Use AI to speed up product development.

If this statement evokes images of androids in lab coats cooking mechanically in a test kitchen, you’re not alone.

But after the initial scifi image dissipates, maybe something more akin to cautious hope arises: could it actually work? Well, yes and no.

Many food companies are using AI in some capacity. Unilever, Nestle, and Kraft are all public about their investments and achievements in leveraging the technology to ultimately improve their products while getting them into the hands of consumers faster. However, how to actually leverage this kind of technology for your company remains somewhat of a mystery.

If we look behind the curtain, we often see research and product development data scattered in spreadsheets, some formatted, some not — and much of it in point systems like LIMS, ELN, PLM and PPM. All holding a wealth of information inaccessible for the combined analysis required to make decisions that will actually speed up product development.

Hypothetically, you spend 6 months to format and clean this data up, feed it into a cool algorithm from a fancy tech company and voila! You have all the AI-generated predictions needed to speed up product development.

This never materializes. Why?

Clicking a button is not going to magically solve the latest ingredient supply disruption, or satisfy the consumer’s drive for less sugar in their favorite melt-in-your-mouth candy bar, especially if the right data hasn’t been captured along the way. And 6 months of cleanup isn’t going to produce results if the data wasn’t there to begin with.

AI requires clean, connected, and consistent data.

But let’s set AI aside for a moment. What some companies fail to realize is that if they just start to focus on collecting more structured data, giving their data the attention it deserves, they will start reaping benefits immediately.

Projects become easier to monitor and collaborate on interdepartmentally, and the data is easier to visualize and less prone to error. That alone will speed up product development more than you think.

In order for such an effort to be successful, it has to be a company-wide initiative. Data can no longer be siloed by department, and everyone generating data must follow the same structure and understand that all of their inputs matter.

Data scientists can create impressive algorithms, but they aren’t generating the data needed to make them useful — your food scientists and technicians are. If you can make their jobs more doable with systems that are easy to use and integrate with other software and equipment, they can pull and aggregate the data they need for effective analysis.

Then your entire organization will experience the benefits of operating within the same structure, and the company’s data will be well on its way to AI readiness.

But it all starts with the day-to-day data contributions, made by every person at every level of your company, to turn the magic of AI into a reality.

Ready to accelerate your innovation?
Scientific AI Roadmap >Request a Demo >