(If you are wondering why…the hippodrum costs $90 which is ridiculous for a plastic garbage can with a stick through it; not to mention that it doesn’t work half the time because most of the paths people walk on are not flat, and a less than full drum is very uncomfortable to roll around; the playpump desperately needs to update their model…again costs simply too much to install and the design is horribly outdated).

I take your point with regard to business. You’re right that ideas are tested, results are measured, and so on. I guess that in business sometimes these things are more easily measurable: sales, revenue, profits, etc. Your idea to try to find good measurements for the LHF venture is really good.

My point about business was maybe more low-level, however. A lot of work goes in to test over-arching strategies and big product launches, but successful organisations are those that allow people at all levels to try to make positive contributions where they see a need. Actually, your idea is just that, and so I’m a supporter, with the caveat that people need to be able to act quickly and flexibly, and so everything needs to be done to ensure that that’s possible.

1. In my experience, donor priorities cover big overarching goals like reducing poverty, improving access to health care, or improving the environment for microfinance. Fitting one small solution into that kind of comprehensive effort is difficult.

2. That is a great question, and not one I have an easy answer to. In some cases, a solution has been identified by a local community but they lack the resources to implement. For example, farmers everywhere will tell you they need enclosed irrigation.

3. I don’t think that documentation or choosing a control community would slow down implementation. Choosing a control could be done using existing demographic and geographic data. Documentation doesn’t need to be arduous, either. Plenty of rapid assessment tools exist. Well-chosen evaluation does not impede implementation, it improves it.

I disagree with the premise of your business example. Business do market assessments before they introduce new products or services, and they have metrics to make sure that their product/service is actually being purchased. No one would run a business with a way of knowing whether or not they were making money, growing, capturing new customers and so on. That’s exactly what monitoring and evaluation is.

1. Why are these not part of donor priorities in the first place? They really should be.

2. A bit more sceptical, but the obvious (Easterly-inspired?) question is, if what people needed was rolling water containers wouldn’t they have invented them a long time ago? And if they are so cheap, why aren’t people willing to pay for them? I guess this question is quite specific, but can be applied more broadly…why are concepts we think of as quick wins not thought of as such by poor people? Or maybe I’m wrong, and they are, but there is some other problem…

3. I think that documenting “everything to pieces” and using a “paired control community” risks the viability of the idea. Its merit is in being quick and flexible. Covering it with so many other things risks slowing it down. As an example, in business (and I realise I may be tiring you with comparisons to the business world…) you simply wouldn’t see that kind of approach. You would come up with an idea, run it by a few people, try it out, and then either run with it or bin it. There would be some evaluation, of course, but nothing to seriously impede the idea being implemented.

Chris – I keep forgetting open license isn’t an automatic idea…

I’m all about low hanging fruit, and documenting everything to the point where replication is possible, but I _really_ think we have to be very, very precise in selecting the technologies to scale.

But apart from those quibbles, with it 100%.

I’d be inclined to choose some different technologies – e.g. treadle pumps are probably more cost effective than play pumps. But we can let the LHF people and the target communities decide the details.