Why an externally produced materiality assessment is a first draft, not an answer

A growing number of ESG data providers sell materiality assessments while positioning themselves as data platforms rather than consultants, a distinction that's mostly commercial: it lets them sidestep the obligations of a consulting engagement. What a company actually receives back is a structured, prioritised judgement about which sustainability topics matter to them specifically, built from sector averages and generic operational inputs, arriving with an authority that discourages anyone from questioning it.

In assurance work across hundreds of engagements, the pattern is consistent: implemented without internal review, these assessments run roughly 75% accurate. The 25% that's wrong isn't evenly distributed, it concentrates in misidentified or entirely missing topics an algorithm has no way to know about (a company-specific supply chain vulnerability, a recent strategic decision), incorrect prioritisation against sector benchmarks that don't reflect actual exposure, and one half of a double materiality assessment, impact or financial, getting genuinely thin treatment while the other gets real rigour.

The fix isn't to stop using external providers, their sector knowledge and benchmarking data are genuinely useful. It's to treat their output as an input to an internal process rather than the process itself: cross-functional review against operational knowledge, real stakeholder validation instead of proxy data, and a documented trail from external draft to internally-owned final assessment. Under CSRD the materiality process is itself subject to assurance, which means 'the data provider said this was material' stops being an acceptable answer the moment an auditor asks why a topic made the list.