Use Marketing Attribution To Optimize Your Ad Budget
Every purchase is the result of a consumer journey. That journey begins with a desire—I want a new hat to make me look cool in front of my friends!—and ends with an exchange that benefits your business—if you give me that hat, I’ll give you more money than it cost to make and market!
While that journey used to be somewhat linear, the path of today’s consumer is longer and more convoluted—one study of online buying patterns included purchasing journeys of between 20 and 500 touchpoints.
While 500 may be on the high end for most businesses, the truth is that consumers are exposed to more and more messages in general, and utilize more channels to access those messages than ever before. As a marketing manager or small business owner, it’s your job to manage those channels and touchpoints, and to choose where to funnel your ad budget. But with such a dizzying array of options, that task can be extremely daunting. How do you know the best way to distribute your resources?
This is where touch-based marketing attribution comes into play. Attribution allows you to trace the consumer journey to evaluate the performance of your marketing efforts. It helps you better understand how your efforts impact the entire sales cycle, and how your various channels and touchpoints work together to produce optimal results. By measuring and evaluating the performance of your marketing channels on a user level, you can distribute resources in a way that drives the most conversions at the lowest cost to your business.
It’s a critical component to building successful marketing campaigns—according to Google Think, 72% of marketers agree it leads to better budget allocation.
What is marketing attribution?
Marketing attribution is the process of determining which of your marketing efforts are responsible for causing consumers to take actions that create value for your business.
Touch-based attribution modelling
Touch-based attribution modelling uses individual data gathered from specific users to perform attribution analysis. As mentioned above, each purchase is the result of a journey. By following the consumer along this purchasing path, analyzing how they interact with the channels and touchpoints they encounter along the way, and assigning credit to each based on its ability to impact conversions, you can decide the most profitable way to distribute your marketing resources. Touch-based attribution is particularly useful for small and medium businesses that have significant online marketing efforts.
When thinking about how to implement marketing attribution, it’s helpful to keep some basic vocabulary to keep in mind:
Marketing channels are the media through which you deliver messages. Search advertising, television commercials, social media platforms, various print publications—these are all types of marketing channels. Depending on the type of analysis you’re doing, channels can be looked at in the aggregate (i.e., you could analyze the performance of all social media channels at once), or on a one-by-one basis (i.e., you could focus on the performance of Facebook ads only). Both would be considered channel-level analysis.
Marketing touchpoints are marketing tactics—materials and experiences that engage consumers and communicate messages about your brand. Examples include pretty much any type of marketing effort you can think of: a billboard on the side of the interstate, a TV commercial during the big game, a text ad at the top of your search results on Google, a sponsored instagram post featuring your favorite celebrity. All of these would count as a touchpoint.
Conversions refer to any action performed by the consumer that increases their lifetime value (LTV) to your business. In many instances, the desired action is making a purchase, which increases the value of the consumer to your business by directly contributing to revenue.
In some cases, the desired action is something different; for a social media company, for instance, a conversion may occur when a consumer signs up for a free account. There’s no immediate revenue associated with this action, but it allows the company to show ads to that consumer down the line. Since ad sales are how many social media companies generate revenue, this cashless conversion still increases LTV.
Types of marketing attribution models
Marketing attribution models are the systems with which marketers evaluate the performance of a marketing effort in order to determine its impact in generating conversions. For touch-based attribution, this analysis takes place at the user level, by looking at the purchasing journeys of specific, individual consumers. These journeys are tracked and evaluated to draw conclusions about which touchpoints influenced their decisions to convert.
These models are typically applied to digital marketing campaigns, because online campaigns can utilize digital trackers, such as cookies, to track the entire journey of each consumer. Cookies are small pieces of data sent by a website or advertisement when a user interacts with it, and stored on the user’s computer. The first time a user interacts with one of your touchpoints, they accept your cookie. When they encounter each subsequent touchpoint, the cookie will tell you about all of their prior encounters. Cookies and other online trackers form the basis of online marketing attribution.
Touch-based marketing attribution models are often divided into two categories: single-touch and multi-touch. They provide the starting points for your marketing attribution program.
Single-touch attribution models
Single-touch attribution models look at just one touchpoint in a consumer’s purchasing journey, and assign all the credit to that touchpoint if the consumer converts. Here are two examples:
1. First-touch model
The first-touch model gives all the credit for a conversion to the very first touchpoint a consumer encounters on their purchasing journey. For instance, if a consumer first encounters your brand by clicking on a paid search ad after searching for a keyword you’ve won at auction, that touchpoint gets all the credit for that consumer’s conversion—even if the consumer interacts with more of your touchpoints before converting.
2. Last-touch model
The last-touch model, as you might imagine, gives all the credit to the final touchpoint a consumer encounters on the purchasing journey. If a consumer first encounters your marketing efforts by clicking on a search ad, then opens one of your promotional emails, then watches one of your ads before a Youtube video, and converts by clicking on the Youtube ad and purchasing something from your website, only the Youtube ad gets credit for the conversion in the last-touch model.
Multi-touch attribution models
Multi-touch attribution models look at multiple touchpoints in a consumer’s purchasing journey, and if the customer converts, assign credit for that conversion across more than one touchpoint. Depending on which model you employ, touchpoints are weighted as being more or less responsible for a conversion.
The linear attribution model weighs every tracked touchpoint along the purchasing path equally.
If a consumer first encounters your brand by clicking on a link in a sponsored tweet, then several days later watches an ad before a youtube video as part of your remarketing program, then two weeks after that clicks on a digital coupon in one of your marketing emails and uses it to make a purchase—each of those touchpoints are valued as having the same impact on the conversion and are therefore funded equally during budget allocation.
2. Time decay attribution model
The time decay attribution model values recency in determining which touchpoints most impacted a conversion. Starting with the conversion and working backwards, each touchpoint is assigned less and less credit for causing the conversion. Using the example above, the email marketing piece would receive the most credit, the youtube ad would receive second most, and the sponsored tweet would receive the least.
3. Position based attribution model
Position based attribution models assign weight to touchpoints based on the stage of the purchasing journey in which they are encountered by the consumer.
4. U-shaped attribution model
One type, called the U-shaped model, values the first and last touchpoints as being most important—they each get 40% of the credit. The remaining 20% of the credit is distributed evenly among the all of the touchpoints in between.
5. W-shaped attribution model
In another, called the W-shaped version, the first and last touchpoints remain highest valued, along with a third touchpoint that takes place somewhere in the middle of the journey. This middle step is called the opportunity stage. It occurs after a consumer’s first encounter with your brand but before a conversion decision is made: it is seen as having a large impact on convincing the consumer to seek out the final touchpoint that leads to the conversion. In the W-shaped model, the first, middle, and final touchpoints are each assigned 30% of the credit, while the remaining 10% of credit is assigned to each of the points in between.
Using attribution data to make strategic optimizations in digital campaigns
1. Customize your attribution model
Keep in mind: Each of the attribution models described above should serve as a starting point for tracking and measuring the performance of your touchpoints. In practice, most businesses end up using a custom attribution model comprised of variations on one or more of the models described above. Every campaign and touchpoint is unique, and it takes time and experimentation to figure out the most effective way to allocate budget resources.
By adjusting these established models, recording the results, and adjusting allocation accordingly, you can find a way to distribute credit and funding among touchpoints that maximizes ROI.
2. Follow a framework that matches your strategy and goals
Follow a framework of best practices to allocate your ad budget. Thoughtful evaluation, ROI tracking, and accurate data are important keys to taking the guesswork out of allocating your marketing budget.
3. Understand your sales cycle
Interview your customers and sales team and develop a clear idea of what the average customer experiences in their purchasing cycle. This helps you develop a rough outline upon which you can build your attribution model. Use information directly from customer interviews regarding the importance of your touchpoints in their decision making, feed that into the way your model assigns credit to touchpoints and channels, and adjust marketing spend accordingly.
4. Use experiments to inform your attribution modeling
Apply various attribution models and weighting schemes, adjust your ad spend, and monitor the results. Try experimental methods such as switchback and synthetic control. In the switchback method, you turn off certain marketing efforts for a period of time, and compare your performance metrics, such as sales and conversions, to a period when those efforts were active.
With synthetic control, instead of time periods, you’re comparing two similar audience groups. Turn the marketing off on one, then compare to the other and note the differences in performance.
Doing these types of experiments will help you determine how to weight touchpoints in your attribution models.
5. Account for missing touchpoints
Touch-based attribution suffers from one significant flaw: difficulty in collecting data on offline touchpoints.
To produce the fullest possible picture of each consumer’s purchasing journey, you need to collect a ton of data. This includes a step-by-step view of the purchasing journey. Your view of each and every buying path should be as complete as possible, as each missing puzzle piece detracts from the accuracy of your analysis when it comes to assigning relative values to each component.
Since cookies can’t be passed to a customer in the real world when they see a television ad or visit a brick-and-mortar store, attribution reporting systems often fail to account for their influence.
While it’s impossible to collect every touchpoint along every consumer journey, do your best to incorporate offline interactions. Use tools to track mobile usage or in-store visits.
Phone calls, in particular, are a vital touchpoint for most businesses, and can be easily incorporated by using of the right tools.
Ultimately, marketing attribution requires constant tinkering over multiple periods to perfect. By lining up your organizational goals and marketing strategy, understanding your sales cycle, experimenting with various models and capturing as much data as you can—online and off—you’ll be armed with the information you need to optimize your marketing spend.