The Evolution of Materials Science in the Commercial Sector
Materials informatics started as a research idea and became a commercial advantage. What began as government and academic efforts to structure materials data has turned into a competitive requirement for the companies that develop chemicals, coatings, batteries, and advanced materials at scale. The organizations pulling ahead are the ones that treat their experimental data as a durable asset rather than a byproduct of the lab.
Where Did Materials Informatics Begin?
Materials informatics began in government and academia, led by scientists who wanted to structure materials data well enough to learn from it. Early work at the National Institute of Standards and Technology, including the Materials Genome Initiative and the Materials Data Repository, set out to make materials properties searchable and reusable. The University of Bonn's Pearson's Crystal Database followed the same instinct at scale, cataloging hundreds of thousands of data sets and chemical formulas on crystal structure. The shared premise was simple: when materials data is centralized and structured, it becomes something teams can query, correlate, and build on.
How Did Materials Informatics Move Into Industry?
Materials informatics moved into industry when manufacturers saw that structured data could shorten the path from lab to market. As academic studies drew commercial attention, many industrial manufacturers began building their own centralized databases of formulations, properties, and experiment results. The goal was to apply statistical learning and data analytics to their own research, so they could get better materials to market faster instead of rediscovering work they had already done.
Many of today's largest multinational materials companies, including Dow Chemical, BASF, and Evonik, have announced major partnerships with technology firms. These initiatives center on applying machine learning, data mining, and analytics to materials development, a clear signal that data-driven R&D has moved from experiment to expectation.
What Are Materials Companies Doing With Their Data Today?
Today, leading materials companies are consolidating decades of experimental data into single, searchable platforms and measuring the payoff in time and cost. The pattern across industries is consistent: centralize the data, structure it so it can be queried by content, and connect R&D to quality control and product lifecycle management so nothing is lost in handoff.
The results are quantifiable. Sika, in its published November 2025 investor materials, reports roughly 75% fewer experiments, more than 50% faster time to market, and a library of more than 100,000 data points that expands daily. Repsol consolidated more than 10,000 data points from two decades of formulation work into one searchable platform and cut its experimental workload by 30 to 40% per project. SCG Chemicals reports a 20% increase in R&D output alongside a 45% reduction in DOE workload. These are not projections. They are outcomes from materials organizations that decided to structure their data first.
Why Does Structured Data Come Before AI in Materials Development?
Structured data comes before AI because machine learning cannot find patterns in data it cannot read.
What Does the Commercial Payoff Look Like?
The commercial payoff shows up as faster cycles, recovered time, and institutional knowledge that compounds instead of resetting. AGC Chemicals cut weeks from its experimentation cycle after centralizing its data globally. Covestro reports development timelines up to six months faster. Ripple Foods reduced time spent on data reconciliation by at least 30% and freed roughly half a day per scientist each week. The common thread is that a structured data layer turns each experiment into a reusable asset, so teams stop repeating work and start compounding what they learn.
This is the shift materials science has been building toward for two decades. The science has always been rigorous. What is new is the ability to centralize, structure, and leverage that science across an entire organization, and to do it in one connected platform rather than a patchwork of tools.
