How DataRobot Built Data-Driven Enterprise Marketing Strategy with HockeyStack

The Who

DataRobot is an enterprise AI platform that helps organizations build, deploy, and manage agentic and generative AI applications at scale. It automates much of the data science process—like feature engineering, model selection, and tuning—so teams can develop AI solutions faster while ensuring governance, transparency, and monitoring. DataRobot equips teams with the tools and insights needed to speed up AI adoption, reduce complexity, and make agentic and generative AI accessible beyond expert data scientists.

Doug Cone spearheads growth marketing and operations at DataRobot, with a focus on connecting marketing efforts to revenue impact. Alongside Ryan Moore, who leads marketing data strategy, their team ensures that campaigns are optimized across channels, attribution is deeply understood, and marketing investments are aligned with business growth. Together, they enable DataRobot’s marketing organization to move beyond gut instinct and toward a culture of measurable, data-backed decision making.

‍

The Problem

DataRobot’s marketing team had already experimented with attribution—using both in-house last-touch models and a prior vendor—but the results weren’t reliable enough to drive confident budget decisions. Last-touch reporting meant that spend was often shifted based on incomplete or misleading signals, causing the team to double down on channels that looked successful in isolation but ignored the broader customer journey.

“We were making bad decisions because of the lack of quality. One data source would say something was performing, but it missed the aggregation of all the touches that actually led to an opportunity.”

A key gap was the inability to track and connect anonymous touchpoints that didn’t tie directly to a form fill. Much of the buying journey remained invisible, leaving the team without a full picture of what influenced pipeline. On top of that, large data volumes created technical challenges making it difficult to get actionable insights at scale.

Without a platform that could consolidate all touchpoints, compare multiple attribution models at once, and reliably handle their enterprise-level data, DataRobot risked continuing to invest in campaigns based on partial truths rather than the full story of what drove opportunities and revenue.

‍

The Requirements

• Anonymous Touchpoint Tracking: A way to capture and connect touches that weren’t tied to form fills or known users, so the team could finally understand how the full journey influenced opportunities and revenue.
• Multi-Model Attribution: Flexibility to view and compare multiple attribution models at once, instead of relying on a single model that risked oversimplifying the story and leading to bad decisions.
• Enterprise-Grade Data Handling: A platform capable of ingesting and processing DataRobot’s large data volumes without lag or delays, unlike competitors that struggled to keep up during trials.
• Journey & Lift Analysis: Visibility into how touches worked together across the funnel—beyond just attribution outputs—to answer questions like which campaigns, keywords, or channels actually drove meaningful impact.
• Ease of Use for Teams: A solution that could serve as both an executive-level dashboard and a day-to-day workspace for channel owners to embed data-driven decision-making into marketing culture.

“A must-have was tracking touchpoints that weren’t necessarily tied to a form fill or a known user. Most of our touches don’t drive a form fill—how do we get that into our tracking to really understand the journey?”

‍

The Alternatives

Before choosing HockeyStack, DataRobot evaluated several options. The team had prior experience with Bizible (Marketo Measure), which integrated tightly with Marketo but lacked the ability to capture anonymous touchpoints (one of the biggest requirements for DataRobot’s marketing organization). Having already seen the limitations of relying on Bizible’s attribution outputs, they knew it wasn’t the right fit.

They also trialed Dreamdata, but quickly ran into performance issues. The platform struggled to ingest and process DataRobot’s large data volumes, leaving the team lagging behind in their reporting. More importantly, Dreamdata’s platform didn’t offer lift reporting or time-period comparisons, limiting the utility of the platform.

‍

“We knew what we would get with Bizible, and we knew we did not want that. Anonymous touchpoints were a huge part of the gap. And when working with Dreamdata, we were always lagging behind on data just because of the size of our dataset.”

‍

Custom in-house models, such as last-touch attribution, were already familiar but had proven insufficient, leading to biased budget allocation and missed insights about the true buyer journey. Ultimately, these alternatives highlighted the need for a platform that could combine robust data handling, multi-model attribution views, and the ability to track anonymous touches in one place.

‍

The Insights

• Branded Search Insights: What appeared to be strong Paid Search performance was actually concentrated in branded keywords at the bottom of the funnel. HockeyStack revealed the hidden impact of earlier non-branded touches that had been overlooked.
• Attribution Models Tell Different Stories: By comparing multiple models side by side, the team learned how attribution perspective changes outcomes—helping educate stakeholders that there isn’t one “truth” but multiple valid lenses.
• Journey Visualization: Trend and journey charts gave visibility into how different touchpoints worked together, showing the importance of sequencing rather than isolated conversions.
• Team Education: Dashboards became tools not just for reporting, but led the way for organization-wide mindset shifts.

‍

The Results

“The return on HockeyStack isn’t just numbers. It’s our team becoming data-driven, making decisions based on evidence instead of gut instinct. With HockeyStack, our marketers can tell the story behind the data — not just read a chart.”

• Smarter Paid Search Investment: With new clarity into branded vs. non-branded performance, the team reallocated spend more effectively, ensuring money went toward campaigns driving demand.
• Faster, More Confident Decision-Making: Having reliable attribution views eliminated second-guessing and allowed marketers to defend spend strategies with leadership using data, not anecdotes.
• Improved Executive Alignment: The CMO dashboard provided at-a-glance visibility for leadership, while tailored attribution views empowered individual teams—bridging the gap between high-level strategy and day-to-day execution.
• Cultural Shift Toward Data-Driven Marketing: Bi-weekly metrics reviews forced every team member to analyze and present their own data, embedding accountability and boosting organizational comfort with analytics.