The Bees’ Knees

[6 May 2015 Note: The article on Bees’ Quadratic Voting mentioned below has drawn some stinging criticism from a certain Warren D. Smith. Knowing nothing about bees, I have chosen to flag the contention over these ideas.]

When it comes to product development, I have long felt uncomfortable about how most organisations go about choosing which products, and features within products, to prioritise.

The most common approach – HiPPO – and indeed many other approaches, only seem serve the egos of those involved. Rarely does an approach come even close to selecting the most promising product or feature, with respect to the goals, ambitions and yes, shared purpose, of the organisation.

Software and product teams rarely if ever get to choose what they work on. So the key decisions have been made even before the work flows into their remit. Little wonder then the frustrations of such folks, who often find themselves working on stuff in which they have little faith, from a commercial standpoint. And little wonder, too, they these folks find it, ahem, a challenge to feel enthusiastic and committed to their work.

Experiments, MVPs, A/B Testing, Etc.

We have evolved ways to address this issue, mostly centred around the idea of trying things out and seeing what works. This causes me discomfort, too. Not least because in involve some investment in work (sunk costs), and inevitable waste.

And it still doesn’t address the up-front question of just WHAT to experiment on, or try out.

Of course, if your organisation is old-school, then it’ll have specific people – the single, wringable necks – whose job it is to make these kinds of decisions. Product Managers, for example.

Yet some companies are finding that crowdsourcing such decisions, whether using an internal crowd (a.k.. Fellowship), or involving outside folks too, can lead to much more effective decision-making.

Quadratic Voting

This recent article about how bees make decisions [link updated 22/Apr/2022 to use a Wayback Machine reference] points the way to a more effective approach to such crowdsourcing. In essence, we might imagine a prioritisation approach that uses many heads, yet allows for an efficient outcome:

“All type-symmetric Bayes-Nash equilibria of an independent private values Quadratic Voting game converge to an efficient price-taking outcome as the population size grows large.”

~ Steven P. Lalley

Mathematically, Quadratic Voting is the only pricing rule that gives individuals an incentive truthfully to report their preferences.

Maybe you might like to experiment with using Quadratic Voting (QV) in how your organisation decides which product and feature to work on next? Or, God Forbid, if you’re attending to folks’ needs, then QV may help you all agree on which needs come first.

And maybe it’s a perfect complement to the Advice Process as seen and reported by Frederic Laloux?

– Bob

Further Reading

The Principles Of Product Development Flow ~ Donald “The Don” Reinertsen