The Tuna and Beef Commoditization Play
Find the two engines of value, mass-produce them, and slap a licensed character on top.
- Difficulty
- Advanced
- Time to result
- ~months to results
- Steps
- 4
- Confidence
- 90%
O'Leary learned as an MBA-summer product manager on Miss Mew cat food that all cat food reduces to two protein engines — beef renderings and tuna — and everything else is just marketing flavor names, because cats don't have credit cards, their owners buy the flavor. Years later at The Learning Company he applied the same insight: 80% of the business was advancing math and reading scores, so he built two core code engines (math and reading) and licensed characters like Big Bird and Barbie to sit on top. He fired two-thirds of the expensive coders, cut prices from $99 to $12.99, and multiplied volume 5x. The move dramatically raised margins and lowered cost of capital, letting them acquire competitors with stock.
Origin
Between his first and second MBA years, O'Leary worked a summer for a Dutch product manager on Miss Mew cat food, who taught him that cat food has only two engines of protein — beef renderings and tuna — and everything else is marketing. A decade later, sitting with his logistics partner at 6pm in Boston, O'Leary realized The Learning Company's business was really just two engines — math and reading — with licensed characters on top, and the 'magic moment' reshaped the whole company.
Core principles
- 01Most products reduce to a small number of core value 'engines' — everything else is marketing.
- 02Identify your two (or few) real engines of value and build reusable core assets around them.
- 03Everything layered on top — brand, character, packaging — can be licensed or swapped cheaply.
- 04Commoditizing the expensive core lets you cut cost, drop price, and multiply volume.
- 05Higher margins and lower cost of capital let you out-acquire and out-price competitors.
How to run it
- 1
Deconstruct the product to its engines
Strip the product down to the few irreducible sources of the value customers actually pay for — the 'tuna and beef.' For The Learning Company it was advancing math and reading scores.
Pro tip Ask what the customer is really buying versus what merely decorates it — parents bought score improvement, kids only wanted the character.
- 2
Separate core from surface
Split the durable core engines from the interchangeable surface layer (brand, character, flavor name) that can be swapped without the customer caring.
- 3
Consolidate to reusable core teams
Build a small number of reusable core teams or assets instead of many bespoke ones. O'Leary consolidated to two code teams — math and reading — and fired two-thirds of the expensive coders.
Watch out Expect internal and industry shock — no one had done this in consumer software before and it made the Wall Street Journal.
- 4
License the surface and scale
License recognizable brands or characters to sit on top of the core, then cut price and multiply volume on the fatter margin. They dropped software from $99 to $12.99 and grew volume 5x.
Pro tip Fatter margins lower your cost of capital, which lets you use your stock to acquire competitors.
In the wild
O'Leary reduced the company to two engines (math and reading core code), licensed characters like Big Bird and Barbie on top, fired two-thirds of coders, dropped prices from $99 to $12.99, and increased volume 5x while staying profitable.
→ Margins jumped, cost of capital dropped, and the company used its stock to make hostile acquisitions of competitors, selling hundreds of millions of copies worldwide before its $4.2 billion sale.
As a summer product manager, O'Leary learned that all cat food is built on just two protein engines — beef renderings and tuna — with new 'flavors' being pure marketing aimed at owners, since cats don't buy their own food.
→ The insight became a mental model he inventoried for a decade and later deployed to transform The Learning Company.
Common mistakes
Paying premium cost for a swappable layer
Building bespoke, expensive production for what is really an interchangeable surface layer leaves enormous margin on the table.
Refusing to commoditize out of pride in the craft
The acquiring company later refused to run titles off the shared engines, insisting on two-year studies, and left hundreds of millions in profit on the table.
Is it for you?
Best for
Product companies with high production cost and interchangeable, brand-driven demand.
Not ideal for
Products where the deep technical differentiation itself is what customers buy.
From the transcript
“There's only two engines of protein. It's beef renderings and chicken faces... or it's tuna from the sea of Japan.”
“We can license the character. We can fire two-thirds of the coders, which are really expensive, and just use the two tuna and beef.”
From the episode
Kevin O'Leary: The Game-Changing Habits That Set Top Entrepreneurs Apart
Kevin O'Leary