Walking Away from Geographic Data

I recently attended the 2009 annual conference of The Funders' Network for Smart Growth and Livable Communities (TFN) in Miami. It was a well run, thought provoking couple of days featuring great speakers, good panel and group discussions, and lots of very smart, committed people. Almost every session presented - responsible development, growth in South Florida, the foreclosure crisis, how the stimulus money might be used for smart growth, transportation oriented development in many forms, and how climate change and sustainability issues affect all of the above – had a strong geographic focus. Since the participants were almost all foundation community development or environment program officers, professional planners, local government officials, or representatives of smart growth oriented non-profits, every session was full of experts on the topics covered who really knew the issues in their geographic areas.

In every session, I witnessed fascinating exchanges, real learning, “AHA!” moments, and networking. I also noticed that we, as a group, were walking away from a lot of potentially useful data on topics we'd just spent 90 minutes discussing, that we care about deeply, and on which we spend our working lives. Like all organizations, TFN has its leaders, long-time participants, natural networkers, and subject area experts. It is through these “thought leaders” that information seems to be exchanged, partnerships formed, and agendas developed. While TFN is very inclusive and goes to great lengths to make new participants feel welcome, realistically not everyone speaks up and even the most gregarious and committed of the networkers can't begin to talk with everyone in a given session, so there is ALWAYS important data “left on the table” at the end of each session.

Some of this lost data is around policy, some around subject expertise, some around personal connections pertinent to the work discussed, but a great deal of the data is geographic. For some reason, many foundation people are “data shy” when it comes to sharing the geography of their own work. I really want to be proved wrong on this point, but I've seen the same scenario many times: foundation program officers meeting their colleagues in attempts to coordinate their efforts in a program area and locale they share (say community development in a given city) and admitting that they're hampered in their work by a lack of shared knowledge around the geography of their mutual grant-making. Everyone acknowledges this as a problem, everyone thinks that something needs to be done to correct it, but no one seems able to come up with a solution that solves the problem with reasonable time and resources (and therefore, it never gets solved).

There are two hurdles that need to be overcome here: first, there is some fear around actually mapping one's grants and risking having the data seen by the press or the public in a more understandable format than the 990-PF's; second, most foundation people still think that mapping their grants is time consuming, technically complicated, and the exclusive realm of their “technical people”.

The fear will go away as soon as people acknowledge it and see that the benefits of mapping and sharing the data around their place-based grants far outweigh the potential for negative publicity – this is, admittedly, a chicken/egg problem. As for the technical challenge of producing maps of place-based grants, there have been major changes in the last few years. With the introduction of mapping applications like Google Earth, it is now possible to map place-based grants, view them on a free cross-platform application, and share any part of the data with colleagues via email. This can be done by anyone almost as quickly and easily as a grant can be described.

So how can this technology help smart, committed people who've come together to share information and learn more about their field avoid walking away from “lost” data in the future in venues like TFN's annual conference or any gatherings of field-specific grant-makers?

First one needs to acknowledge at the start that walking away from potentially useful geographic data is a problem and that there is a non-threatening solution. One approach might be distributing simple forms at the beginning of each session that allow participants to note ANY place-based grants, programs, or initiatives that they are aware of that are pertinent to the topic of the session or any of the subsequent discussion. A form could be as simple as a description of the grant's locale (neighborhood, CDC, city, etc), the program area (environment, community development, etc), the amount and duration of the grant, a short description of its purpose, and the name and contact number (not to be mapped) for the submitter in case of questions. The data could be mapped the same day, available for review, and with a conference such as TFN's, presented to all interested attendees at a session before the last event of the conference.

There's no question that some of the elements mapped under these circumstances would be “rough” (approximate locations when neighborhood overlay maps weren't readily available are an example) but these could easily be cleaned up after the conference and shared with the participants if they proved useful.

This is a simple, easy-to-implement approach to maximizing the potential place-based knowledge that can be gained whenever experts convene. Trying this approach costs almost nothing, the gains could be remarkable. Don't walk away from your geographic data!

The Emperor's New Maps – Part 3

Analysis: What Does This Mean?

Some notable causes of the negative response to the idea of mapping and sharing place-based grants data seem to be too much money, a fascination with complexity, and little understanding of information technology. When presented with the idea of mapping their place-based grants, most people who make the strategic decisions for foundations understand the potential value but have no understanding of the technology. The decision makers invariably see complex and expensive systems as being superior to easy to use and free, or inexpensive, alternatives. Perversely, since the time and cost of implementing mapping solutions is directly related to their complexity, this leads to the inevitable conclusion that they don't have the time, staff, or discretionary dollars to do what they acknowledge would be incredibly useful to their grant-making programs. The tendency to hand the ultimate decision off to technical staff (who have every career reason to want to see complex systems, requiring ongoing technical support introduced into their organizations) further handicaps the idea of free, shared mapping of place-based foundation grants and the benefit that would bring the sector.

However, I now believe that the most significant reasons foundations, largely self-proclaimed paragons of openness and transparency, are so hesitant to map and share their place-based grant information has little to do with time, technology, re-granting, or ongoing projects. Rather it has everything to do with the control of information and the power imbalances under which they have grown accustomed to operating. While I believe most foundation people are largely motivated by a desire to make the world a better place, I also believe that the unique conditions under which they work can lead to an unusual and “out of balance” relationship with their applicants and grantees. Foundations and foundation staff are almost always dealing with supplicants and this does something to one's view of the world and one's sense of control. When one is always on the dominant side of any relationship, it is hard to hold oneself accountable.

The increased scrutiny of foundation grant-making that we've seen in the last several years has also made foundation employees less comfortable with making public data more accessible. This is true internally, whether it is a concern with one's performance in the eyes of a supervisor, how a program fared in a grantee satisfaction survey, or with the opinions of one's board. It is also true externally, whether it comes from the reaction of unsuccessful community grant-seekers, the local press, or city, state, or federal government.

All of this, coupled with the view engendered by the inherent power imbalance in philanthropy, conspires to make it very hard for foundations to voluntarily make access to their grant-making data more accessible and to use it as the powerful tool that it can be.

Mapping place-based grants should be about the right way to visualize and use data that is already public. Once you've answered all the questions about time, technology, resources, and cost, the objections will always boil down to who gets access to the newly visualized data and the power it imparts.

Thinking About Solutions

So how does one map place-based grants in any location when there are so many grants that could be mapped and when foundations themselves resist the idea of the place-based grants data becoming public in a useful format?

I'm proposing several steps to show the ease and utility of the process. Since it is way beyond the ability of any individual or small group to map all these grants, it makes sense to concentrate on a few cities and a few place-based program areas first – show the value without diluting the effort too much.
Building on work already finished, it makes sense to initially concentrate on community development grants in Baltimore, New York, and Boston by posting and expanding the mapping work already completed.

The mapped grant data will be publicly available at several sites including a Foundation Mapping Project site and possibly Google Earth and ESRI sites. Since I want to encourage foundations to participate in mapping their own work here, I'll continue to offer to teach any foundation staff how to map their place-based grants data at no charge if they'll agree to make the data public and share it with their colleagues in other foundations. For a variety of reasons, many larger non-profits have expressed interest in mapping the grants they receive and there's no reason not to include them in the educational effort. Step-by-step instructions and sample maps and links to tutorials will be available through the project site.

The next step will be to demonstrate how much of the process can be automated so that grant data can be dumped from spreadsheets and grants management systems (initially Gifts) making drawing of the polygons the only manual task in the process.

Another step involves the realization that there is a big difference between mapping the data that is already available through the 990-PFs and annual reports and mapping real-time grant-making. 990-PF data is almost always at least 18 months out of date. The ability to see who is putting money in an area of interest right now, when you are considering making a grant, is even more powerful than the retrospective look we've considered up until now. If we succeed in getting foundation “buy-in” we can move toward making this close to a real time process by having foundation support staff map grants as they are approved.

A fourth step could be the addition of underlying demographic data, most of which is freely available from the Census Bureau. This would another dimension to real-time mapping as a decision making tool and would be another step in making mapping widely used for place-based grant-making. While it is impossible without community, program goal, and grant specifics to speculate on which measures might be useful in which communities, the easy availability of this data raises many possibilities. Are the right neighborhoods being served by your grant-making? Will a better objective understanding of the community's issues improve your grant-making? Are there over-funded or under-funded non-profits serving the areas in which you're working (and what could you do about it)?

Finally, in the future, with the addition of demographic data, tracking changes in this data over time might provide measurable indicators for the success of your program's funding.

Conclusion

I hope it is obvious that the mapping of place-based grant-making by foundations could provide tremendous gains in foundation effectiveness and add real transparency to the process. The recent availability of free, easy to use, and powerful mapping tools should convince the foundation world to consider their values and adopt public mapping and sharing of information as a way to improve the effectiveness of their place-based grant-making.

The Emperor's New Maps – Part 2

A Chronology of Conversations with Foundations About Mapping Grants

The first thing I did was ask questions and listen to foundation staff. Going back to the common theme of program staff being unaware of similar, place-based work being done in their communities, it is almost universal to hear “if only we knew where our colleagues (funding locally) were putting their money, it would make a huge difference in the way we do business”. Clearly, foundation staff understand the gain to be had from shared geographic knowledge of their place-based funding. Over a three year period I heard this from foundation presidents, program directors, and staff in Baltimore, Chicago, Minneapolis, Pittsburgh, New York City, New Hampshire, and Vermont. I also heard it from the staff and leadership in three separate grant-makers' associations and two non-profits whose work is foundation effective

I learned that almost all foundation people think that mapping and sharing data on their place-based grant-making is a wonderful idea. They acknowledge that it would help them make better place-based grants, aid collaboration with their colleagues in the field, and demonstrate how transparent they are to their communities, to the press, and to to those involved in governmental oversight.

However, in my experience over the past several years, a surprising and contradictory fact is that the same foundations and foundation people, who express such interest, will almost never voluntarily carry through on mapping and sharing data on their place-based grants. I may be wrong - I hope so - but in years of working on this issue, I've never come across a foundation that consistently maps its place-based grants and shares them with anyone outside of the organization. I've certainly seen isolated cases of shared mapping on single projects and a few cases of internal mapping, but never any mapping of place-based grants that is consistent and consistently shared with anyone outside the organization.

In attempting to encourage foundations to do this mapping and sharing (that they readily acknowledge would be so useful), I heard the following - and variations - many, many times:

we don't have the time,
we can't free up any one to do this work,
we don't have the technical staff we'd need,
our technical people need to make this decision,
we already do that,
we don't make any place-based grants,
a lot of our dollars are re-granted so this doesn't apply to us,
we're already working on a project (usually knowledge management) that will do this for us,

and on a slightly different level,

the media would misuse this information,
people would be upset if they thought we were putting our money in the wrong neighborhoods,
the local non-profits would ask too many questions and be too upset if they weren't funded,
our board would want to know why we aren't putting our money in more distressed areas.

The answers to the first set of objections are easy:

it doesn't take more than a few minutes per grant,
it isn't technical, anyone can learn how to map place-based grants in thirty minutes, and your technical people shouldn't be the ones making this sort of strategic decision,
occasionally requiring a grantee to produce a map or using a service like Foundation Search is not the same as mapping and sharing your place-based grant information,
almost everyone makes some place-based grants,
ask the re-granting agency to do the mapping (it's still grant dollars you should care about),
none of the knowledge management projects I've seen have ever been finished with any shared mapping component in place.

The second set of common objections, which I will address in a following section, I believe reflects a deeper and more difficult problem that goes to the heart of the resistance to mapping and sharing place-based grants data.

Next Steps: Proof of Concept

After consistently hearing the same first set of objections for over a year, I changed tactics, and rather than trying to get foundations to embrace the idea and begin mapping themselves, I volunteered to do the mapping for them at no cost (including mapping of re-granted dollars) to see if that would satisfy their objections. I thought that they would readily embrace the idea once they saw how useful it was and what little strain it put on their time and their budgets.

I chose as my test case of this new strategy, a group of half a dozen foundations in Baltimore who met three time in 2005-2006 to talk with each other about their community development grant-making in the city. I attended two of their meetings in 2006. At both meetings several of the attendees bemoaned the fact that they didn't have a needed comprehensive overview of where the other attending foundations were making community development grants in the city of Baltimore. After both of the meetings, I contacted all of the attendees and offered to map their grants and those of their colleagues at no cost if they'd agree to share the information with each other.

While no one refused the offer, their participation was far from enthusiastic (despite the fact that they were getting what they claimed to need), and in the end, while I was able to map the community development grants of the majority of the foundations and the major re-granting agency, not a single foundation was willing to share the information or make it publicly available. Remember that the only information that wasn't already available in their 990's or their annual reports, was the actual mapping of the locations where the grant dollars were spent (usually just a polygon on a map encompassing a neighborhood in the city). I simply gathered it all together and mapped it onto city of Baltimore.

During the time I was gathering and mapping the data, I was contacted by a journalist writing the weekly philanthropy column for the Wall Street Journal. She'd heard about the mapping project and thought that foundations that were using mapping to improve their grant-making might make a good story. Despite the fact that it was all public data and I'd done all the work at no cost to the foundations, in order to be politic, I felt that I really needed to get agreement from the foundations before I went ahead with any substantive discussion with the journalist.

To my surprise, none of the foundations were comfortable with my continuing any discussions that might lead to a column in a national paper (even when it would reflect positively on them and make them seem more open and transparent in their work).

[Next: Analysis – What Does This Mean?]

The Emperor's New Maps - Part 1

Overview: Mapping Place-based Grants

A significant number of charitable foundation grants to non-profits are place-based, that is, the grant dollars are spent in a specific , definable geographic area or areas that can be represented on a map. Different foundations and different program areas within foundations will have different mixes of place-based vs non place-based grant-making. Community development, housing, the environment, and public health may have a large number of place-based grants, while other foundation funding areas (medical research, advocacy, social action) seek outcomes that may not be limited to specific geographic areas. In discussing this with the research department at Guidestar, we reached the conclusion that, conservatively, 40% of foundation grants were place-based.

Surprisingly, given the emphasis on openness, transparency, and collaboration, one of the common themes one hears in talking with groups of foundation staff doing similar work in the same geographic areas is that – for the most part - none of them know where the others are spending their money. This is true both within, and between, grant-making foundations. Even funders groups that form around specific place-based community issues will bring up the lack of common knowledge in group meetings, yet there is almost never effective follow through. My purpose in writing this paper is to look at the reasons this is happening and to suggest ways in which mapping geographic information can be used to help us all make more effective place-based grants at almost no cost.

Rationale

A geographic approach just makes sense with place-based grants. There isn't an area of business or government where geographic information isn't used to save time and resources, track assets, increase efficiency, accuracy, and productivity, generate revenue, increase collaboration, aid in budgeting, add to transparency, and support decision making. Yet, as far as I've been able to determine in several years of work, geographic information is completely absent as a shared, open decision-making tool in the foundation world. Mapping certainly exists as a tool in some program areas of some foundations and as an external, third party service, but I've been unable to find any examples of consistent, open, and shared mapping of place-based grant-making in the foundation world.

If a foundation program officer can see on a map where her money will be spent, which foundations have spent money in the specific area in the past, who is spending money in adjoining neighborhoods or communities, there is a tremendous opportunity to communicate and collaborate with other grant-makers, to leverage one’s investment, to avoid redundancy and waste, to make intelligent use of up-to-date demographic information, and ultimately – when one takes a longitudinal point of view - to gain a better understanding of the effectiveness of one’s grant-making.

All but one critical piece of information to make this approach possible is already available in each foundation's form 990-PF and, in many cases, in their annual reports and grants management systems – that is, exactly (down to the neighborhood level) where the place-based grant dollars were spent. All grants management systems will contain the addresses of the offices of non-profits that received grants and it is usually a simple matter to get these addresses given the information in 990's and annual reports, but that doesn't tell where the money was actually spent. Mapping the office location of non-profits that received grants can be automated and gives foundations a certain “window dressing” level of information (which is frequently confused with mapping where the money is spent) , but it doesn't provide the extra data that can enable a host of additional and substantial benefits.

Mapping: Staus Quo, The Problem, and a Way Forward

Today there are several organizations that offer geographic information in one form or another. Knowledgeplex's DataPlace project offers free access to housing and demographic maps and data. TRF's PolicyMap offers a web-based GIS (geographic information system) with access to 4000+ indicators. PolicyMap's Premium subscription level adds the ability to upload and plot one's own data-sets for a substantial fee. Foundation Search, a for-profit company, offers analysis and geographic visualization of foundation grants on a state, county, and city level, but there is no detail available at finer than the city-wide level (in other words, if you care about neighborhood or block level data, you're out of luck). Even when the existing services list all the grants to non-profits down to the city or zip-code level, they only show the location of the office of the non-profit that received the grant – not where the money was actually spent. These are frequently very different locations.

What none of these solutions do is enable the average foundation or non-profit employee to easily and freely add and share place-based grant data at any level of detail with any of her colleagues anywhere in the world. Because they use custom-developed software platforms, the least expensive of the alternatives mentioned above is still several thousand dollars per user per year and they are sufficiently complex that most people will need to have their IT departments handle the data transfer.

What is needed is a powerful, free mapping application that will run on any system connected to the Internet, has access to huge amounts of public data, has the ability to save and share user-created data-sets, and can be learned in 20 minutes by anyone who can type and use a mouse. When Google released Google Earth and the .kml file format in 2006, we got just such an application. Both ESRI (ArcGIS Explorer), the world leader in GIS, and Microsoft (Virtual Earth), have added support for the .kml file format in their recent releases.

I've been working for several years through the non-profit Foundation Mapping Project to get foundation staff to use inexpensive mapping tools and techniques, such as Google Earth, ArcGIS Explorer, or Virtual Earth to help in making more effective grants.

[Next: A Chronology of Conversations with Foundations About Mapping Grants]