In the rapidly evolving advertising landscape, understanding how and where to invest your resources is paramount for guaranteeing maximum return. Here’s where the concept of marketing attribution comes in. Simply put, it’s about quantifying the effectiveness of different marketing strategies across a variety of platforms by crediting the end result (e.g., a sale or a lead) to resource allocation (i.e., spend) on one or more marketing channels. Its ultimate goal? To streamline marketing efforts, cut waste, and boost profitability.
Broadly, the two primary methods are user action-based and top-down type models. Both types can lead to better decision-making.
So Many Models to Choose From
If you run a quick online search for “marketing attribution models,” you will quickly find several flavors. You will see fancy terms like first touch, last touch, linear attribution, etc. Each of these differs in term of how it mathematically counts or attributes specific user actions to desired goals such as a sale. Each of the aforementioned examples are what I call “user action-based” attribution models. For example: a specific user came to our website after clicking on a banner ad served up by our recently-launched digital ad campaign, filled out a web form and later made a purchase. Note that in this case, we are tracking that specific traffic source (i.e., a click on a banner ad at a specific point in time by a specific user) and linking them to a particular sale.
But what if you are unable to directly track how the user came to your website, such as when advertising via billboard or radio? Or, what if you are running multiple advertising channels at once and you are unsure just WHICH one made the actual difference in getting a particular customer sale or hot lead? In these cases, you might need to switch away from user action-based models and instead employ a top-down – or econometric – model.
In terms of these two broad classifications, let’s review each model type in light of what might be best for your business.
User Action-Based Attribution Models
User action-based attribution models, as the name suggests, begin from the very first interaction the customer has with a brand’s assets, such as their website. Examples include the first-click method that assigns all the credit to the initial user engagement point, the last-click method, which recognizes the final touchpoint before a conversion, and the linear model, that apportions equal credit to all touch points in the journey. Besides these, there’s the U-shaped model that gives more weight to the first and last interactions considering them as key pivot points in the customer’s decision-making process. But each of these shares the common trait of linking that user action to a specific result.
Pros:
The granular approach of user action-based models is a game-changer, especially in our ultra-digitized world. It supports specific targeting, aids real-time adjustments of marketing actions, and is highly customer-centric, providing a better understanding of individual customer behavior.
Cons:
However, it’s essential to be aware that user action-based models may inadvertently overlook the interaction effects of marketing campaigns that are composed of multiple advertising channels. Lastly, they may not account for offline marketing activities – those whereby we are not able to automatically link a user action to a desired result in the first place.
Top-Down Attribution Models
On the other hand, top-down – or econometric – attribution models take a holistic view of marketing initiatives. Using statistical techniques such as correlation and regressions analysis, these models measure the impact and interplay of various simultaneous marketing efforts, such as radio ads, print campaigns, or digital ads in driving conversions. They attempt to find causality between these activities and business outcomes.
Pros:
Adopting a top-down model can offer several benefits to your company, like the ability to track multiple channels – both offline and online – and the option to account for external factors such as market trends and competitor actions. They provide a big-picture view, allowing the opportunity to optimize large-scale campaigns. And, quite simply, this type of model may be your only option, since user action-based models may not even be available for the marketing channels you are using.
Cons:
On the downside, top-down models often require starting by revisiting pre-campaign, historical sales data to develop a baseline before advertising started. This approach can be more time-consuming to properly set up a test, measure results and come to a conclusion. These models overlook individual customer journeys and, due to its broad-based approach, they may not be beneficial for micro-targeting strategies whereby your goal is to measure attribution of sales of specific products to specific target audiences.
Choosing between user action-based and top-down attribution models – or adopting a hybrid model – isn’t simplistically about picking the “best” one. It is about understanding what speaks to your business needs. The size of your company, industry, target audience and the complexity of your marketing strategy all play a part. In both cases, you are harnessing and letting it guide your marketing strategy towards success.
How to Choose the Right Type of Model for Your Company
Here are the factors to take into account when choosing the right type of marketing attribution model for your company.
Choose a user action-based attribution model when:
- You are employing marketing methods amenable to algorithmic or automatic conversion tracking, such as digital ads with a well-defined, trackable user action
- You are able to estimate or directly measure the monetary value of a conversion
- You have a limited number of advertising channels running at the same time
Choose a top-down attribution model when:
- You are unable to link specific user actions to specific conversions
- You are running multiple advertising channels at once and want to discern which are making the biggest difference to your sales or leads
- You want to take into account external factors such as seasonality, the economy and competition
You may ultimately choose to employ a combination of these two types of marketing attribution models for best results.
Reach out to Ambient Array to get our help in building an attribution model for your next campaign.