Innovation.

Co-Innovation Risk

5 min read·Apply by lunch

The question

Take the thing you plan to launch next year and count everyone outside your business who has to do something new before a customer gets any value from it at all — the supplier who has to make a part that doesn't exist yet, the installer who has to learn a procedure, the regulator who has to write a rule, the retailer who has to stock something nobody has asked for — and then ask what each of them would privately put their own odds at. You've costed your own risk carefully. The rest of the chain sits in the plan as one word: partners.

The idea

Ron Adner was teaching strategy at INSEAD when the Harvard Business Review published Match Your Innovation Strategy to Your Innovation Ecosystem in April 2006, where he named three risks: initiative, interdependence and integration. He rebuilt the argument as The Wide Lens in 2012, by then at Dartmouth's Tuck School, and renamed them execution, co-innovation and adoption chain — the names they travel under now. Execution risk is what your project plan is made of: can we build it, on time, to spec. Co-innovation risk is how far your success waits on other innovations succeeding first, elsewhere, on somebody else's schedule. Adoption chain risk is how many intermediaries must take the thing up before an end customer sees any value. Michelin's run-flat tyre carried both of the last two. The tyre talked to the car's electronics, so it fitted only vehicles designed to accommodate it, and Michelin waited on a willing manufacturer's design window — three to four years before volume production. Behind the carmakers sat dealers who had to support the system and repair shops that had to buy service equipment. Nine years after its introduction, the article reports, it was standard on a handful of car models. High-definition television had the same shape, over decades. Then do the arithmetic. Give yourself a generous nine in ten, and four partners nine in ten each. Nine-tenths to the fifth is fifty-nine per cent. Nobody in the room ever multiplies.

The framework

01
Draw the chain to the end customer, and put a name on every link.
Start at the person who eventually pays and work backwards to your desk, writing down every party the thing has to pass through: component suppliers, standards bodies, installers, distributors, retailers, the IT team inside the buyer, whoever signs off a regulation. Then strike out everyone whose part is routine. What's left is the real chain — the links that have to do something they have never done before. Most plans stop at the first tier and file the rest under partners, which is how a dozen separate dependencies end up held in your head as one comfortable noun.
02
Co-innovation risk is the part that gets invented somewhere else.
This is the first risk your project plan doesn't contain, and the test is narrow: does something that doesn't exist yet have to exist, built by somebody who doesn't report to you, before your thing works at all? A battery at a density nobody has shipped. A standard still being argued over in a committee. An integration on another company's roadmap, sitting below their own priorities. You cannot manage this with a schedule, because it is not your project, and it runs on a board meeting you will never attend. Assumption Mapping plotted the beliefs inside your own walls; these are held by people who don't work for you.
03
The middlemen who gain nothing are where it usually dies.
The adoption chain is the second, and it asks something narrower than it looks. Not what a distributor's margin buys, and not who finally captures the value — both come round elsewhere. Only this: will a party who has to do something new actually do it? So go link by link and write two lines, what this party gains and what it gives up. Michelin's repair shops weren't obstructive; they were asked to buy service equipment for a product hardly anyone owned. One negative line anywhere stops the chain, and the link that gains least is rarely the one you've been calling the partner.
04
Multiply the odds, then shorten the chain rather than the odds.
Put an honest number against each link — your own included, and be less generous with yourself than you'd like — and multiply. The result is usually so unpleasant that people go back and inflate the inputs, which is the wrong move: the number is the finding. You have three real responses. Take a dependency inside your own walls and build it yourself. Cut a link out by changing the design so the thing works without it. Or start with the shortest chain you can find — one setting, one geography, one buyer type where fewer parties have to move — and earn the right to the longer one later.

Members only · 4 principles + template + AI mentor

364 more frameworks are waiting.

Co-Innovation Risk is Day 1 of 365. One framework every morning for a year — across strategy, sales, negotiation, leadership, and more.

$1 /day

Billed $365/yr · cancel any time

365 frameworks

12 topics · templates · AI mentor

Reserve your place — $365/yr →

One payment a year · the whole library · no auto-renewal.

More Innovation frameworks

All 24 innovation frameworks →