Modern Marketing Deserves Modern Attribution

Give credit where credit’s due with data-backed attribution for our multi-media world.

The State of Attribution Today

Attribution has been a hot topic for years. As advertisers spend more on digital advertising, we also want to know more about how customers interact with it. Is it working? What’s the impact on sales? Is the customer journey what we thought it was? Attribution allows us to analyze the impact of ads, assign credit where credit’s due, document customer purchase journeys, and gain insights on campaigns, strategies, digital tactics, and publishers.

The problem is marketing campaigns are increasingly complex, with data gaps and blind spots that make it difficult to accurately attribute sales. If we want to REALLY know how our marketing is working, we have to track customers across their devices and connect offline customer behaviors to their online activity. And when you throw social media into the mix, things get even more challenging. That’s because social networks act as “walled gardens,” restricting marketers’ access to data. And data is necessary for accurate attribution.

In an attempt to put together the full picture of a customer’s journey, some hands-on attribution rules have become commonplace, including:

• Last Touch: 100% of the credit goes to the last touch.
• First Touch: 100% of the credit goes to the first touch.
• Linear Attribution (Equal Weights): Credit is evenly distributed to touches along the customer’s purchase journey.
• Time-Decay Attribution: Earlier touches receive smaller weights and later touches receive larger weights.
• Other Position-Based Attribution: e.g., a U-shaped attribution rule that assigns 40% of the credit to the first touch, 40% to the last touch, and evenly distributes the remaining 20% to the rest of the touches.

But here’s where those go wrong. The Last- and the First-Touch approaches ignore significant portions of the campaign. And while the last 3 multi-touch rules are slightly better, they can still be inaccurate because they’re predetermined.

And have you heard about algorithmic attribution? Don’t let the fancy name fool you. While it might seem appealing in its complexity, it typically lacks transparency and doesn’t allow any user input or adjustments.

How We Fixed It

With our hybrid attribution modeling approach, Attribution Advisor, we build a more precise, data-backed attribution rooted in a complete, accurate view of each customer’s online and offline journey. With it, we overcome both the inaccuracy of hands-on rules and the mystery of algorithmic solutions.

How Attribution Advisor Works

Leveraging strategic partnerships, we employ a de-identified universal ID that bridges the gap between online (anonymous device) and offline (PII customer) data points to create a comprehensive, secure view of all marketing efforts and outcomes. This enables us to:

• Recognize the same customer across multiple devices and assemble pieces of the same digital journey while maintaining customer privacy
• Connect data collected from different online platforms at the individual level
• Integrate each customer’s exposure to online advertising and offline purchase behavior to capture their complete purchase journey

The Process

We begin with customer journeys. For example, a customer-first sees a display ad on one of his favorite websites, which prompts him to search for the advertised product and then watch a video listed as one of the top search results. A week later he sees a coupon for the product in his social media feed and makes a purchase with that coupon.

From there, we evaluate the entire customer journey without any preset weights, allowing the data to “speak for itself.” We analyze journeys that lead to sales and those that don’t in order to quantify the true probability of conversion for each touchpoint.

Then we consider distinctions among different customer groups. Not all customers are created equal, nor do they share the same purchase journey. For example, customers who are already close to the bottom of the purchase funnel will probably have a shorter journey. By identifying distinct customer groups and developing an attribution model for each, we’re able to reveal precise and actionable insights to guide future digital strategies.

Next, we measure the correlation between different touchpoints and uncover how they interact, how each one depends on the one before it. Without considering these correlations, the attribution model may underestimate the importance of some channels and give misleading insights.

After that, we develop alternative data-driven models to find the most accurate, precise solution.

The Result

With all this, you gain accurate, actionable insights into how your traditional and digital marketing is working, how your customers are interacting with it, and its impact on your bottom line.

Want to know more about how Attribution Advisor can improve your digital strategy? Contact hello@havashelia.com.