From the Archive

What Is The Real-Time Supply Of Human Services?

 At Markets For Good, we explore a wide range of themes on the use and sharing of data, but not for the sake of discussion itself: Instead, we want to discover what’s going on and how to connect, align, and accelerate what’s working. Think of this as upgrading the information infrastructure in the social sector. With that in mind, we have seen that standards, protocols, and taxonomies, etc. comprise a theme (common language, methods, and categorizations) that has solid momentum, not only in discussion and needed advocacy, but also in practice.

Erine Gray, founder of Aunt Bertha, is a case-in-point as he walks us from the food pantry to an open taxonomy for classifying and counting human services. The idea? Making it much easier for people to find and use the services they need.

What if we knew the number of food pantries currently open – right this minute – in Tuscaloosa, AL? What if researchers could compare that figure with the number of food pantries in Bloomington, IN? (Both are similar-sized towns.)

To realize this vision, we need a clear understanding of what constitutes a food pantry. That can be more complicated than you think. But it’s vital.

At we focus on connecting people in need with government and charitable programs through the internet and mobile devices. We launched our service in Austin, TX in June 2011 and this year launched in every zip code in Texas (which covers roughly 1/12 of the American Population). Our goal is roll out our service to the whole country.

We quickly learned that to achieve our mission of making human services information accessible to everyone, we must agree on a common language so we know the simplest thing: how to count these programs.

Existing Taxonomies

Ever shopped for shoes on It’s pretty simple. You search by keyword or by shop by category. There are shoes for men, shoes for women. There are sneakers, high heels, boots, etc. That’s a taxonomy: a way of categorizing shoes.

At Aunt Bertha, we started looking for existing taxonomies that allow us to classify human service programs. We recently reviewed the AIRS/211 LA County taxonomy and it is perhaps the most thorough of those we found: but we feel it is deeply flawed for two reasons.

It is a hierarchical taxonomy: The problem with a hierarchical taxonomy is that you have to know the route: if you don’t, it may take you a while to get to where you want to go. If you’re looking for a food pantry for people with HIV or AIDS, you may need at least seven clicks to find that information. If hierarchical taxonomies are too rigid, they can make users feel as if they’re navigating a corn maze.

It costs money to use: According to the AIRS/211 LA County website, the taxonomy subscription fee is between $200 and $650 per year.

Let’s Do Something Different

When we were deciding how to classify our programs, we chose a different route:

We wanted to develop a taxonomy guided by two important principles: It should be multifaceted, and it should be open.

Let’s go back to our Zappos example. We’ve decided we want men’s boots. But we’re not sure which color or style.

Should we drill down (click through) every possible style of men’s boots until we find the one we want? What about color? Would we also list the colors under every category and style of boots? Of course not.

Zappos let’s you navigate to Men’s Shoes >> Boots, using a partially hierarchical taxonomy, yes, but also allowing you to add other facets including colors and styles. When you select those, magic happens: you get a list of all the boots that match the color and style you’re interested in.

Let’s translate that to human service programs.

If you’re looking for a food pantry, we think you should click Food >> Food Pantry. If there are quite a few of them in the list (as there are quite a few boots on Zappos), then we think you should be able to narrow them down based on your own situation. We call these “situational tags.”

If you want a food pantry that caters to people with HIV or AIDS that should be an attribute just like color or style is an attribute of our boot search. This allows you to narrow down your search in seconds.

Think about how this would apply to people in need that suffer from rare illnesses and are trying to find resources provided by charities and governments that help with those illnesses.

“OPEN. (This a no-brainer.)

In fact, we figured we’d get the ball rolling by releasing our draft taxonomy with a Creative Commons Attribution License. We want you to use it and to help us make it better. And we’re willing to do the work that it takes to help coordinate that effort.

If we agree to an open and free taxonomy to classify human service programs then we can finally begin to count the services that are available by location.

This is the first step to knowing the number of open food pantries – right now – in Tuscaloosa, AL.

Curator’s Note: Think on the scale of the entire social sector (for-profit, nonprofit, impact investors, public sector): Tell us what have you seen that is working in taxonomies and standardization, or just access to data.

Here’s another idea, for our private sector readers: It is clear that we are talking about Customer Service. What is working for you to make it easy for your customers to find and use the information they need? Let us know here in the comments. And, for all, if there is a particular initiative we should know about, you can also share it here or email me directly: Thank you!


  1. Clive Jones says:

    Just to give a counter-view as a long time user of the existing Taxonomy of Human Services, in particular regarding its two “flaws”:

    1. Yes, the Taxonomy is organized as a hierarchy but it is not intended to be searched or used as a hierarchy (except by indexing experts). If you want to search all of the terms that involve the word “Food”, you just have to type that single word to see all of the choices involved. The Taxonomy also has tens of thousands of Use References and See Also terms that allow you to find terms more easily using natural language and also to jump across hierarchical branches. In fact, most users do not know that there is a hierarchy.

    2. Probably 98% of users pay the $200 a year license. And they do that to ensure that the system is properly maintained. Thousands of small and major changes are made every year and all users are encouraged to feed in suggestions.

    The Taxonomy of Human Services is supported by thousands of nonprofit and government organizations across the United States and Canada who see the value of a common language that allows for consistent reporting and use across jurisdictions and human services silos. It allows an agency in New Jersey to provide support to an agency handling a disaster in Louisiana because they know the indexing structure of their database and can provide reporting information back again on what people are asking for. And it allows those agencies to reverse supporting roles during an event such as Hurricane Sandy.

    I find it odd that when there is so much disparity in the human services system, that the Taxonomy of Human Services which has flourished from a nonprofit grassroots initiative over 30 years ago and has now taken such firm root within the human services infrastructure, is worth the effort of developing a competitive system.

    The Information and Referral sector would no doubt welcome the energy and technical alternatives that Mr Gray is capable of developing, with the goal of improving the existing system rather than creating new systems.

    1. Erine Gray says:

      We appreciate you taking the time to give us your feedback. When we launched the Open Eligibility Project our goal was to provide a platform for conversations like this and the more perspectives we can get the better. We are working on a big problem and it will take diverse views and experiences to solve. Thank you for helping us start this conversation.

      We understand and respect the commitment of everyone in the Information & Referrals sector. In times of natural disaster and during every day hustle and bustle there is seemingly endless work to be done. We’re lucky to have passionate people dedicated to solving these problems.

      We studied the AIRS taxonomy (and the implementations) and believe there is a lot to be learned. We decided to take a more targeted, simpler approach. Our goal is to say things as concisely as possible, not in as many ways as possible.

      Just recently, according to the AIRS website, a new term was added to the taxonomy: Personalized License Plates. While, we can appreciate AIRS comprehensive approach we must ask: where is the tradeoff between comprehensiveness and usability? It only takes a little imagination to understand why we need to dedicate expensive training and certification processes to confidently point someone in need in the right direction.

      America was built on standards from railroad tracks, to door frames, to the World Wide Web. The best standards limit our ability to choose some things in order to broaden others. The call center model of providing 24-hour help has served us well during times of crisis. The reality is, as a society, we can’t afford enough call centers to meet the rising demand for human services. We must do things differently in order to keep up and it’s going to take creative thinking on all fronts (both technology and standards).

      Our vision is make human service information accessible to people and programs through simple, intuitive web and mobile software. We believe simplifying the taxonomy will open up a wealth of innovation in the Information and Referral sector. This means more people in need find what they’re looking for and non-profits can focus more energy on direct services.

      1. Clive Jones says:

        Not disagreeing with you on much of your reply …

        Regarding the example of Personalized Number Plates … somebody in some community asked for this to be made a term and so it was. However, folks can ‘switch off” any term in the Taxonomy and never see it and it would never come up as a result of a search unless someone switched in on and then used it ,,,,

        I don’t think anyone working in I&R has a vision of “Less accessibility” …

Comments are closed.