Non-Profit Tech Blog

I had a quick phone interview with Joe Green and Keith Rarick from the Project Agape team. Joe Green is the head of Project Agape and Keith Rarick’s “title” is Lead Engineer. I first wanted to ask the questions that were submitted on the blog. However, I think you can find many of the business-oriented questions that readers submitted answered in a more direct format over at Micah Sifry’s interview with Joe Green over at the Personal Democracy Forum.

The biggest and most important question for Michelle Murrain was the business model for Project Agape. I have to say this is the most salient question for many nonprofits who want to know where their money is coming from and deciding on whether or not they should play in the Project Agape space.

Joe had two different answers. One of them is simply practical: that as a nonprofit they would never have been able to raise the hundreds of thousands of dollars needed to buy the servers they need to run Project Agape. His second answer was that the social mission goes far and beyond the costs that have been incurred by Project Agape so far. Apparently, Project Agape charges 4.5% on every donation which doesn’t even cover his costs. Actually, Joe had this to say during his interview with Micah Sifry:

JG: At this point we aren’t even coming close to covering our costs. Of the 4.5% that gets taken out, 3% goes to JustGive, and then of the 1.5% we receive we have to pay most of that to JustGive for the processing of the checks to the nonprofits.

If the cause is political in nature, Joe said the money gets sent to ABCPac if it’s a right-leaning or Republican cause or ActBlue if it’s a left-leaning or Democratic cause.

I had to press Joe on how Project Agape planned to stick around in the nonprofit sector as many nonprofit managers will be wanting to know how serious Project Agape was before they jumped in with both feet marketing their nonprofit with the Causes application. Joe eventually answered that they’ll probably do some advertising and perhaps partner with corporations regarding causes they care about. Even Joe admits that they are not currently focused on revenue opportunities right now. They seem to be interested in first growing the application and then worrying about revenue later. It’s much like a 1990s model of Internet development but they’ve only been around for a few months so let’s hope Joe figures this out by next year.

As a brief aside, Joe and Keith were proud to announce that their growth rate is now exponential meaning that it is rising as more users join Causes. I checked appaholic briefly and they’re right. They’ve spiked the last few days and are up to 39,000 new users per day or basically the entire Second Life audience twice over per day. In fact, they now have more users on Causes than the number of logged-in users to Second Life in the last sixty days. Obviously, Second Life and Causes are not competitors and Joe took pain to make sure that observation was noted. This is more of my own particular pet peeve about the love that Second Life gets in disproportion to its actual user base.

So I then asked the OTHER question that people are dying to know about Causes. How much money has it raised since its inception? Causes has raised… get ready for it… $300,000. Not bad for an application that didn’t exist a few months ago. They have yet to spend money on optimizing donations. Joe says they’ve got plenty of ideas but haven’t yet gotten around to defining what the idea giving experience is going to be.

I then asked Joe about how nonprofits should approach the Causes app and how they should market themselves there. He had an interesting argument that Causes is an experiment that is optimal when the organization is as concrete and transparent as possible. He pointed to the largest Cause which is called “Support Breast Cancer Research”. As of this writing, it has 868,940 members and raised $22,890. Apparently, the 501c3 that will be benefiting from the funds raised for this cause will be to a single hospital in the Boston area. Eric Ding, who started the cause, made a concrete pitch centered on the need for research to help stop breast cancer. As a result, the donors may not really care much about the Brigham & Womens Hospital Inc which is the cause receiving that $22,890. Instead, the donors care about the cause that they are donating to.

Of course, this is contrary to the way that many fundraisers operate. They worry about their nonprofit’s brand and worry about donor loyalty. In fact, Project Agape does NOT release donor lists to the nonprofits that are receiving money from Causes. However, they can find out who the donors are by looking at their Facebook profiles since each donor and fundraiser is listed. It’s a whole new ballgame for the nonprofit fundraisers. Will nonprofit development folks “get it” when people like Joe Green tell them that they can’t get donor lists and that brand loyalty is not particularly important? I hope so because that’s how Web 2.0 is siding on privacy issues for donors.

So for all you techies out there… here are, to me, the most fascinating details about Project Agape. Keith gave me the following details…

Previously, Project Agape was running about 14 servers to service the one million users they had back then. Believe it or not, they have not increased the number of boxes. In fact, they’ve actually reduced the number of application servers from 14 to 11 and made three of them utility boxes. The application is a lot more efficient now and their biggest win was when they rewrote the SQL queries they originally used to access the database. In fact, Keith is now clearly saying that the biggest bottleneck was never in the Rails application that they built but in the database bottleneck that resulted from SQL queries that were inefficient. Apparently, Project Agape was consulting with other Rails-based developers such as iLike and soon, Twitter, and sharing code and techniques that would allow them to scale more efficiently. In fact, Keith is hoping that they’ll be able to release this as an open-source toolset someday. (Release the code! I’d love to see it.)

One thing that I always thought was interesting about Project Agape was its reliance on FreeBSD. FreeBSD is no longer the OS of choice at Project Agape and they have moved over to Linux “purely for reasons of support and availability”, says Keith. They’re having a tough time hiring people with FreeBSD experience hence Linux it is. Keith was a little stingy in his praise of Linux, calling it the second-best OS out there but hey, it’s obviously good enough to run their app.

In fact, they’re going to start componentizing their codebase and the way it gets distributed over their little server farm so that it would be easier for them to scale going forward in the future. One interesting tidbit that Keith shared with me was that their current mysql server configuration should put them in good position to scale upwards to around twenty million users (or two a half times the entire community for Second Life hehe).

However, the Causes team is not exactly sitting on their laurels. They recently added a “Media Board” to a Cause. This allows users to post media related to that cause up to the rest of the users who have joined that cause. There’s finally a serious incentive for nonprofits to post publicly available audio-visual media somewhere on the Web in the hopes that Facebook users can pick it up and post it to a Cause. Obviously, I’m enamored of what the Causes app is doing. I even admitted that I was a Project Agape fanboy to Joe and Keith, thus obliterating any sense of journalistic neutrality that I ever had.

Be that as it may, I strongly urge nonprofits to seriously consider trying out Causes especially if that nonprofit is not as well established. Who knows? One key proponent of your nonprofit could be that informal marketing genius who could vault you to the top of the heap in Causes. Utilize your offline social networks and get them to support you on Facebook. Do it now before everybody else figures out a good approach thus raising a barrier to your nonprofit’s entry.