The Challenge
The brand's marketing team was trapped in an attribution nightmare that had compounded over multiple iterations of their multi-touch attribution model. Data flowing through the system was impossible to trace, and the team could not answer fundamental questions: which channels actually drove revenue, where to increase spend, or what their real customer acquisition cost was by channel. As the premium linen brand expanded into retail through Nordstrom and its own store locations, a critical gap widened. Customers discovering the brand through online ads but purchasing in-store generated zero attribution credit for the digital channels that drove awareness. Marketing was being judged solely on direct e-commerce conversions, even as campaigns drove significant retail traffic.
The Constraint
The structural problems ran deep. Only 2-4% of user sessions could be attributed to actual customers. Four percent of orders vanished entirely in the data pipeline because the model assumed every order had a corresponding web session. Duplicate records inflated order counts by 9%. The attribution logic had grown through patches and workarounds, including Blisspoint probability overrides, coupon logic, and influencer tracking, creating spaghetti code where changes to one area broke three others. Leadership asked which channels to invest in, and marketing responded with guesses dressed up as analysis.
The Approach
We mapped every customer touchpoint across digital, offline, and retail channels before writing a single line of code. A critical data audit revealed user identification was broken at the source. We rebuilt the attribution foundation around four independent, auditable components: touches, spend, revenue, and attribution logic, each modular enough to evolve without breaking the others.
The Outcome
User attribution coverage improved from 2-4% to 13-17%, a 4-7x improvement that captured dramatically more of the actual customer journey. The 4% order loss and 9% duplicate inflation were eliminated, giving finance and marketing a single source of truth across all channels. Adding new touchpoint sources went from multi-month rebuilds to same-day deployment. Reporting that previously took three weeks became instant self-service dashboard refreshes in Looker. Marketing gained full visibility into online-to-offline journeys for the first time.
What This Unlocked
Confident retail expansion. With clear attribution connecting digital spend to store traffic, leadership scaled both the Nordstrom partnership and owned retail with conviction.
Strategic budget conversations. Defensive justifications replaced by proposals backed by omnichannel attribution data, including proof of digital marketing's ROI on retail sales.
Continuous channel optimization. Marketing tests new channels and measures performance immediately across e-commerce and retail, no longer waiting months for attribution updates.
Cross-functional trust on one set of numbers. Finance, marketing, retail operations, and growth teams all work from the same auditable data, shifting debates from accuracy to strategic direction.
Services
Tech Stack
Results
User match rate
2–4% → 13–17% (4–7x improvement)
Order data accuracy
Eliminated 4% loss + 9% duplicate inflation
New channel onboarding
Same-day deployment (was multi-month)
Cross-functional alignment
Finance, marketing, and retail on single source of truth