To the hackers and painters,
To those who believe well thought out sequences of 0’s and 1’s can change the world,
And to those interested in transforming the nonprofit space,
This post is going to draft a big problem in the nonprofit arena, and provide what we think are potential steps to solve it.
The short: the current platforms used to discover and evaluate charities are great in some ways but do a poor job in helping people find charities that address causes they care about. As we’ll outline below, we believe there are concrete lean steps to build a product to classify and connect nonprofit knowledge across the web that will provide greater transparency, accountability, and understanding to both prospective donors and nonprofits. Turning the current chaos into catalytic power for greater social impact.
Sounds like something of interest? Join us in building the nonprofit discovery platform of the future. Today.
The Background/Problem
Good St. is a microdonation platform where we send out a daily email to our subscribers who can donate to one of two featured charities addressing the cause (example here). That means we gotta find and vet 730 charities a year (and we look at thousands more we don’t end up featuring). You can read more about how we vet in a previous post, but in short it’s a lot of work for an all-volunteer nonprofit and it means hours at nights, on weekends, and during vacation. Imagine you’re just trying to find out a nice charity to donate to that fits your interest — all these platforms can add to the confusion. It’s hard to blame anyone for falling prey to nonprofit scams when the space is a tangled mess.
The current platforms are helpful to find and evaluate charities but have much to be desired with poor search capability, outdated filters, misleading information, and paywalls.
Take Charity Navigator, one of the premier platforms in the space. They provide an amazing service, and there are many useful features. But if you are searching for a particular issue that you want donate to, it’s search function is really poor. Let’s say you are trying to find an organization supports heart disease prevention. Search “heart disease” and this is what comes up:
7 charities in all, and none are actually rated by Charity Navigator (you’re on your own to comb their tax returns, do all the needed vetting, and you still won’t necessarily know if you should donate to them). That’s crazy — amazing nonprofits addressing heart disease don’t see the light of day. The American Heart Association doesn’t even come up in that search, you have to specify “American Heart Association” (or a variation of) to find it.
Or let’s say you want a charity that provides healthcare to those in Uganda. This is what you get when searching “uganda healthcare”:
Only two charities come up, and they don’t even address the issue.
Charity Navigator has some really great tools that we use. But when it comes to finding charities addressing a particular cause, it can be pretty difficult to navigate.
GuideStar is another major platform to find charities. It’s served us well at Good St., and we are grateful for the site. But there’s room for improvement. While their search function captures plenty of nonprofits, sometimes it can get overwhelming. Search “heart disease” and you get this:
875 results. Few people have the time to comb through it all and figure out which one you should support (which is what we do manually on Good St. on a daily basis).
Or suppose you are looking for a charity that addresses AIDS. So you search, “AIDS”:
17,258 charities come up. Basically you are also getting organizations with the word “aid” in it. 5/6 of the first results have nothing to do with AIDS itself. (To actually find an AIDS charity you gotta specify “HIV AIDS”.) Like Charity Navigator, there are ways to filter by organization size/rating and other advanced queries. Which is awesome. It helps narrow down the nonprofits a lot. But at the end of the day we are left with a lot of work to sift through.
GuideStar has another problem we often face: paywalls. Let’s say you want to filter by organizations that are defunct — reasonable request, right? So you go to the sidebar on the site and are about to check off “exclude defunct or merged organizations.”
But once you hover over to check that box you get a little notice —
Bummer. You can’t actually filter that criteria without upgrading.
But I mean it can’t be that expensive of a subscription, you’d think. They’re a nonprofit platform for the public, right?
Another bummer. You have to pay $125 a month (or $250 month to month) to do things like filter out organizations that are defunct.
We aren’t here to blame them for trying to make a living selling things, but having it more accessible to the average person would be nice. Like how we got a nice email from GuideStar about what sounded like an awesome report with really insightful and helpful information:
Looked great, and we tried to “learn more”. But the only thing we actually learned after clicking the link was you that you can’t learn more unless you have $374 to spare (and $999 for multiple people / teams).
All this said, however buggy the existing platforms, we are thankful for their tremendous efforts and appreciate all the people who dedicated their lives to the work. We do not want to replace them (to the contrary want to work with them), we just need a better solution layered on top of it for people who want to find good charities they’ll care about. We chose to poke at Charity Navigator’s and GuideStar’s room for growth over the other platforms we use because they are still very helpful. It’s by standing on their shoulders that we’ve been able to come up with what we think is an improvement layered on top. Which brings us to the next point: what we want to do to fix these issues.
The goal
We want to put the tricks/tips/insights we’ve learned into building an overarching nonprofit discovery and vetting platform. A real help to not only us but the public at large. Free and open for anyone. No strings attached. If someone wants to know how they can contribute to cancer research, help the homeless in their neighborhood, or aid starving children—we want a platform that provides reasonable answers on what you’d actually be giving to and empowered with knowledge that matters to you.
Let’s build a platform to bring unprecedented power and clarity to the nonprofit space. If done right, this can be a massive step for how everyone from federal/state governments, private foundations, NGOs, businesses, and individuals find, vet, and more broadly relate to nonprofits.
The Rough Roadmap
A sketched roadmap is below to give you an idea of what we’re talking about. Step 1 is the minimal viable product (MVP). If only that was built, it would still help us — and we think the public — to a great degree. A bunch of the steps can be done out of order too.
- 1st step (MVP): Collecting information from GuideStar, Charity Navigator, and Give.org on their respective rating for each platform and secondary available content such as email contact, website, etc. Then building a basic frontend whether through google sheets or otherwise where the database can be queried to pull up charities that meet basic filtering criteria (ex. meets criteria of “3+ stars on Charity Navigator, is BBB accredited & has a gold rating on GuideStar).
How this helps: Allows people to find charities that rise to the top of each platform (reducing reliance on one site’s rating quirks) and saves a lot of time cross-referencing ratings between platforms. - 2nd step: Going wide by gathering key information from other platforms including GreatNonprofits, Yelp, Glassdoor, Facebook, foundation grantee lists, Twitter, organization LinkedIn profiles, staff LinkedIn pages, Google News hits, legal suits database, and more. Also in going wide considering international organizations and their respective platforms (UK and Canada).
How this helps: The other non-vetting platforms can provide a lot of powerful information on a nonprofit and serves as a way to flag scams. Something as simple as flagging a nonprofit that has less than a certain ratio of people-associated on LinkedIn to its annual-revenue for example. - 3rd step: Going deep by starting to classify charities whether and how they are either religiously or politically affiliated based on public data (such as through an estimation of the number of hot button keywords associated with the given topic). Also, beginning to auto-classify the type of nonprofit issue(s) addressed (“environmental”, “domestic”, etc.).
How this helps: Whether a charity is religiously or politically affiliated is very important to many people. And starting to auto-classify will be important for helping navigate people’s preferences. - 4th step: Introducing a basic visualization layer so that the charities can be viewed in context of the entire nonprofit space.
How this helps: Basic text readouts are great, but visuals can do a lot of incredible work in developing a more informed perspective (like seeing a scatter plot of an organization’s fundraising expense % compared to other nonprofits). - 5th step: Incorporating information from the nonprofits themselves, essentially allowing nonprofits to contribute to their own ‘wiki’ profile.
How this helps: The nonprofits can help self-police any mistakes on their own profile (with a public change-log for transparency), and add any metrics or information they want to share with the public. - 6th Step: A more advanced method of classification through weighing keyword searches. So if in searching for “heart disease” the term comes up on the organization’s website it carries more weight to classify the nonprofit as addressing heart disease than if the keyword is associated with a news article or online review.
How this helps: Reduces errors and friction while searching. - 7th step: Further advanced search methods that would allow someone to easily filter and find a “small charity” that’s “non-political” addressing “healthcare” for those in “Uganda”.
How this helps: The more tailored the discovery/search can be, the easier for users to find charities they’ll love. - 8th step: Introducing some machine learning to allow the classifications to get ‘smarter’ over time.
How this helps: Classifying charities will run into some issues. Being able to learn from the distinctions based on un/supervised feedback can go far in reducing errors. - 9th step: Building a direct payment and volunteering engine. Whether donations are sent direct from us or funneled through a partner organization. So people that find an organization they like can choose to donate or volunteer (the latter can be just an email automatically sent on behalf of the user to the organization or some other integration).
How this helps: Further encourages donations and volunteerism. - 10th step: Introducing advanced NLP so you can just ask what you want and we pull up the charities for you.
How this helps: Making it as super easy as possible for the public to find and give to charities they love. - 11th step: Being proactive through contextual micro-moment notifications. So if a person who opted-in visits a new city, we can automatically serve them up local charities they can donate to. Or if the person watches a National Geographic documentary, we can offer a nonprofit that helps conserve the habitats of the animals in the documentary description.
How this helps: Not waiting for people to reach out, but catering to them directly to integrate more opportunities to do good into their lives. - 12th+ step: Let’s hear your ideas!
We need you
This is all an ongoing conversation and we want you to join us. We aren’t gonna bother listing qualifications — if you’re reading this and know what needs to get done, that’s it. And beyond having the skill set, it goes without saying we need someone relentlessly resourceful, with deep integrity, and with a passion for concrete impact. A hacker and hustler for social good.
If you aren’t ready to join this initiative, or if you aren’t a programmer but still find what we are trying to do interesting enough — be sure to drop us a note and we’ll keep you in the loop.
Email hey@goodst.org to get involved.
— Joe Benun
Team Good St.
