Concurrent Production

I've talked before about how Product Management is becoming an increasingly pivotal function in many digitally facing organisations. Reading this Quora answer about some 'best-in-class' product management processes led me to Marty Cagan's thought about 'Product Discovery Teams' which he explains in the short video above.

In essence, the Product Discovery Team are the core team working on product development, and comprise the product manager (responsible for functionality), the lead interaction designer (looking after usability) and the lead developer (focused on feasability). Having each of these present when vision is elaborated and new products designed ensures that the right minds are there from the beginning and reduces the risk of work being started on features that can't work.

This reminded me of a point made by Steven Johnson in his excellent book Where Good Ideas Come From about how Apple work - using a process they call concurrent or parallel production. Traditional models of development see designers passing an initial look and feature set to the engineers (who find ways to make it actually work), and then on to manufacturing (who work out how to produce it), and then sales and marketing (who decide how to sell it). As Johnson says, the ubiquity of this model comes from the fact that it works very well in situations where efficiency is the most important consideration but it kills creativity since original ideas get chipped away at at every stage.

In contrast, Apple's development model involves all the groups (design, engineering, manufacturing, sales) meeting continuously throughout the process in an approach that is "messier and more chaotic at the beginning", but which avoids great ideas being slowly hollowed out and becoming ghosts of themselves: "The process is noisy and involves far more open-ended and contentious meetings than traditional production cycles— and far more dialogue between people versed in different disciplines, with all the translation difficulties that creates".

The idea of concurrent working rather than departments (or indeed different types of agency) handing off projects and briefs to each other is a challenging one for agencies to think about. In this Google 'Agile Creativity' hangout (which also features Greg Anderson of BBH), John Boiler (founder of 72 and Sunny) talks about how they work to actively break down silos within the agency and how when a brief comes in, they discuss it as a large, multi-discplinary group ("not knowing necessarily where the centre of gravity for the solution will come from") and use a 'work wall' as the centrepiece for a process that takes account of the benefits of authorship and specialist expertise, but also enables them to work more concurrently and "get everyone around the problem" to work far more iteratively.

When I was researching the The Progression Of Agency Value report earlier this year, looking at how agencies were developing their use of technology and the structures, skills and behaviours around that, there were a few technology based agencies who I spoke to who seemed to be working in interesting ways. Teams that typically incorporated a key triumvirate of talent - business/strategy, creative/design, and technology - would work concurrently on developing solutions. It was a structure that allowed for consistency and commitment to long-term projects and clients, but also flexibility as talent was encouraged to contribute to other relevant work or even unrelated projects (perhaps in a Google-like 70, 20, 10 way). It makes a lot of sense to me that in response to such a rapidly changing communications environment we should consider not just the how the work is evolving, but more fundamental changes in how we produce that work.

HT Gerald Breatnach for the hangout link

Comments

[Charles Frith]

Human like might be a more accessible word than concurrent production. Departmentalisation, hierarchy, explicit command and control are all artificial constructs to dehumanise. We're tribal hunter gathers. That's it.

It's about Jazz not orchestras. Bazaars not cathedrals. This all.

[neilperkin]

"It's about Jazz not orchestras. Bazaars not cathedrals."
Love that

[Don]

Concurrent engineering is a work methodology based on the parallelization of ... the basic conceptual design can be altered before actual physical production ...