Boards demand credible carbon numbers. Regulators require auditable disclosures. Operational leaders need forward-looking climate risk models embedded in day-to-day decisions. Behind every one of these deliverables sits a data practitioner, the engineer, architect, steward, scientist, or analyst, quietly shouldering the burden of making it all work.

For data professionals, sustainability presents a uniquely demanding challenge: the outputs are highly visible, the inputs are notoriously messy, and the domain expertise cuts across nearly every function in the enterprise. In a recent SAP Let's Talk Data, Sustainability Analytics Unlocked: From out in the Field to into the Data, Ridwan Bhuiyan, Vice President of Digital Products at Chubb's Global Resilience and Risk Consulting practice, and Tina Rosario, Chief Data Officer at SAP, shared hard won lessons on what it actually takes to build a modern data foundation capable of supporting sustainability at enterprise scale.

Here's what practitioners need to know.

Sustainability is a downstream data consumer, plan accordingly

An important architectural insight in the conversation is this: sustainability analytics almost always sit downstream of a wide array of systems that generate the source data. That has profound implications for how you design pipelines, negotiate data contracts, and set quality expectations.

"A lot of times sustainability and resilience topics are kind of downstream from other activities in a business," Bhuiyan explained. "If we're thinking about a car manufacturer, a car manufacturer has to deal with procurement, operations at a facility, energy, so many different other components. And sustainability, specifically for measuring impact, has to go to those functions and collect data. So, they [sustainability and resilience] are downstream consumers, and a lot of times have to deal with the downstream carry over in terms of data quality."

For practitioners, this means sustainability inherits, and often amplifies, data quality issues that upstream systems tolerated because they didn't need the same level of detail. Take carbon accounting for a single automotive component: "They need to know the total quantity, the material type of that component, the life cycle, where it traveled from, how many miles it traveled, and what kind of fuel was used in that travel. The procurement function would know how to handle and ask for specific data points from their supplier. But the sustainability team would need far more than what was initially asked." The procurement team's dataset isn't "bad" - it's just incomplete relative to sustainability's requirements. "Data completeness is inherently a data quality issue," Bhuiyan noted. "They [sustainability and resilience] are essentially inheriting that data quality issue from another function of the business."

Data professional takeaway: Profile your upstream sources against sustainability schema requirements before you build downstream models. Identify completeness and quality gaps early, and treat every gap as a business conversation, not a technical patch.

Breaking down the ivory tower

Historically, sustainability functions have operated in isolation, what Bhuiyan called "an ivory tower, almost like a niche topic." Sophisticated analyses were performed, elegant reports were written, and then… nothing. The insights rarely made it back into operational decision-making.

"A lot of really cool, complicated analysis was performed in this space, but it kind of just stayed there," Bhuiyan reflected. "The vehicle for delivery was a report that we evaluated at the end of the year. For me, that’s a lot of wasted energy."

The solution is collaboration and integration. Sustainability teams must embed themselves with the functions that generate the underlying data. "Breaking down that tower and going and embedding yourself with the procurement teams and making a business case about how collecting more data could be beneficial to them in terms of carbon tracking is a way to overcome those data quality issues," Bhuiyan said. "As owners of that procurement data, they have the power to make data more complete. And then sustainability teams will inherit that benefit downstream."

Data profession takeaway: Data governance is essential, not as a bureaucratic overlay, but as a mechanism for aligning incentives and creating shared ownership across functions.

Building the data foundation starts with evangelism, not tooling

Data practitioners are often eager to jump to architecture: data fabric, data mesh, lakehouse, semantic layers, catalog tooling. But when asked what it takes to build a trustworthy data foundation for sustainability, Bhuiyan offered a "spoiler alert" that will not surprise any seasoned data leader: "Ironically, it doesn’t start with technology. At least initially, it has nothing to do with what kind of tech stack you use."

Instead, it starts with evangelizing, communicating the commercial and strategic benefits of embedding sustainability into the business so that downstream data owners become genuinely invested. So "when you go to a data producer, a data owner, … they’re not approaching it [sustainability] as a side project. They’re really invested in the idea of managing sustainability data comprehensively."

Only then does technology enter the picture. "How do you centralize assets across a business into one place so that when you need to run critical analytics for measuring sustainability performance or running through scenarios of climate resilience, you have all the bits and pieces you need in one place, and it’s easy to integrate others if you need to?" Bhuiyan explained. "Making a modern data platform approach is creating the critical foundation that you see in other corporate functions like finance, marketing, risk, and compliance - specifically for sustainability and resilience topics."

Data professional takeaway: Build stakeholder alignment and data contracts before optimizing the pipeline. A well-tuned data flow, data fabric, or data product can't compensate for a source system whose owners don't care about the data you're extracting.

Treating data as a product - the path to adoption

Data products play a critical role in transforming sustainability from a niche reporting exercise into an operational discipline that drives real business outcomes.  Sustainability analytics don’t create value if they sit unused. Bhuiyan emphasized that data leaders must wear "a product development cap,” thinking beyond the analysis to how insights will actually be adopted. "When you come across a business problem, it’s not all about dissecting it from a strategic standpoint but really thinking about how is this going to be translated into a product that’s going to be used and adopted and drive value," he said.

"We don't want to produce operational exhaust and fill our clouds with data that's not going to be used," Bhuiyan said. "Stewardship is really important, particularly for corporate resilience, climate resilience, sustainability, carbon tracking... by productizing those results, applying some stewardship, governance, and making them readily available to the rest of the business, they can be reused and really embedded with those teams. We can't really ask a company to reduce its impact if the functions within that business cannot necessarily measure [it]."

Rosario extended the point to ownership: "How do you empower those people downstream to take ownership in those data products so that they can be consumed upstream in an effective way? Stewardship, governance, attention to quality, understanding the source - it all goes hand in hand."

Data professional takeaway: Assign business ownership for every sustainability relevant data product. Publish SLAs, definitions, and lineage. Instrument usage. If a product isn't being consumed, treat that as a signal to investigate it, and see if there is a reason to keep it around.

Get your house in order before looking outward

Sustainability data doesn’t exist solely within corporate walls. Suppliers, weather patterns, geographic factors, and community-level dynamics all influence a company’s footprint and resilience. Yet Bhuiyan cautioned against jumping straight to external data as a shortcut.

"For me, it’s almost more important to get your house in order before you look externally," he said. "Once you know your house is in order, you can more effectively leverage external data or external insights or incorporate standards like industry standards in the right way."

Rosario echoed this, emphasizing the value of external data as validation rather than substitution: "To be using some of that external data as, okay, let’s do a benchmark, let’s understand what others are doing well, let’s make it a learning exercise, not necessarily just a collection exercise."

Geospatial - A Power Tool for Climate Resilience

For climate risk and resilience use cases, geospatial capabilities have become essential. Bhuiyan described geographic information systems (GIS) as "a form, an approach for analytics, but more and more it's becoming a way of telling stories about data."

For Bhuiyan, geospatial analytics allows his team to "assess vulnerabilities at multiple locations across different geographies" while integrating weather data, location, time-series information, and "forward-looking climate scenario data."

Data professional takeaway: Practitioners working in this space should invest in geospatial ready data models, coordinate reference standards, and semantic alignment between location entities across systems; otherwise, spatial joins become a source of silent data quality failures.

Intentional strategy, actionable roadmap

The conversation closed with a call for intentionality. Bhuiyan urged that "Being super intentional about your data strategy, your corporate data strategy - making sure that you’re bringing everybody to the table, understanding their needs, finding patterns, giving them a North Star to align to, and then a roadmap on how you’re going to get there.”  Rosario added, “Any leader in today’s business - data or otherwise - really needs to understand what are the value drivers, what are the objectives of the organization, and how can they move the needle in the right direction. Focus on those outcomes and let those outcomes drive the priorities."

Data professional takeaway: Sustainability, corporate resilience, and risk must have a seat at the strategy table, but a seat alone isn’t enough. Data strategies live or die on execution, and without an actionable roadmap to operationalize them, even the best designed strategy becomes a document that "just sits there for everyone to look at," as Bhuiyan put it.

The bottom line

Sustainability analytics succeed or fail on the strength of the underlying data foundation. That means:

• Map downstream dependencies and address quality and completeness gaps as business problems, not technical ones.
• Evangelize before you architect. Getting buy in from data owners determines quality far more than any tool.
• Productize data assets with clear ownership, stewardship, and consumption metrics.
• Get internal data right before layering in external sources.
• Bake sustainability, resilience, and risk into the enterprise data strategy with an actionable roadmap.

Sustainability is no longer a niche side project or an “ivory tower exercise,” it’s one of the most demanding tests your data foundation will ever face. It stresses every layer of the stack: source system completeness, cross-functional collaboration and governance, stewardship, product thinking, and execution discipline. If your platform can deliver trusted, reusable, decision-ready data for sustainability, it can likely deliver on any promise the business asks of it.