Attribution Challenges
As marketing teams adopt increasingly sophisticated models, the need for clarity, alignment, and actionable insights has never been more critical. In a recent roundtable discussion hosted by Joaquin Dominguez, Head of Marketing at Adzact, three senior marketing leaders shared their experiences, challenges, and solutions in tackling attribution complexity. The conversation underscored the importance of data integration, multi-touch measurement, and balancing automation with human oversight to drive meaningful revenue impact.
Hosts
Joaquin Dominguez was joined by marketing experts to discuss the topic.
Guests
Meghan Harris – Head of Marketing Operations, SoftServe
Manish Rai – VP of Product Marketing, SnapLogic
Yuliya Maystruk – Senior Manager, Marketing Operations, Redpanda Data
The Complexities of Attribution in B2B
Attribution remains one of the most debated aspects of B2B marketing. Unlike B2C models, where attribution is relatively straightforward—such as tracking a direct purchase following a Google ad—B2B attribution involves longer sales cycles, multiple stakeholders, and a variety of touchpoints that contribute to deal progression. The discussion highlighted a range of challenges, from data fragmentation to differing internal expectations on what should be measured and credited.
A major hurdle in attribution is the lack of a one-size-fits-all approach. First-touch and last-touch models provide limited visibility, often failing to capture the full customer journey. Multi-touch attribution, while more comprehensive, presents its own difficulties—particularly in ensuring data accuracy and defining the weight of each marketing touchpoint.
"Attribution is only as strong as the data foundation behind it. By ensuring a seamless integration of touchpoints within our CRM, we create a system that not only tracks interactions but also informs strategic decision-making across marketing and sales" - Yuliya Maystruk.
Building a Unified Data Foundation
One of the key themes that emerged from the discussion was the necessity of a robust data infrastructure. Yuliya Maystruk outlined how Redpanda Data centralises attribution within Salesforce, integrating multiple touch models—first touch, last touch, and multi-touch—through an ETL vendor. This ensures that every relevant interaction, from event attendance to content downloads, is accounted for within a single system.
*Extract, transform, and load (ETL) is the process of combining data from multiple sources into a large, central repository called a data warehouse. (https://aws.amazon.com/what-is/etl/)
To refine attribution further, Redpanda Data has developed a structured approach that layers single-touch, last-touch, and multi-touch models, each providing a different lens on engagement. The framework accounts for both contact-level opportunity attribution and an ABM-driven perspective, ensuring all relevant interactions across an account contribute to the overall measurement. The ability to toggle between these models offers deeper insight into where marketing efforts yield the highest ROI, allowing for more strategic investment decisions.
Meghan Harris echoed this approach, explaining that SoftServe is transitioning from a basic first-touch model towards a customer journey-based framework. By mapping interactions across different departments—including sales and customer success—their goal is to create a more holistic attribution model that informs both strategic decisions and day-to-day campaign adjustments.
However, both Meghan and Yuliya acknowledged that integrating sales data into attribution models remains a challenge. Many organisations struggle to merge sales and marketing activities within the same framework, leading to misalignment in understanding which efforts truly drive conversions.
"AI is transforming attribution by allowing us to connect fragmented data sources and uncover deeper insights into engagement patterns. However, its real value lies in its ability to adapt and refine models over time, ensuring we remain responsive to evolving customer behaviours" - Manish Rai.
Automation and AI: A Double-Edged Sword
The conversation then shifted to the role of automation and AI in improving attribution accuracy. Manish Rai highlighted SnapLogic’s use of AI-driven tools to track the impact of content and campaign performance. With vast amounts of content being deployed across different platforms, AI is becoming increasingly essential in measuring which assets contribute to pipeline growth.
However, AI alone is not a silver bullet. Meghan pointed out that many AI-based attribution tools, while promising, are still in their early stages. The challenge lies in ensuring that AI-driven insights align with the realities of human decision-making in complex B2B sales cycles. Additionally, manual data entry remains a persistent issue, particularly in large sales organisations where top-performing reps often bypass CRM updates.
"Attribution should serve as a tool for decision-making rather than a scoreboard for marketing. The real value comes from using it to refine strategies, optimise spend, and drive meaningful business outcomes" - Meghan Harris.
Aligning Attribution with Business Objectives
A critical takeaway from the discussion was the importance of aligning attribution models with broader business goals. Meghan warned against using attribution merely as a means to ‘give credit’ to marketing efforts, arguing instead for a more pragmatic approach—one that helps inform budget allocation, campaign optimisation, and revenue forecasting.
Manish elaborated on this point, describing how SnapLogic analyses SEO impact, campaign engagement, and sales interactions to refine its Ideal Customer Profile (ICP). By clustering customer data based on industry, company size, and behaviour, they are able to prioritise accounts that are more likely to convert. This approach not only improves targeting but also helps in tailoring messaging that resonates with high-intent prospects.
Measuring Offline Interactions and Events
Another pressing issue raised was the challenge of tracking digital engagements before a user is identified, often referred to as the 'digital dark web.' Many interactions occur anonymously until a user is cookied, making early attribution difficult. In contrast, offline interactions, such as trade shows and customer dinners, are often more straightforward to track through list imports and real-time form automations. Yuliya Maystruk shared how Redpanda Data has analysed attribution beyond pre-open stages, examining engagement to gain deeper insight into how various touchpoints contribute to deal progression.
Meghan reinforced the need for flexible attribution frameworks, particularly for high-touch initiatives. She cautioned against applying uniform attribution metrics to vastly different types of engagements, advocating instead for tailored measurement approaches that align with each campaign’s strategic objectives.
Final Takeaways: Practical Advice for Marketing Leaders
As the discussion drew to a close, each speaker shared their key piece of advice for marketing leaders grappling with attribution:
Meghan Harris: Avoid striving for perfection—attribution will never be 100% accurate. Instead, focus on directionally useful insights that inform better decision-making.
Yuliya Maystruk: Invest in building a strong data foundation from the outset. Ensuring CRM integrations are properly structured will save countless hours in the long run.
Manish Rai: Prioritise key data points, automate processes wherever possible, and adopt an iterative approach to attribution improvements—continuous refinement is more valuable than seeking an elusive ‘perfect’ model.
Join the Precision Punks Community
Access to the slack community of serious B2B advertisers
Chatham House rules ‘Adzact confidential’ events with peers
Private in-person events in London and New York
Exclusive data & learning resources to build your career
