Weekly Recap: Co-packer & Applications Lab Progress
The first Soylent factory was in Rob, Matt, and John’s (already) cramped San Francisco kitchen, which came to look a lot like a storeroom for a busy bakery: big bags of oat flour and maltodextrin stacked chest-high and a light beige dusting on countertops piled with tubs of powder. We needed a larger production space, and fast.
There are two basic options for producing a foodstuff at an industrial scale. The first is acquiring and operating your own manufacturing facility, which would give us complete control over every aspect of the production process. However, to operate our own facility would add a host of other concerns, including an array of expenses, a major time commitment, increased overhead, staffing, production and inventory management, regulatory compliance, developing relationships with suppliers, and more. With a ship date on the close side of 3 months and no capital to invest in a facility, this wasn’t a reasonable choice.
The second option is the one that we found best for Soylent: contracting with an existing manufacturer, known in the industry as a co-packer. A co-packer already has the facility, equipment, and expertise to create a product that meets our high standards, which allows us the time and energy to concentrate on the countless other essential activities of starting a company.
In the past two months, we’ve been working with a co-packer to finalize our formula, a process which includes:
- Establishing target nutritional values
- Deciding upon ingredients to best meet our nutritional goals (based on nutritional content, taste, physical characteristics, and market availability)
- Creating initial versions optimized for nutrition (flavor/texture comes later)
- Distributing formulas to beta testers and gathering feedback
- Iterating formula until nutritional profile is finalized
- Working with co-packer’s applications lab to improve the flavor and texture of the product while maintaining established nutritional profile
- Initiating the final round of beta testing to work out any remaining issues
- Finalizing the formula
Currently, we are working through Step 6. Last week, the team flew to New York in order to work at our co-packer’s applications lab and resolve the formula’s remaining issues. These issues include:
In order to make Soylent a complete source of nutrition, we include a mix of vitamins and minerals in the formula. Vitamins and minerals tend to have very potent bitter and sour flavors, so we’re working on how to best suppress their tastes. The best way to do this is by a process known as encapsulation in which the vitamin and mineral particles are coated with an inert substance (often maltodextrin) to inhibit their flavor and smell from overpowering the rest of the formula. This is still a proprietary process, and we’re working on finding a suitable supplier for our vitamin blend.
Soylent is thick and creamy, kind of like a milkshake. The beta versions also have a mild to moderate chalky texture, which is caused by the mesh size of the rice protein we are currently using. Some testers haven’t minded the texture, but others have had issues with it, so we’re testing higher mesh-size rice proteins to find one that is smoother.
We are excited about the progress we have made resolving these last bugs in the Soylent formula and have established a solid roadmap for achieving a final formula. We understand that all of you are eager to see the exact ingredient and nutritional breakdown of Soylent 1.0, and the progress we’ve made this week marks a significant step towards that goal.
For a discussion on vitamin and mineral encapsulation featuring Ram Chaudhari, PhD & Chief Scientific Officer at Fortitech, read this post: http://www.nutraingredients.com/Research/Encapsulation-is-key-to-working-out-vitamin-solubility-issues
Although state-of-the-art rice protein manufacturing processes are proprietary (and we can’t discuss them due to NDA’s with ingredient suppliers), this paper describes the process at a high level: http://www.aaccnet.org/publications/cc/backissues/1966/Documents/Chem43_145.pdf