Reducing the cost of failure in waste innovation

In my last post "How to drive innovation in waste", I wrote about a foment of ideas being critical to innovation proceeding. From this foment, good ideas emerge and can then be developed.

Having written it, this model of innovation troubles me a little. Not because it doesn't work, but because it glides straight over something really important - and that is how to select ideas. But first, let me go back a step or two.

Publish then filter

I've been reading Here Comes Everybody by Clay Shirky lately, and the key point that sticks in my mind is the notion of "publish, then filter". Rather than filtering before publishing, the social model for innovation is to publish first, then let something like natural selection filter.

Now this model works where the cost of publishing is very low. It simply happens because it is cheaper (time, money, whatever) to publish than it is to evaluate. In fact, it is so cheap to publish that you can use the audience to filter for you.

Now how might you run a similar model in a waste processing business? Keep in mind that my aspiration for Garbologie is to make waste redundant. Everything becomes reincorporated into the productive economy rather than being siphoned off into large and growing waste streams at landfill, incineration or whatever.

Top-down and bottom-up

Attempting this through centralised data leads to a particular type of top-down approach. That approach is governed by thinking at a scale it can conceive of (large infrastructure), and is motivated by a drive to achieve targets whilst minimising environmental impacts. If you are in the business for long enough, this seems the only way to proceed.

Top down view, Empire State Building. Source: notmyholiday.com

Frankly, having been in the business for a while, this approach is paralysing and stupefying. It's paralysing because it requires massive capital investment, and this investment means that you have to be right. And stupefying because the only way to be right in waste is to go for the broadest, least value adding base.

My thinking is to attempt a bottom-up solution. It would seek to carve out particular materials from the waste stream and reprocess them. It would, necessarily, be formed of a network of loosely connected operations. Businesses even.

Bottom up view, Empire State Building. Source: Aquirkyblog.com

The challenge in bottom-up

The trouble is, how do you identify the materials to be carved out? And what is there to do with them? In building and sharing a scoop.it resource regarding innovation in waste management ("The Future of Waste"), I've come to realise that there is little shortage of ideas for processing a whole slew of materials.

There is no simple source of data that shows how much of a particular material is in the waste stream, and certainly nothing sufficiently fine-grained to enable decision making. There is no way to get help that isn't mediated by the same professionals who are locked into a "bigger is better" mindset. And the costs of failure are large - to test something you have to build it and hope they come.

So to me, the challenge is threefold:

1. Collect data. Lots of it, cutting across the problem on a range of different angle, and making that data as publicly available as possible
2. Be guided by the wisdom of non experts seeing the data with fresh eyes
3. Make the cost of failure low so that ideas can be attempted easily

My proposed solution

To me, many of the challenges are well on their way to being resolved by starting a business that is both broad and shallow. I have decided to do this through Tip/Shop, a site that is both: 

1. A customer oriented waste facility, clean, friendly and utterly focused on recycling as much as possible. As far as the customer is concerned, this is a place that is nice to visit, nothing like a waste facility; and
2. A test bed and incubator, gathering data on the materials coming in, seeking opinions from staff and customers on which materials could be next processes, and testing out micro-scale ventures to process them/

By having a steady stream of materials coming through, and by having sufficient space to undertake small scale trials, you can tinker until something works. That then develops into its own business, perhaps moves away to develop on its own trajectory. 

If you could do this as a myriad of employee centred ventures, each given support to come up with something that might eventually work and they can own a piece of, then you've created a business that is porous, interconnected and retains that "surface area" of a lot of people testing a lot of ideas on the ground. 

So this is how I propose to reduce the cost of failure in innovation. Provide a space where there is plenty of opportunity, and where experiments can be run regularly and easily. Make those experiments deeply personal to the people running them, and you might, just might, reduce the cost of failure.