Data for the Greater Good: Tough to Say No

Have you ever gone to a conference and, in presentation after presentation, there seems to be a subtheme that isn't on the agenda? It's just something that comes up as a key point with multiple speakers and sticks in your brain.

Jake Porway
Jake Porway

Maybe it's a subtle strategy by the conference planners, or maybe we get focused on one of the first speaker's points and imagine it being echoed by other presenters. Whatever the source, one key takeaway from this week's SAS Analytics Experience 2016 was that it's time for organizations to make sure that their analytics initiatives aren't just about probing data, but that they have the means to put that data into action.

It was a message hammered home by keynoter Jake Porway, founder and CEO of Datakind, the nonprofit that coordinates volunteer work by analytics professionals to use data for good, helping social organizations address some of the world's challenges: disease, food production, troubled children, racism, and more. It works with organizations that typically don't have the resources to hire their own data scientists.

Porway is what you might call a high-energy guy, as you can see in the video of his speech, which was an update on the presentation he gave at SAS Global Forum in 2015.

He outlined the progress that Datakind has made, having signed up more than 12,000 volunteers and helped more than 135 organizations. In his keynote and in a later interview, Porway discussed successes and failures, and lessons he has learned in the past two years.

The lessons include some relating to how Datakind and other nonprofits put data into action; but I will add, that lesson applies to the business world, too. He listed six key elements of a successful project, with an emphasis on how social organizations will use the resulting data.

"At the end of the day, we don't care if we just build something. We have to make sure it is something they are going to use that they can apply, something that they are going to make a difference with," he said.

The six elements are:

  • A "super clear" problem statement
  • Datasets (belonging to the organization or in the public domain)
  • Data scientists (the Datakind volunteers)
  • Funding (most social organizations never have enough)
  • Subject matter expertise
  • Social actors on the ground

The first four elements on that list are obvious within any data initiative -- for profit or not -- although we all know that the one about the "super clear" problem statement sometimes gets ignored in many realms.

The last two elements -- subject matter expertise and social actors -- seem to be key to putting data to work in any organization. Porway uses the phrase "design with, not for."

The subject matter expertise supplements the analytical and data skills that a data scientist brings to the project. That subject matter expertise could come from someone such as an expert from the social organization, an experienced field worker, or maybe an independent expert. It's someone who helps define the real problem that needs to be solved, perhaps can identify untapped data sources, and can help to analyze the data as it comes together.

"You really have to make sure you have the constituents and social actors there helping you design it in the first place and design the user experience piece. If you don't, you're going to build something that is useless or worse -- dangerous," said Porway.

What he calls the "social actors" are representatives of the people who will use the data in the field. They will provide the input on what actions the users of the application will be able to take, what types of information they need, and what they cannot do for various logistical reasons.

In a way, the world of Datakind and the social organizations it seeks to help isn't that different from the business world.

"It's really parallel to business applications. For-profit companies work to serve their customers, innovate on their products, and ultimately make a certain amount money. Nonprofits aren't that different. Their end goal might be reducing homelessness or pollution. They need to serve their customers and come up with products that do that," he said.

However, there is one key difference. In the business world, while it's possible that a really bad failure could break the company, in most instances the worst-case scenario is that a couple of individuals could lose their jobs. It stinks for them, but the company rolls on. For a nonprofit that fails in its analytics initiative, the wasted funds could mean loss of a sponsorship and the demise of the whole organization. Then those in the most need, those that the organization serves, suffer the most.

Yet, Porway noted, a successful data project could serve as a proof point that encourages sponsors to invest in still more analytics projects. A string of successes could mean that the organization can hire its own data team.

Being the head of a nonprofit that relies heavily on volunteers, Porway used the Analytics Experience platform for recruiting, calling on the SAS professionals in the audience to step up and get involved in Datakind projects. For those who need a job, he added that Datakind is still growing, and he recommended a visit to

Watch the video of his presentation on demand, and then try to think up excuses for why you wouldn't heed his call for volunteers to spend a weekend working on a Datakind project. It might be hard to say no.

James M. Connolly, Editor of All Analytics

Jim Connolly is a versatile and experienced technology journalist who has reported on IT trends for more than two decades. As editor of All Analytics he writes about the move to big data analytics and data-driven decision making. Over the years he has covered enterprise computing, the PC revolution, client/server, the evolution of the Internet, the rise of web-based business, and IT management. He has covered breaking industry news and has led teams focused on product reviews and technology trends. Throughout his tech journalism career, he has concentrated on serving the information needs of IT decision-makers in large organizations and has worked with those managers to help them learn from their peers and share their experiences in implementing leading-edge technologies through publications including Computerworld. Jim also has helped to launch a technology-focused startup, as one of the founding editors at TechTarget, and has served as editor of an established news organization focused on technology startups and the Boston-area venture capital sector at MassHighTech. A former crime reporter for the Boston Herald, he majored in journalism at Northeastern University.

Big Data Success Essentials: Tech, People, and Process

While there is an increasing focus on the role of people alongside technology in analytics initiatives, let's not forget that process -- business rules -- play an important role in big data success.

Why You Should Remember Equifax

The Equifax breach raises multiple concerns about how companies respond to hacks, but also how they handle third-party consumer data.

Re: Growth
  • 9/21/2016 10:04:41 AM

@kq4ym. Good observation about the difference between the number of volunteers and the number of organizations. I think one explanation would be that the model they use for a lot of projects is like a hackathon. They bring together five or six (or more) volunteers for 48 hours over a weekend to explore data sources and come up with an application.

They may have longer engagements with some non-profits but that probably would involve part time work by just one or two people.

Re: Growth
  • 9/21/2016 9:29:19 AM

Yes it does look like a good organization doing good. But, it seemed surprising to me that there were so many volunteers but only about a tenth of that number of orgainzations they were helping. It seem like they could spread out the talent a bit further.

  • 9/20/2016 8:24:36 PM

I read about Data Kind a few years ago, I believe. It's good to see it growing like this. I've seen some of the founder's work on TV. He's high energy for sure!

Re: Go fish
  • 9/20/2016 5:30:42 PM

I think more must be done in teaching managers how to form those clear expectations. There can be too much emphasis on the technology and process involve to overook how to frame questions quickly as well as clearly.

Re: Go fish
  • 9/20/2016 10:22:19 AM

I'm going to play devil's advocate here and say that the fishing is a good first step.  Even if someone has a pre-conceived idea of the story they want to tell they are still collecting data.  If you can piggy back off of their work and remove any bias then I don't see the problem.  Someone has to do the initial leg work but as long as the data is valid it can be re-used.  I know that in some cases data collection will be skewed based on the bias but if you understand the limitations that those cause there is still a chance you can use that data.  Calling it a fishing expedition is actually a good way to describe what I'm getting at, there are researchers that ride along on fishing vessels even though they may disagree with some of their practices.  Without riding along, they wouldn't get any of the data that they need because it's incredibly expensive to spend weeks or months at sea.  

Re: Go fish
  • 9/19/2016 3:00:49 PM

Fishing expedition has its place and can be beneficial, but not as an analytics strategy. I can see it in the initial discussion probing possibilities, but once the 'super clean statement ' is formulated, fishing ceases. Now, once the defined goals have been achieved, fishing can resume in identifying add-ons. Discipline is imperative for success.

Re: Go fish
  • 9/19/2016 12:44:02 PM

Thanks, Lyndon... I think I have a title for my next book pitch: "Trawling For Insights, Or How Not To Do Analytics."


Re: Go fish
  • 9/19/2016 11:12:15 AM

@SaneIT. Would it make sense to view any fishing expeditions as phase two of an analytics strategy? Use the well-defined business problems to get the initiative off the ground. The more massive data collection would help the organization grow the initiative and justify additional resources. Thoughts?

Re: Go fish
  • 9/19/2016 8:13:41 AM

"In other words, don't treat an analytics project as a fishing expedition."

Unfortunately, I think part of this tendency is that people need to eat.  If you're not working with at least a little bit of a lean in one direction, chances are you don't have much financial backing.  Without financial backing it gets very difficult to assemble the resources needed to do extensive data collection and analysis.  


Re: Go fish
  • 9/18/2016 1:40:33 PM


Terry writes

I always appreciate these back-to-basics reminders but am once again struck by the need to remind stakeholders for the need to have "a super-clear problem statement."

In other words, don't treat an analytics project as a fishing expedition. 


Cogent point. It makes a lot of sense to define what kind of data set and analytics you need rather than drag a trawling net and hope to glean some enlightenment out of it.


Page 1 / 2   >   >>