email newsletter. Then, you would have to follow that customer through each email campaign to see if they click through on any parts of your email or redeem any special offers. Finally, you would have to track the email address through to the sale. With privacy laws limiting certain types of data collection, this information can be hard to track.

Since custom attribution models are so complex, it’s easy to misinterpret the data and end up crediting your ad campaigns incorrectly.

Before You Begin: Understand Your Business Goals

If you are interested in a custom attribution model, set clear goals first, as these will help you determine what data to collect and how to allocate credit for your ad campaigns. Your goals will also help you decide which metrics to measure as you evaluate each ad campaign.

Step 1: Capture Accurate Data

Once you have your goals set and your marketing metrics established, you can start collecting data. Map the customer journey to determine how many touchpoints your average customer has with your brand before they make a purchase.

Make a list of your hot leads to identify trends on how these customers typically find your brand. Additionally, use a client relationship management tool that integrates with multiple touchpoints, including email, social media, and website.

Step 2: Compile Data Together

After you have your data, compile it into a single source. You may need a platform to aggregate your data and help you visualize certain trends. If you have a tool that generates reports, use charts and other data visualization features to help you identify patterns and consistencies among your best customers.

This will give you a clearer picture of how your best customers interact with you and how successful your digital marketing efforts are in driving traffic and conversions.

Step 3: Determine Your Attribution Window

Even if a customer interacts with your brand multiple times, each touchpoint may not contribute to a sale. If a person doesn’t buy anything from you within a month or so after seeing your ad, you can assume it didn’t inspire them to convert.

If they see another ad two weeks after your first one and then click on your website, it’s likely that they remembered the first ad, but the second inspired them to click on your website. Use your average customer journey and their buying process to determine your attribution window.

Some businesses have longer buying cycles. For example, if you sell big products like cars or appliances, you might set your attribution window at 90 days. And if you’re selling clothing or accessories, you might set it at 14 days.

Best Marketing Attribution Models

Custom attribution can be overwhelming. If you’re just getting started, consider these other, more common attribution models:

First-Touch Attribution: assigns all credit to the customer’s first touchpoint
Last-Touch Attribution: assigns all credit to a customer’s last touchpoint before purchase
Linear Attribution: assigns equal credit across touchpoints
Lead Conversion Touch Attribution: assigns 100% credit to the touchpoint that turned the prospect into a qualifiable lead
Time-Decay Attribution: assigns credit to touchpoints based on how close they were to the actual sale
Position-Based Attribution Model: assigns most credit to the first and last touchpoints, then distributes the rest equally throughout each touchpoint in the middle
W-Shaped Attribution: distributes credit equally to the first, middle, and last touchpoint, then evenly distributes the remainder across other touchpoints

Best Practices for Custom Attribution Models

When building a custom attribution model, it is more accurate if you capture multiple pieces of data. You should collect data from various tools, including Google Analytics, your CRM platform, your social media pages, and other sources from which you interact with customers.

Once you get started, keep analyzing customer behavior patterns to adjust your attribution model and make it more accurate. Finally, stay flexible. Your customer journey may change as your industry shifts and adapts to new trends. Assess and update your model from time to time.

The Rise of TV Attribution

With more and more audiences flocking to Connected TV to watch their favorite shows, advertising on these streaming platforms gives brands a captive and precise audience. Working with a CTV advertising partner like MNTN can offer more refined attribution models like ours, which we call Verified Visits™.

With Verified Visits^TM, marketers are able to track their CTV ad’s performance as accurately as they can with social media and other digital channels. This solution tracks viewers who see your MNTN ad on CTV and then click on your website within a defined window — all on a device in their household. Verified Visits^TM gives you a better idea of how your CTV ads are doing so you can optimize your strategy based on real data and easily build touchpoints into your custom attribution model.

Final Thoughts: Custom Attribution

Although custom attribution gives you the best picture of your customers, it can be a complex process. Unless you have a large marketing team that includes data analysts, you may want to use a traditional attribution model or hire someone to come in and create a custom model for you (note: if you do the latter, make sure you know how to update it as your customer base and behavior evolves). But if you’re more interested in accurately allocating your marketing time and budget, a custom model is a safe bet.