I first heard about the Donation Dashboard, created as part of a research project by the Berkeley Centre for New Media, through my US-based former colleague Jeff Brookes’ Donor Power blog, and I’ve been meaning to mention it for a while. Then it struck me that it could be of particular interest to any American donors (or donors anywhere I guess, but the organisations included are all US non-profits) wanting to get the new year off to a good start by refreshing the portfolio of charities they support.
Donation Dashboard uses a ‘collaborative filtering algorithm’ to recommend a portfolio of non-profits to support, including what percentage of the amount you wish to donate should go to each, based on an analysis of your stated interests combined with those of everyone else who has taken part in the research project (not dissimilar to Amazon’s ‘people who bought that also bought this’ recommendation approach).
To generate a personal giving portfolio, you’re presented with an initial set of 15 non-profits, with summary information for each, which you rate on a sliding scale from ‘Not Interested’ to ‘Very Interested’. The site then generates a statistical model of your giving preferences and a recommended portfolio based on this, which you can then refine through the rating of more non-profits.
My portfolio is shown above. No surprise that it includes Kiva, which I’m a big fan of, and NPR/PBS which I used a lot when living in the States. But One Laptop Per Child was an unexpected yet interesting, and personally relevant, recommendation.
It’s a fun tool to play with and if you’re the type of person who plans their charitable giving in terms of a balanced portfolio then it may well introduce you to some new organisations doing work that’s of real interest to you. However, I’m not sure how many donors actually approach donation decisions quite as rationally as that?
Where I could see this type of collaborative recommendation approach being more effectively used is perhaps in online ‘charity supermarket’ sites like GlobalGiving and Play it Forward, where potential donors are offered the opportunity to choose projects to support from a wide range of different organisations. There the diverse range of choices might be more easily, and interestingly, navigated through some form of collaborative recommendation approach – rather than just via the usual drop down menu choices.