Materiality assessments have never been more scrutinised. With ESRS now setting a clear expectation for double materiality, companies are under pressure to show that their process is not only compliant, but credible. Yet in practice, there is a growing gap between robust, decision-useful assessments and those that are either overcomplicated or too superficial.  At the same time, the context in which these assessments are being carried out is shifting rapidly. From geopolitical instability and supply chain disruption to energy price volatility and workforce risks, the assumptions underpinning materiality are moving targets. Add to this the rapid emergence of AI tools promising to automate the process, and it is easy to see how companies can either overcomplicate or oversimplify their approach.

Don’t prejudge the outcome

One of the most common pitfalls is starting with a conclusion and working backwards. Whether driven by internal politics, legacy topic lists, or a desire to avoid difficult issues, some companies are effectively validating what they already believe.  This is increasingly risky. ESRS requires companies to assess impacts, risks and opportunities with a clear audit trail. If the process appears selective or biased, it is unlikely to stand up to scrutiny, from auditors, regulators or investors.

Good practice starts with openness. That means allowing for the possibility that topics move, some rising in importance, others falling away, and being ready to explain why. In a more uncertain world, this also means being prepared for greater movement than in the past.

Getting the balance right

There is a wide range in how much effort companies are putting into materiality. Some are running very large, complex exercises that take significant time and resource. Others are doing very light updates, relying mainly on desk research and concluding that little has changed. Neither approach works particularly well.  Claims that materiality can be reduced to a purely objective, judgement-free exercise are misleading. The real risk is not the presence of judgement, but unstructured or hidden judgement.

A strong materiality assessment does not need to be overly complex, but it does need to be grounded in evidence, structured in how it gathers input, and clear in how conclusions are reached. It should reflect what is happening in the business and the wider world.  This is where some of the current hype around AI needs to be treated carefully. Digital tools can help process large volumes of information and identify patterns, but they cannot replace judgement. A materiality assessment still requires companies to interpret evidence, challenge assumptions and make defensible decisions.

In practice, the most effective approaches sit somewhere in the middle: building on what already exists but taking the time to gather fresh input and test assumptions, particularly where external conditions are changing.

What good looks like in practice

Leading companies are moving away from one-off exercises and embedding materiality into how they manage risk and strategy. A few examples stand out:

• Unilever has long integrated materiality into its business strategy, linking material topics directly to business priorities and regularly updating its assessment based on stakeholder and market changes.
• Ørsted provides clear documentation of how it assesses impacts, risks and opportunities, with strong alignment between its materiality process and its risk management framework, helping it meet both reporting and strategic needs.
• Nestlé combines global consistency with local insight, recognising that issues such as water or land use can vary significantly by geography, and reflecting this in its analysis and disclosures.

Across these examples, a few common themes emerge; clear links to strategy and risk, structured input from across the business, and transparency in how decisions are made.

Designing a robust process

A strong process should build on what already exists, rather than starting from scratch. Most organisations already have a wealth of relevant insight: risk teams maintain risk maps, sites engage regularly with local stakeholders, investor relations teams are in constant dialogue with investors, and commercial teams understand shifting customer expectations. The challenge is not generating more input, but bringing these perspectives together in a structured and consistent way.

This means:

• Using existing data and insight as the foundation, supplemented with targeted additional input where needed
• Collecting input from across the business in a consistent format, so it can be compared, aggregated and explained
• Linking materiality clearly to risk management and internal processes, ensuring conclusions are grounded in evidence
• Applying clear and consistent criteria when assessing topics, so judgements are comparable across functions and geographies
• Using AI and digital tools to support the process, for example in synthesising inputs or identifying patterns, but not relying on them to replace judgement or run the process end-to-end
• Regularly testing underlying assumptions, particularly in a context of heightened global uncertainty, ensuring that conclusions remain relevant as conditions change
• Documenting how judgements are made, so the process is transparent, repeatable and can stand up to scrutiny

For large and complex organisations, it is also important to capture regional differences in a structured way, highlighting where issues are more significant in certain geographies, while maintaining an overall group-level view. This avoids spending a disproportionate amount of time on minor regional detail, while still allowing patterns and insights to emerge, without distorting or overriding the group-level conclusions.

A shift in mindset

Materiality is no longer just a reporting exercise. It is about making informed judgements in an uncertain world and being able to stand behind them. That means being open to change, investing the right level of effort, and recognising that the outcome is not fixed. It also means resisting the temptation to either outsource decisions to tools or assume that last year’s answers still hold.  A good materiality process doesn’t just confirm what you already think, it helps you see what might be changing.