The Hidden Secrets To Improving Attribution Models
Attribution is one of the most heavily discussed and debated topics among digital marketers. Interestingly, a recent study conducted by eConsultancy found that only 54 percent of companies are yet using any form of marketing attribution. The reasons for this vary, and they range from staffing to cost to time. Add to that the complexities of […]
Attribution is one of the most heavily discussed and debated topics among digital marketers. Interestingly, a recent study conducted by eConsultancy found that only 54 percent of companies are yet using any form of marketing attribution.
The reasons for this vary, and they range from staffing to cost to time. Add to that the complexities of attribution, which have multiplied with the growing number of customer touch points. Today’s customer journey now spans multiple devices and marketing channels as well as the online-to-offline experience.
That’s not to say that we don’t have plenty of sophisticated models and platforms available to help with attribution.
At one end of the spectrum, there are a slew of new technology tools offering advanced attribution modeling. Those systems offer a logical approach to better understanding attribution, and there’s no denying their value. But for a lot of companies, they’re not a feasible approach to analyzing the results of every campaign because they tend to be time consuming and costly. For many companies, these systems may be well beyond their budget.
At the other end of the spectrum is the path of least resistance, i.e., last-click attribution. While we’d all like to think we’ve evolved beyond the days of last-click, the reality is that it’s still the most common attribution model used (pdf).
If you’re like a lot of marketers today, you’re likely to fall somewhere in between those two extremes. You want to apply more sophisticated attribution models to better understand your audience, yet you don’t have a wealth of resources at your disposal.
While there is no lack of expert opinions when it comes to the topic of attribution, following are some tangible actions you can put in place today that don’t require additional resources. More specifically, below you’ll find:
- Three critical, yet often overlooked attribution models that will ideally help you refine your approach
- Three ways attribution goes awry so that you can avoid missteps
3 Critical, Yet Often Overlooked Attribution Models
When it comes to attribution, we’re inclined to focus on the specific drivers that impact the bottom line. While we can all agree that there are many channels that contribute to a sale, it’s easy to overlook those channels that weren’t on the direct path to conversion. With this in mind, here are three non-linear dimensions that are worth a closer look as you evolve your attribution models.
Contributions Versus Conversions
Just because you can’t track a channel with a hard dollar metric doesn’t mean you can’t account for it. For this reason, don’t leave out the channels that contributed to the sale.
In fairness, to create a balanced contribution versus attribution model, divide your channels into “contributions” and “conversions,” so that you can see the impact of each effort throughout the customer journey. This also allows you to provide your vendors with a clear understanding of how they will be rewarded for their efforts.
It’s easy to get into the habit of only looking at conversions. However, this can lead to a myopic view of your performance because it doesn’t take brand awareness into account — or the reality that some campaigns may have reached your audience; yet, for one reason or another, they didn’t resonate.
For this reason, you want to include engagement as a critical attribution dimension. In this context, “engagement” refers to a consumer’s interaction with an ad, video or image by hovering, clicking or otherwise interacting.
Along with proving your ad was seen and having a way to determine how consumers responded to it, engagement is also a more compelling alternative to simply measuring clicks.
Complementary, Overlapping Channels
In the ideal scenario, all of your channels are working together; yet, when you have so many customer touch points, the overlaps may cause you to rethink how you’re allocating your budget.
For example, let’s imagine your campaign results show that affiliate and display drove the same customers to conversion. Based on this information, you may come to the conclusion that you can scale back one of those channels. However, what that bottom line data is not telling you is that the affiliate and display campaigns were a dynamic, one-two punch.
Before you abandon one channel in favor of another, you want to go deeper into your analytics to develop an attribution model that reflects which channels complemented each other versus those that cancelled each other out.
3 Ways Attribution Goes Awry
In determining attribution, some marketers may elect to use the same attribution model for every campaign. That’s the first reason for attribution going awry.
Using The Same Attribution Models For Each Campaign
Most marketers establish an attribution model based on the type of products they sell, their cost, and the length of the sales cycle. This often leads them to adopt attribution models such as first-touch, last -touch, or position-based, which prioritizes first and last interactions over actions that occur in the middle.
While these models are relevant and have likely been customized based on the needs of your business, using the same model for each campaign can limit your ability to get a more comprehensive picture of your target audience. If you want to get a better understanding of your audience, consider applying a different attribution strategy to your next campaign.
For example, if you’ve always used the last-touch attribution model and are coming out with a new product, you may want to consider emphasizing the first touch. The reason being that a primary driver for the campaign may be awareness for the new product. In this scenario, by assigning the first touch attribution model, you are essentially giving more credit to those interactions that support the primary strategy.
Underestimating The Importance Of Performance
Of course, attribution is all about performance and the pay-for-performance model is one of the most effective and reliable ways to measure your success. However, let’s not lose sight of how the campaign goals were achieved.
For example, if you’ve ever hired a contractor that billed by the hour, you know how important efficiency and expertise are to getting the job done. The same applies to understanding your campaign performance and properly acknowledging how the results were achieved.
Prioritizing Payment Over ROI
When you’re determining how you’ll allocate your budget across the various marketing channels, consider the importance of the ROI from each channel as opposed to just looking at the cost.
As my Rakuten MediaForge colleague Danny Kourianos pointed out, when it comes to attribution, focus on ROI first and the payments later. This speaks to the industry’s call for a more systematic and weighted approach to attribution. It only makes sense that the ROI comes before the payment.
The attribution conversation is far from over, and we’ll all continue to refine our approaches as we move further away from the last-click model.