Part one of this blog series looked at how #GivingTuesday grew from concept to global event in just four years. While we often hear about scale, #GivingTuesday is a living example of just how quickly an idea can spread. From its first year to its third the aggregate giving tagged and counted as part of the effort grew to almost $46 million.
Of course, I also could have led this story with a recent survey finding that only 18% of American consumers were familiar with the idea of #GivingTuesday. Or that the estimates of giving are known to be undercounts, but no one knows by how much. Or that while fewer than 1 in 5 shoppers say they’ve heard about #GivingTuesday, those of us who identify more as givers than shoppers wonder if our email will collapse under the weight of donation requests.
In other words, #GivingTuesday is a born-digital effort with a measurement challenge. That should strike you as counterintuitive– doesn’t every digital click, tweet, and “like,” get counted? Well, yes. And no. And therein lies one of the paradoxes of born digital events – they may need to tradeoff scale for metrics. It’s a true buzzword throwdown.
Let’s step back for a second. #GivingTuesday was designed to grow, spread, and encourage adoption and adaptation. The goal of the effort is not simply to encourage people to give to charity, but to foster and provide an outlet for a culture of generosity — a time and a place when we all collectively celebrate our ability and desire to help others. #GivingTuesday aspires to create the mix and the moment when we both give and benefit from the joy of collective participation.
To do this, #GivingTuesday relies on digital systems and behaviors. From hashtags and “likes” to remixable branding collateral and mobile giving platforms, the infrastructure of #GivingTuesday consists entirely of people using digital tools. The logo is made to be copied, there’s no “fee” or required MOU or commitment necessary to launch a #GivingTuesday effort. Anyone, anywhere with a giving campaign can “ride along” on the tide of attention that the day seeks to generate.
The participation jump from zero to 18% of the U.S shopping population in 3 years actually seems pretty good.
But that same distributed strategy – to make it your own – means there is no centralized repository of data.
Every #GivingTuesday campaign is its own project. It relies on whatever software it wants, counts its own contributions, and tracks its own participation levels (or not). The online payment platforms that help nonprofits process donations do collect the data – but their first data loyalty (as it should be) is between the person who gave the money and the organization to whom it was given. The data on #GivingTuesday transactions are not (by design) funneled through one particular channel. In-depth information on funds raised rests primarily with individual organizations. Online payment platforms capture a chunk of donations – and it may be a large chunk – but it’s not all that gets raised. Plus, there are lots of payment platforms. In its first year, #GivingTuesday was able to report out on the funds raised through only one platform (Blackbaud). In year two, they were able to collect aggregate data on transactions from five major platforms. In the third year, a more coordinated and anticipated accounting was made possible by a Case Foundation partnership with Indiana University. This year, at least 50 payment platforms have agreed to provide aggregated data in a uniform way.
This effort to bring together data from so many platforms means there will be an aggregate number reported out to the media on Wednesday. That number will be more accurate than in the past. Remember, though, there are dozens of giving platforms. Plus dozens of crowdfunding platforms. And some people give directly. Some people still write checks, or (gasp!) donate cash. I’m sure this year will see plenty of Bitcoin donations too.
These donations matter to the organization or effort that raised them (and to the person who donated them), but they’re not going to get counted in an aggregate tally from payment providers.
The collection of data on transactions, popularity, and reach of #GivingTuesday might seem to be simple. It’s not. #GivingTuesday is a classic example of the peril and potential of digital data. Yes, the resource (data) is abundant and cheap. That means it is easy to find the haystack; not so easy to find the needle. Digital data for #GivingTuesday raises key questions of ownership, privacy, consent, and analytic capacity – questions which just so happen to be relevant everywhere else in digital civil society.
The organizers of #GivingTuesday have responded to the demand for an aggregate number. To do so, they have to try to cast the widest net possible to gather data from multiple partners. The expansion of these reporting partners each year is testament to those efforts.
So, here’s the question for #GivingTuesday? What can/should be collected centrally and what should be the purview of the local projects? In 2014 a small group of #GivingTuesday founders and friends met at the Stanford Design School for a “measurepalooza,” to try to identify meaningful metrics and means of gathering them. The group considered many of the same questions I hear people ask about #GivingTuesday – does it favor big organizations over small? Does it just shift giving or add to the pie? Is it new donors or old friends? The effort led to loose partnerships with payment platforms in 2014 and the expanded effort in 2015.
But I have a slightly different question – can #GivingTuesday help us figure out how to distribute data collection efforts in the same way it’s showing us how to distribute participation, activities, and the creation of best practices? Can we figure out a way to keep the data where it is most appropriate (locally) and still answer questions nationally (and internationally)? The people who care most about whether their #GT effort leads to new donors or encourages old friends to make a gift are the nonprofits leading those efforts. They may well have looked through their data, refined their pitches, and found new ways to say “thank you,” that are shaping both their own campaigns and gathering answers to the questions the public is asking. These nonprofit partners can analyze their own data; they know their donors, they know their giving history and, at the end of the day, the most important lessons about #GivingTuesday’s behavior change are going to matter most to them.
Working it this way also respects the existing conditions under which the data are generated and collected. We have two choices about how to get more fine grained answers to the national questions about new donors or old, new dollars or substitution effects, one-off gifts or returning donors. We could try to work out new data agreements at a centralized level, with many different payment platforms, agreeing to share lots of disaggregated data (that means data points on individual givers). In order to answer these kinds of questions at a national or international level, we’d need access to raw data from dozens of payment platforms over time. Getting this information while respecting the privacy rights of donors and the contractual agreements between nonprofits and their vendors is not a task to be undertaken lightly. Quite the opposite, it is one which raises all a host of key ethical issues regarding data use in civil society.
There is an alternative. We could leave the data where it is and distribute the analytic capacity. Rather than building a centralized data infrastructure under a (by design) distributed effort, we could focus on building distributed, high quality analysis skills. Encourage, allow, and facilitate the tens of thousands of nonprofit and campaign partners that participate in #GivingTuesday to make sense of their own donors’ behaviors, and then share what they learn. This approach focuses on aggregating insights, not data.
Either way, we can answer the questions that people are asking and that can help improve the awareness and effectiveness of #GivingTuesday. In making the decision, we should keep in mind the same design principles and values that have powered the effort from the start.