AI revenue management in CPG: From hindsight to decisioning
A conversation with Mike Milanowski on real-life examples of CPG decisioning automation
When CPG organizations navigate volatile demand, shifting shopper behavior, and relentless margin pressure, “more reporting” rarely changes outcomes. What changes performance is decisioning: the ability to turn commercial signals into clear, repeatable actions across pricing and promotions — fast enough to matter. This is the core challenge of AI revenue management in CPG.
From XTEL’s VP Strategic Advisory Mike Milanowski’s experience in global CPG environments, a pattern emerges repeatedly: when data is fragmented, workflows are spreadsheet-driven, decision cycles slow down, teams debate definitions, and execution becomes inconsistent across markets.
In a recent digital transformation engagement, Mike and the XTEL team partnered with a global consumer health company in North America — where those same constraints were preventing the organization from making real-time, fact-based decisions across a large consumer footprint.
Fragmentation, manual work, and decision latency: A common challenge among CPGs
“The commercial teams were operating in complex pricing and promotion dynamics, but with an operating model still heavily dependent on manual workflows,” says Milanowski. Over time, this created a familiar set of symptoms: slow insight generation and distribution, limited cross-functional alignment, and inconsistent execution across pricing, promotions, portfolio, and customer profitability.
His takeaway is pragmatic: most revenue management programs don’t stall because teams lack expertise. They stall because the organizational and technological ecosystem around them can’t deliver clarity and speed at scale. The mandate, therefore, should not be “let’s get better dashboards.” CPGs need to build a unified revenue management foundation — one that supports faster, smarter execution in a dynamic market.
Implementing the right approach: AI that drives actions, not just analysis
In this engagement, the objective wasn’t to add another analytics layer. The aim was to move beyond manual work and enable automated, insight-rich decision flows — the foundation of effective AI revenue management in CPG, especially around promotion effectiveness and pricing strategy.
The targeted capability set anchored Revenue Growth Management outcomes: promo ROI optimization, predictive promotion uplift forecasting, pricing insights, and assortment analytics. The goal was to help commercial teams make better decisions faster — not simply explain performance after the fact.
Building the backbone: a scalable architecture anchored in the data lake
A field-tested lesson Mike consistently emphasizes is that architectural choices can accelerate or delay value generation. In this case, XTEL designed the solution to operate independently of direct integrations into the broader IT ecosystem. The company could capitalize on XTEL’s ability to capture data directly from the data lake and ERP — Snowflake and SAP, among others — to build a clean, scalable backbone for revenue growth management.
This design principle matters for CPGs operating at multi-region scale: reduce complexity where possible, standardize definitions early, and build a data foundation that supports consistent decisioning over time — avoiding one-off analyses.
Beyond technology: governance, adoption, and shared KPIs
In Mike’s experience, the most underestimated element of revenue management transformation is adoption. Even strong analytics won’t translate into better commercial performance unless teams align on how decisions are made and executed. That’s why the project included structured change management, capability building, and strategic advisory support — market assessments, tailored workshops, and more. Driving adoption across Sales, Finance, Strategy, and RGM teams demands exactly that level of focus.
XTEL designed the sessions to generate insights into pricing elasticity, promotion efficiency, and customer profit pools — while improving alignment on shared KPIs and data foundations. These are the essential building blocks for scaling decision-making beyond a single market or team.
The rollout: multi-region TPM, TPO, and AI-driven decision support
The initiative was structured as a multi-region release spanning three continents, and all the key XTEL solutions: TPM, TPO, and ADAM.
Within TPO, the organization leveraged AI-infused boosters to support portfolio, pricing, and promotion decision-making — a practical lever for improving commercial planning quality and efficiency at scale, particularly when teams need to make comparable decisions across markets with varying levels of maturity.
Implementation discipline: quality, trust, and “day-one” readiness
Mike knew that if a platform doesn’t earn its users’ trust early, it becomes shelfware. That’s why the team deliberately prioritized quality, accuracy, and user readiness — through rigorous UAT cycles, cross-functional testing, and iterative refinement. The goal: a platform that functions as an AI-driven decisioning engine from day one.
This implementation discipline matters because revenue management is not just a model output. It is a set of repeatable commercial decisions, made and executed by people who need confidence in the numbers, the definitions, and the workflows.
Impact: turning insight into action faster
Although the transformation was multi-region in scope, early results already demonstrated impact in the U.S. market — driven by faster conversion of insights into action. The program projects more than $20M in five-year net sales uplift: a clear signal of what happens when AI revenue management in CPG shifts from hindsight reporting to active decisioning and execution.
Conclusion
AI-powered Revenue Management delivers real value only when treated as an operating model — not a reporting layer. In this global consumer health case, the shift required a scalable data backbone (Snowflake + SAP ERP), clear governance and shared KPIs, and a serious adoption plan supported by rigorous UAT and a disciplined multi-region rollout. The result: a path from explaining what happened to deciding what to do next — and executing consistently across markets.
Ready to move from reporting to decisioning?
Key takeaways:
- Decisioning beats reporting: performance improves when insights translate into repeatable pricing and promotional actions.
- Data foundation is a growth enabler: reducing integration complexity while anchoring directly on data lakes and ERPs (SAP, Snowflake…) helps scale across regions.
- Adoption is not optional: workshops, capability building, and shared KPIs are what make the tools usable in day-to-day decisions.
- Quality earns trust early: rigorous UAT and cross-functional validation prevent “shelfware” and protect day-one value.
- Scale needs structure: multi-region rollouts work best when definitions and workflows are standardized upfront.
Good to know:
What is AI Revenue Management (or AI RGM) in a CPG context?
AI Revenue Management applies advanced analytics to commercial decisions, especially pricing and promotions—so teams can move from hindsight reporting to decisioning: choosing actions based on predictive signals, consistent KPIs, and repeatable workflows.
Why do revenue management initiatives stall even when the analytics are strong?
Because the bottleneck is often the operating system around decisions: fragmented data, manual workflows, inconsistent definitions, and limited cross-functional alignment. Without clarity and speed at scale, execution becomes uneven across markets.
What are the non-negotiables for scaling RGM across regions?
A scalable data backbone, early standardization of definitions and KPIs, and a structured enablement plan for Sales, Finance, Strategy, and RGM teams. Multi-region scalability is usually won or lost on governance and adoption, not on dashboards.
How do you build trust in an AI decisioning platform quickly?
By prioritizing quality and user readiness: rigorous UAT cycles, cross-functional testing, and iterative refinement—so users see consistent outputs, transparent definitions, and reliable workflows from day one.
What kind of value can AI-driven pricing and promotions unlock?
In this example, early impact was visible in the U.S. market, with more than $20M projected five-year net sales uplift attributed to faster conversion of insights into action.