Why Multi-Touch Attribution Falls Short: What You Need to Know
If you’re in charge of driving revenue, you’ve likely encountered the concept of multi-touch attribution. At first glance, MTA seems like a powerful solution for understanding how marketing efforts influence customer journeys. But there’s a catch: MTA has limitations that can obscure the real impact of your marketing strategies, leading to poor investment decisions and potential misalignment with your revenue goals.
Let’s break down why MTA often fails to live up to its promise and what you can do to address these shortcomings effectively.
What Multi-Touch Attribution Is—and Isn’t
MTA aims to track each point of contact a customer encounters, from first touch to final purchase. By tracking these touchpoints, the goal is to determine which interactions “drive” revenue. However, MTA isn’t a causal tool; it’s merely a correlation model. And that’s a big deal. Causation implies that one event directly results from another, while correlation only indicates that two things happen together. Assigning revenue to each interaction without knowing if it caused the purchase can lead to misinformed decisions.
Incomplete Data: The Missing Pieces in MTA
Imagine you’re piecing together a puzzle of your customer’s journey. MTA often provides only a fraction of the full picture, especially with organic channels like LinkedIn or offline interactions that it simply can’t track. Think about a recent significant purchase you made—how many of those influencing factors were directly tracked by ads? Likely, very few.
This incomplete view causes businesses to over-invest in the easily tracked bottom-of-the-funnel (conversion-focused) efforts while neglecting the upper funnel activities that fuel long-term growth. Without a comprehensive understanding of the full journey, MTA-driven strategies may overvalue certain touchpoints, potentially hurting long-term revenue.
Misleading Impressions: The Problem with Stochastic Models
Certain MTA tools try to address these gaps by including stochastic (probabilistic) models to predict touchpoints based on past patterns. But this approach only guesses what might have happened, which is inherently unreliable. Unlike deterministic data (which shows real events), stochastic data is a “best guess.” For instance, LinkedIn ads often won’t show impressions due to confidentiality restrictions, leaving the real impact ambiguous.
Without accurate, deterministic data, these predictions are at best an educated guess, which could misdirect your marketing efforts.
Survivor Bias: Learning from What You Don’t See
During WWII, military strategists observed where returning planes were most damaged and initially thought they should reinforce those areas. But statistician Abraham Wald proposed focusing on areas where surviving planes weren’t hit, theorizing that planes hit in those spots likely didn’t return. This is survivor bias: focusing only on visible successes.
In MTA, survivor bias manifests as a focus on touchpoints of converted customers without considering those who didn’t convert. By ignoring the non-converters, MTA risks making misleading assumptions about what actually drives revenue. Just because certain interactions appear more frequently in converted journeys doesn’t mean they caused the conversion.
Misaligned Incentives: The Sales-Marketing Disconnect
When marketing is evaluated on MTA metrics, there’s a natural tendency to prioritize actions that show up easily in attribution models, like retargeting ads. However, channels that may be effective yet hard to track—such as certain PR initiatives or brand-building activities—may be underfunded or ignored entirely.
One example is Apple News, which can drive brand awareness and customer engagement but doesn’t show clear attribution metrics. Companies that reduce investment in such channels risk losing critical top-of-the-funnel visibility, which could hurt long-term growth.
Moreover, MTA can often clash with insights from the sales team, who are closer to the customer. Marketing might attribute a conversion to a LinkedIn ad, while the sales team knows the buyer’s real interest stemmed from past experience with the product. Misalignment here can erode trust between departments, leading to inefficiencies and missed opportunities.
The Incrementality Gap: Proving Real Impact
Incrementality measures the true impact of each touchpoint on a conversion, determining whether a particular action caused a change in behavior or was merely incidental. MTA struggles here. Simply assigning a dollar value to each interaction isn’t enough. To gauge real influence, you need robust incrementality testing methods, like marketing mix modeling or pre/post campaign tests.
Without incrementality testing, MTA can lead you to believe all touchpoints are equally valuable, overlooking the nuanced interactions that genuinely contribute to revenue.
External Influences: The Factors MTA Overlooks
Competition, price changes, new features, and even broader economic trends can significantly influence customer behavior. MTA lacks the depth to account for these external factors, often mistakenly attributing positive results to recent touchpoints instead of recognizing the real external influences. Marketing mix modeling offers a more holistic view by integrating these factors, ensuring you aren’t misled by short-term trends.
The Brand’s Long-Term Impact—and MTA’s Blind Spot
A strong brand presence creates a demand that may not be captured in any attribution model. MTA’s emphasis on direct touchpoints fails to capture the cumulative impact of brand-building activities, which may lead to underinvestment in these vital areas. Over time, this could result in stagnant growth, as the brand’s natural pull on customer demand goes unnoticed.
Word of Mouth: The Unseen Revenue Driver
Word of mouth is one of the most powerful forms of marketing but is largely invisible in MTA. Many companies have found that as much as 30% of revenue comes from referrals, a critical insight that’s missed without self-attribution surveys or brand tracking. Ignoring this key revenue driver can result in MTA crediting the wrong channels, skewing your understanding of where demand truly originates.
Moving Beyond MTA
While MTA provides some insights, it’s crucial to recognize its limitations. To truly understand what drives revenue, consider complementing MTA with more robust tools like MMM, self-reported attribution surveys, and incrementality testing. By doing so, you’ll gain a holistic view of your marketing impact, enabling better decisions that can drive sustainable growth.
In the end, a mix of data-driven insights and practical adjustments will empower you to align sales and marketing efforts, fostering an efficient strategy that truly reflects your market’s dynamics.