Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Episode #43: Marie Whittaker
141 Plays9 years ago
In this week’s episode of The PolicyViz Podcast, I chat with Marie Whittaker from the Office of the Deputy Mayor for Planning and Economic Development (DMPED) in Washington, DC. With a small team responsible for providing economic insight to DC’s...
The post Episode #43: Marie Whittaker appeared first on PolicyViz.
Recommended
Transcript
00:00:00
Speaker
This
Sponsorship Introduction: Juice Analytics
00:00:00
Speaker
episode of the PolicyViz podcast is brought to you by Juice Analytics. Juice is the company behind Juicebox, a new kind of platform for visualizing data. Juicebox is a platform designed to deliver easy to read interactive data applications and dashboards. Juicebox turns your valuable analyses into a story for everyday decision makers. For more information on Juicebox or to schedule a demo, visit juiceanalytics.com.
Guest Introduction: Marie Whitaker from DMPED
00:00:37
Speaker
Welcome back to the Policy Viz Podcast. I'm your host, John Schwabisch. Changing up the format a little bit this week, I'm actually sitting in a room with this week's guest. I'm here with Marie Whitaker from Demped, which is the
00:00:51
Speaker
It is the Deputy Mayor's Office for Planning and Economic Development here in Washington, DC. You're in DC. And a lovely spring day, actually. Actually, yeah. Actually, yes. Marie and I met at a Tableau user group a couple of weeks ago here in DC, and I thought we would talk about the sort of work that you're doing here. So can we start by having you introduce yourself and then talk a little bit about what DMPED does and what you do here? Absolutely,
Marie Whitaker's Background and Role
00:01:12
Speaker
yeah.
00:01:13
Speaker
So John mentioned, hi, I'm Marie Whitaker. I work for DemPed, which is the acronym for the Deputy Mayor's Office for Planning and Economic Development, a funny acronym, very governmenty. My background is in economics, actually. I went to school for economics. Yes, absolutely. Went to school for economics and art, and then came here to DC to work for the federal government.
00:01:38
Speaker
And then switched over to local government in the middle of last year. I've been at DMPED since last May, so almost a year now. DMPED is the agency in DC that handles a lot of real estate and business development. And we also have a cluster of agencies under DMPED. The deputy mayor likes to say that he's been tasked with three jobs right now by the mayor.
00:02:07
Speaker
One is to increase jobs for DC residents. Two is to produce and preserve more affordable housing in the district. Of course, it's a very expensive city and getting more expensive and housing is a huge part of that. And then the third is to increase the tax base so that we can do more of the first two. And the activities that go into all those three things are the activities that
00:02:36
Speaker
DMPED does. There are a few offices within DMPED. And of course, I mentioned the cluster agencies. So we have a real estate development office and a business development office, which tries to attract businesses and attract jobs to DC. We do industrial revenue bond programs to support nonprofits in certain sectors in DC with taxes and financing.
00:03:00
Speaker
And then we have cluster agencies, including the Department of Consumer and Regulatory Affairs, DDOT, the Department of Transportation, Department of Housing and Community Development, among a few others, too. So we are plugged into a lot of different arms of the government, and DMPED is working on a lot of different things all at once. And
Economic Intelligence Team and Their Work
00:03:20
Speaker
my team within DMPED is the Economic Intelligence team.
00:03:24
Speaker
So we support all of those other offices and other agencies in the district as well as, of course, answering to the executive office of the mayor as well. So we're a very small team. Right. And so how many people are on the team? Actually, right now, three. Okay. Three people. Three people. And you are pulling together all of the data that you need to inform the deputy mayor so that the deputy mayor can inform the mayor about what she should do.
00:03:53
Speaker
Okay, and other agencies. And other agencies, right. So what sort of products are you pulling together to inform both the work you do in your office and also to inform the other agencies and the deputy mayor and ultimately the mayor? I would break it into a few categories. One would be dashboarding products. We put out a weekly dashboard for the deputy mayor and for heads of other agencies to
00:04:21
Speaker
that has kind of the vital signs of the district, things that we're tracking to see what's changing, what are the basics that we need to know about the district economy. We also have a public version of that dashboard online because most of it is public data. And that public facing version helps not just DC government, which as like a one-stop shop for economic data, but also members of the public, members of the journalism community,
00:04:51
Speaker
and other data savvy citizens. So are those two separate products, or are they basically the same and you just... The Venn diagram of those two is very, very close to a single. There are a few things in one that's not in the other. The internal one, some of the data we use is proprietary. So you can't make it public, but basically they're about the same. Gotcha. And what technology are you using to build the dashboards?
00:05:15
Speaker
We're using Tableau to host the visualizations themselves online and in the internal version. And we're using actually Tableau Public, although we're building the Visas in Tableau Desktop. So anybody who has free Tableau version can download the data that we're using and play around with it themselves too. Our workflow behind it, we try to make it as automatic as possible.
00:05:42
Speaker
so that it's not a whole lot of manpower going into updating and grabbing the data. So we use a series of Python scripts to scrape data from wherever it might live. Most of it is online, public sources, and then we've automated that process to
00:06:00
Speaker
into Tableau and then really the only manual step is just updating that in the Tableau public. So it seems like you don't really have to think too much about the two different audiences because they sound like they're kind of similar. I mean you have the dashboard is for you know people who are making decisions also for the public but it seems like they all sort of need the same
00:06:20
Speaker
bottom line. So are you thinking about the different audiences differently? Or are you saying, we're targeting people who are interested in this, and this is what we think they need. It doesn't matter if it's the deputy mayor, if it's the head of an agency, or if it's the public. I think for some of the economic vital signs, that is certainly the case. For example, unemployment. Really clearly an important thing to know in DC. It's affecting the policy decisions that the mayor is making.
00:06:46
Speaker
It's affecting what journalists are writing about TC, how they see the state of the world. It's affecting people's lives and how they feel like their experience in the city is. There are deeper dives maybe we can do into unemployment and how job growth is happening in the city that might be different for different audiences. Internally at least, I like to think of my internal data visualization job as kind of a
00:07:15
Speaker
data hypochondriac a little bit. Since I was being educated during the Great Recession and hearing them in the aftermath about how data really wasn't used to its fullest extent to diagnose or even notice the issues. So my job I think internally is to say,
00:07:35
Speaker
what could be going wrong right now and how can we detect it and kind of make sense of it before anything economically bad happens. So that's kind of what my internal mindset is for how we're reporting data and then explaining it out to the audience. Externally, I think we wanna be evangelists for open data and people being engaged with the city's data. Good, bad or indifferent, we wanna put it out
00:08:05
Speaker
whatever it says about DC and get folks to engage on it. So that I think is another difference between the internal and the external products. External,
Transparency Initiatives in DC Government
00:08:15
Speaker
we really want some community and civic engagement.
00:08:19
Speaker
And we want the data dashboard that's our public version to be a platform for that. Yeah. So let's talk about the open data for a minute, because there's a big open data push here in DC, as in lots of other cities. It seems like you're taking open data, you're packaging it up and you're kind of reopening it in a different way. Have you found that DC agencies are more or less on board with that? And if you've had struggles, how do you get people on board with saying, yeah, we're going to open this data because people should have it and people should be able to use it?
00:08:46
Speaker
Yeah. I mentioned earlier, the three jobs that the deputy mayor has and actually there's a, there's a fourth one that kind of underlies all we do. And that's to try to be as transparent as possible with the recognition that that might not have been the case in the past. So we're trying to be the most transparent to DMPED that there's been. And so one of the things that we've done in pursuit of that is to release the, for the first time, the real estate development pipeline data.
00:09:13
Speaker
that we have. So DMPED has a portfolio, a $13 billion portfolio of real estate projects in the city. Last year, for the first time, we released individual project data on all those projects. We've also put on our public dashboard affordable housing data project, their project, so you can get a view of where the affordable housing projects are under construction in the city and where the pipeline is going to be going. And that, I think for the first time,
00:09:39
Speaker
been helpful for the development community, for the business community, for journalists, before those data points were maybe reported on.
00:09:48
Speaker
in separate stories, it was difficult to see the whole picture. But we think it's helpful to put that data out. And within DC government, as you were asking before, there is a tremendous amount of open data put out on the open data portal that DC has. DC open data is primarily map based. So there's a lot you can do with geographic information. And DC has an absolutely fantastic master address repository for standardizing
00:10:18
Speaker
geographic data. So I think that we're kind of leading the pack there. I saw in a recent open data census that DC had a very good score compared to other states. We're not perfect, obviously there's more work to be done. But making products with the data and showing how those can be helpful is I think the best case to be made for getting more data out there.
00:10:45
Speaker
So we've had a tremendous feedback on what we've put out already. And all of that, almost all of it was already public to begin with. But repackaging data, so it's not just data, it's insight, is such a great carrot for getting more data out there, I think.
00:11:04
Speaker
Yeah. So let's talk about the pipeline a little bit. Cause you mentioned it earlier, you're a team of three pulling in all these data. You have some scripts that runs that are updated quickly. Um, what is the biggest challenge throughout that pipeline from, you know, all the way from getting the data at the very start, just sort of publishing something, including, I guess at the end, getting feedback from people, you know, both internally and externally. What are the pressure points in that, in that pipeline? I would say two, two main things. One.
00:11:32
Speaker
When you're producing data as we are, for example, our real estate data, data originates people frequently, if not always. I think, number one, understanding your data and making it accurate as possible.
00:11:50
Speaker
which I'm, I'm really all about automation. Yeah. That's, that's where we'll cut automation, automation, automation. And that even includes not just scraping data, reshaping it, putting it into a good form, but I think getting data from the real world as much as we can automate it and build it into operational processes as much as possible. That's the key for.
00:12:14
Speaker
good data for real time data and for accurate data. So getting the correct milestone dates for real estate projects, building that into processes and people's
00:12:30
Speaker
actual day-to-day work lives, so that it's not an extra step because when data becomes an extra step, it becomes a chore, it becomes the thing that you do later. So you're trying to automate not just the push and pull of the data, but also the automation of the generation. Absolutely. Where that can be done.
00:12:51
Speaker
And how important is that to have the buy-in of, I mean, you clearly have the buy-in of your team because you're doing the work. How important is it to get all the stakeholders to buy into that? Or is it a matter of just demonstrating some success and then you sort of get people to buy in because they see what's happening? I think it's a bit of both. Yeah. I think it's definitely a bit of both. I'm going to demonstrate why data
Team Dynamics and Skill Development at DMPED
00:13:12
Speaker
is useful and valuable by making cool visualizations, but also
00:13:19
Speaker
demonstrate why it's necessary on an ongoing basis. If you have real time data that you're trying to collect, it's not going to do you any good if you can get it one month and then not a month later. So trying to create points where you're regularly looking at the data, you're regularly updating it and to create products that are on a real time basis like our dashboard that we update
00:13:44
Speaker
and then can answer questions when they come up. Because we're always getting questions about real estate data, so to not have to
00:13:52
Speaker
You do the whole thing. Every single request is valuable for people's time. Right, right. Now I'm also curious because you said you have a team of three people and I think a lot of people in organizations think they need 30 or 40 people to have a team that can pull in data and build these things. So let me first ask, what's the background of the different members of the team and how have you been able to find success with what many people might think is a pretty small team?
00:14:20
Speaker
Yeah, we have a very interesting team, I think, background-wise. And I'd say we're more of a hire people with enthusiasm for what we're doing in local government and learn the skills as we go along than a hire people for the skills and have people siloed in their one little function.
00:14:45
Speaker
area. So my background, like I said, is economics. I worked with economic data, various tools, statistical tools for that kind of stuff, some background in visual design, etc. Another one of our team members has a background in database administration and is really, really plugged into
00:15:05
Speaker
DC government and has been here a while. And another one of our team members, our newest team member has a background, more web development, some programming, and has been with various advocacy organizations. So really a hodgepodge of skills. We had an intern just this last semester from GW who is an IT master's program who learned D3.
00:15:33
Speaker
while she was here. So we're more of a get energetic enthusiastic people and then pick a problem and figure out a way to solve it. So do you have any pearls of wisdom for other teams that are trying to do something similar where they're trying to create dashboards, get other people interested in them, get people to use them, I guess is probably the most important thing. Are there any pearls of wisdom or things that you've had to overcome to get this process to work that
00:15:59
Speaker
you would recommend other organizations or people, you know, keep in mind or be aware of.
Communicating Data Insights Effectively
00:16:04
Speaker
Yeah. Yeah. I think that the main thing that I've learned, uh, both here and in my prior life in the federal government making dashboards for policy makers and, uh, getting data insights for policy makers is that when you are trying to build a narrative about what's going on in an economy for me right now, it's DC.
00:16:26
Speaker
It's going to involve a lot of repetition of what the story is. So find a succinct story with the data and repeat it, repeat it, repeat it. Tell anybody who will listen what the so what is. When you are presenting insights from data, I think a lot of the time, unless there is another expert in the room, you're going to be presumed to be the expert on that data.
00:16:53
Speaker
And people really see that your opinion about it is the value add. So figure out a way to succinctly express your opinion about the data and don't rely on anyone else who's reading it to form an opinion based on a chart or a graph. I would say for policymakers, a lot of the time they come from a world of liking data, liking charts and graphs and colorful images.
00:17:19
Speaker
But one sentence, one pithy sentence is what sticks with them the most, I think. Because a lot of them are from the world of policy memos or law briefs or legislation. So words actually do have value. And I think translating data into words is one thing that I have found to be the most valuable. Let me just ask one last question. When you are communicating with decision makers or policymakers,
00:17:45
Speaker
Are they particularly interested in the interactivity of the dashboard or are they more interested in that one sentence pithy thing that you can say, here's the bottom line. Yeah. Here's this graph. Um, here's that thing. Or do they ever want to dive in a little bit deeper? I think it's more on the former description. Yeah. I think when you're designing a dashboard, just assume nobody's going to play around with it unless you're designing it for, and, and
00:18:12
Speaker
This is an audience, we are targeting obviously for the data nerds out there, for the data savvy citizens or data journalists, et cetera. But if you're expecting a policy maker who has a million different things going on, someone who's running a city basically, to sit down and play around with the data that you've spent hours and hours putting together and know all the caveats about, that's just not going to happen. So you need spoon feed your data results.
00:18:42
Speaker
to whoever is going to listen. Let's just finish up with this. So then is the dashboard itself to creating the interactive interactivity and having it built in Tableau?
00:18:52
Speaker
Having that platform, is that primarily to make your life a little bit easier as opposed to the audience? Because it allows you to bring things in more quickly, it allows you to explore it, but they're really just going to look at the initial view and that's pretty much it. So is all the dashboarding pretty much mostly for you and your team and the other folks in your office to dig down? And then the audience, it's really like, here's the one thing you should really know about these three views.
00:19:17
Speaker
Yeah, I think absolutely. And I shouldn't say that policymakers generally are just interested in the answer because frequently we do need to do a drill down. Unemployment stayed the same this month. What's going on? Wages in DC are falling and have been for
00:19:34
Speaker
past several months, what's going on there? What's the story? So it helps springboard my team into action. When we need to do a drill down, we have everything. We have this data library here for us, and we can come up with a few hypotheses, play around really quickly, and either falsify those hypotheses or know which road to travel down for future analysis. And then it's a matter of them getting that drill down deep analysis into a format that's digestible
00:20:03
Speaker
has action items, has the so what, that's a bit deeper. Great. Great. Great. Well, I think you have your work cut out for you here. Oh, absolutely. Well, thanks for coming on the show. This has been a lot of fun. Thank you for having me. And thanks to everyone for listening. If you have any comments or questions, please drop me a line on the site or on Twitter or via email. So until next time, this has been the Policy Viz Podcast. Thanks so much for listening.
00:20:39
Speaker
Again, thanks to our sponsors, Juice Analytics. For 10 years, they've been helping clients like Aetna, the Virginia Chamber of Commerce, Notre Dame University, and U.S. News & World Report create beautiful, easy to understand visualizations. Be sure to learn more about Juicebox, a new kind of platform for visualizing data at juiceanalytics.com. Also, check out their book, Data Fluency, found on Amazon.