Attribution Model Challenges for Marketers

How do you measure the success of your marketing efforts? Are you struggling to design attribution models that give you meaningful insights into the customer journey? Attribution models, flawed as they are, have become popular among marketers struggling to show ROI on their efforts. The challenge for marketers and brands is that most attribution models don’t give an accurate picture of the customer journey and fail to provide the kinds of insights that inform strategic decisions.

While simple to understand and measure, first touch and last touch attribution models only provide a glimpse into the customer journey. Searching for ways to show how marketing efforts are bringing new audience members to the website, marketers have employed first touch methods. Similarly, last touch attribution models became a popular way to show which marketing efforts contribute directly to sales.

The fundamental flaw in each of these approaches is their blindness to which touch points are helping customers out along the way and to the other factors influencing the buying decision. Enter multi-touch attribution models. While these types of models take a more holistic approach, they often fail to provide real, actionable data that informs decisions and improves ROI on marketing efforts.

In a recent episode of Get Found! Erin Robbins, President at GinzaMetrics, and Steve Farnsworth, CMO at Steveology group, discussed the inherent problems with attribution models and some of the trends that are emerging to help marketers show ROI on their efforts.

Beyond single touch attribution

When talking about attribution models, most CMOs would agree that the simple first touch and last touch methods are fundamentally flawed. In a recent discussion with a group of CMOs, Steve found that most people are moving away from single touch attribution and trying to fill in the gaps of the customer journey with a few variations of the traditional multi-touch attribution models.

In most organizations, attribution modeling is taking on one of these forms:

  1. Multi-touch attribution
  2. Decay multi-touch attribution
  3. Cohort attribution
  4. Position-based attribution

Steve mentions that there are some issues regarding how each interaction is weighed with multi-touch models and says that CMOs are trying to address those issues with a model called decay multi-touch attribution. Decay models address common multi-touch challenges by weighting the latest stage activity the heaviest.

Another way to look at the mix is with cohort attribution. Cohort attribution allows you to look at how many people are retained across measurements to track leads month-to-month. By watching how a lead moves through the journey, the amount of touch points it takes to bring them to conversion, and the amount of time required to convert a lead; marketers can calculate the cost of a closed lead and determine the ROI on marketing efforts.

According to Steve, “Cost per closed lead is huge: how much can you spend and still make money? I think that’s kind of where attribution models are. Cohort, multi-touch attribution and time decay is what everyone has to be doing. They may not be doing it yet but I think that’s fundamentally the state of the art.”

He also mentioned position-based attribution, a model that’s a combination of the time decay and multi-touch models. In this model, you assign percentages such as 40 percent to first touch and 20 percent to last touch and then divide the remaining 40 percent up among all the touches in between. The challenge, according to Steve, is that you may end giving too much credit to the first touch. He warns that the solution to a too heavily weighted first touch isn’t necessarily to split the attribution equally between first and last touch. Marketers need to consider the weight a first touch attribution needs in order to make this model effective.

In Occam’s Razor by Avinash Kaushik, he breaks attribution models into three categories:

  1. Online to store – for example: online campaigns that drive in-store purchases.
  2. Multiple screens – for example: TV, desktop, tablet, and smartphone.
  3. Digital channels – for example: social, YouTube, referral, email, and search.

“Most digital marketers are looking at the third model,” according to Erin. “One of the challenges of attribution modeling is how to assign real credit to each channel. The folks I’ve talked to say that everyone who contributes along the conversion path gets credit, which means your total sum credit for a single customer or single cohort equals more than 100% because a lot of people are taking credit for a lot of percentages of conversions.”

Marketers can avoid assigning more than 100 percent to any one conversion by adjusting the percentage of the value that’s assigned to each touch point based on the total number of touch points in any one conversion path. The downside to assigning equal value to all touch points is the inability to determine which touch points have the greatest impact.

Attribution models and the individual agenda

The problems and challenges associated with attribution models is further exasperated by organizations where departments or functions are competing for credit for their efforts on a particular channel.

According to Erin, “I think one of the reasons we’re having such problems measuring with attribution models is that instead of working together to create winning accounts, we’re all concerned with who did the best job and who will get the budget to continue their efforts based on attribution percentages.”

She suggests viewing creating conversions and winning new business as an ecosystem where everyone relies on the efforts of everyone else and realizes that every touch point has led to the final conversion. In this ecosystem, taking any one step out of the process could change the outcome, even if that step is only assigned 12 percent of the overall score. Brands trying to elevate the level of effort from each department have used attribution modeling as a way to create organizational competition and determine budgets and recognition. Using attribution modeling to assign blame or credit results in conflicts over how attribution is assigned and limits the benefits of using measurement to elevate the entire process.

Instead of deciding which department or channel effort has the most value, the method should be used to understand how things are working within the marketing ecosystem. Use it to discover which types of conversions are taking place for certain audiences and then maximizing efforts based on the mix of assets and channels that lead to conversions and the types of target audiences you have. Attribution models can help inform decisions and determine how to deliver the right message on the right channel at the right time.

“The same thing is true within the organization. Instead of using attribution models to assign credit to marketing or sales and award more budget or personal reward, use the findings to maximize the roles and efforts of everyone in the organization to close as many customers as possible and keep them as long as you can,” recommends Erin.

Use measurement to turn up the dial

The number one priority with any measurement model is to map out the customer journey to discover answers to questions such as:

  • Where are leads are coming from?
  • What does it cost to lead a prospect through to conversion?
  • What touch points make up their journey?
  • What happens to leads that don’t close?

According to Steve, “When you focus on those questions, you can start identifying the biggest thing and then you can turn that dial up. That’s what you should be focused on, turning the biggest dials up and then figuring out which of those channels are working best and working together.”

When using attribution models, marketers have to accept that there are flaws in the models that don’t account for other influences beyond the digital footprint people leave when researching and looking for your solution. Instead of using attribution models to assign credit, use them to find solutions, fill gaps in engagement, and amplify the messages and channels that are gaining the most conversions.

If you’re looking for a way to measure engagement, optimize content, and beat the competition, GinzaMetrics has a scalable platform to meet your needs now and in the future. Give us a shout and we’ll show you around.

Watch the entire episode of Get Found with Steve Farnsworth and Erin Robbins.

Categories: Marketing Strategy.
About Karen Scates