Top 5 CPG replaces stringent physical safety testing with high- accuracy prediction from Turing to unlock unprecedented speed


Safety testing for aerosol deodorant was too time-consuming to maintain market leadership position

How do you increase the speed and capacity of R&D across the organization using advanced AI and machine learning methodologies? This was the major question for a top five CPG company, and leadership felt the answer was through automation.

They were specifically concerned about their market-leading aerosol deodorant, which had a 60% market share but required significant safety testing across 11 categories to meet regulatory requirements. The physical testing was slowing down the CPG’s ability to get new aerosol formulations to market quickly and maintain their $2.4 billion market share. The company felt they needed to cut product development time by a significant amount to compete with newer brands that were flooding the market.

There was one additional hurdle: in-silico testing had never been done successfully in this market. The level of accuracy simply wasn’t high enough to meet the strict safety standards needed for aerosols. But if the company could achieve a high level of accuracy through digital testing, it could set a new standard for R&D at scale.

The Product:
Aerosol Deodorant
The Goals:
Maintain market leadership position by going to market faster with an AI testing solution
The Solution:
Automated, in-silico testing that mimics physical testing and reduces testing time
“Our scientists are usually pretty skeptical about predictive tests, but Turing’s accuracy removed any concerns that in-silico was both safe and viable for our business.”
- Head of digital R&D


Use Turing’s AI platform to prove viability of in-silico testing by validating digital formulations with a high level of accuracy

The CPG had somewhat limited data – quantitative and qualitative – that Turing’s AI platform was able to ingest. But since Turing’s AI platform was purpose-built for CPG Lean innovation, it was still able to bring these insights together and simulate optimal formulations that would meet requirements across the 11 key categories of safety. After the AI platform recommended a number of prototypes, the company’s scientists tested them in the lab.

Using out-of-sample validation, the company found that Turing’s AI platform had achieved exceptional levels of accuracy. This proved the viability of in-silico testing for aerosol deodorant for the first time ever, and showed the CPG a clear path towards automation and digital transformation.

With Turing’s AI platform, the company was essentially enabled with a digital R&D lab that automated the manual processes its scientists had been using. The platform filled data gaps that existed between quantitative data and human expertise, created a single source of truth for formulations, and enabled significantly faster product development.


The CPG discovered that Turing’s AI platform achieved out-of-sample validation accuracy between 88-96% across the key safety categories for aerosol deodorant. This level of accuracy alleviated the company’s concerns about going in-silico and will help unlock significant ROI long term, including 3X-4X faster go-to-market, productivity boosts, and faster competitive response. The company is confident that with Turing’s AI platform, they’ll not only maintain their dominant market position in aerosols, but grow in new markets, as well.

accuracy in out-of-sample validation
faster speed to market potential
every predicted test saves 12 weeks
savings realized in 2021
Back to the Resources