I’m thrilled by my latest interview for DMLcentral, re-posted below. When the Dallas Museum of Art (DMA) a few years ago announced it was doing away with museum membership (gasp!), it made big news. Its membership was replaced with an open-badging system called DMA Friends, open to visitors, new and old. I recently contacted Robert Stein, DMA’s deputy director to learn more about DMA Friends, how it empowers visitors, and the ways museum officials analyze the resulting big data to better serve their city.
Robert, Welcome to Mooshme. Please tell us about the Dallas Museum of Art.
The DMA has been in Dallas for 112 years and it’s the largest encyclopedic art collection in the region, the largest cultural venue and organization in North Texas, and it’s really adjacent to the downtown core of Dallas. We’re one of the anchored tenants of the Dallas Arts District, which is the largest arts district in North America.
What challenges were the museum looking to address and how was the design of DMA Friends developed to take that on?
I came to Dallas about three years ago, together with Max Anderson, our director here at the museum [note: Max has now left the DMA and is executive director at the New Cities Foundation]. Max and I had worked together in Indianapolis for quite some time. When we came to Dallas, one of the things we noticed is that while the museum is really a central part of the community, it had a relatively stable attendance over a 10-year period of time. Yet, during that period, Dallas as a city was growing exponentially. We really started to think about what it is like to create a museum that really serves its community, as a city museum. And, in Dallas, like most major metropolitan areas in the U.S., there’s a big disparity between the rich and the poor. A lot of the families here in Dallas, who are less wealthy, just have a hard time bringing their family to any museum, let alone an art museum. And, art museums still have a bit of a stigma of being an elitist sort of institution and a place lots of people think are “not for them.”
So, when we were thinking about that with respect to Dallas, we decided that something that would make a lot of sense for us, and for the museum’s future, is to go to a model of free general admission. The museum had previously charged $10 for access to the permanent collections. So, we decided that what we really wanted to have happen was for the museum to be a place that’s owned by the community. We wanted to be able to facilitate that pop-in visit. So, we dropped the admission charge so that anybody can come to the museum any time we’re open, pop-in for just 5 minutes or 5 hours, and it’s not a matter of $50 to bring your family.
The other thing that we noticed is that it’s sort of strange if we’re saying we want to be open to new people, that the moment you walk in the door, the first question we ask you is: Are you a member? We felt like that probably just reinforces the country club feel that people have when they’re coming to museums. What was and is important to us is that anybody who comes to the museum and wants to belong or participate and plug-in, is welcome to do so. DMA Friends was really a response to that, which said we care more about your participation than your money, and, If you would like to join us and do so through spending time with us, then we’d like you to have that sort of membership for free.
What is DMA Friends?
DMA Friends is a participatory membership program. It’s free to join but focuses on what you do with the museum rather than paying us upfront and getting benefits later. It’s similar to a loyalty program like a frequent flyer program or a shopper’s card or dash card, DMA Friends provides a digital platform to manage your participation with the museum and gives visitors credit for plugging in and participating with all sundry activities across the museum.
The website describes DMA Friends as something that allows visitors to both earn badges and points. Let’s talk about that some more. What might a visitor do to earn points? How does that relate to badges and how might someone looking at someone else’s points or badges understand what they signify?
If you think about all of the things that any kind of museum offers — turns out there are hundreds and hundreds of activities or things to do during a museum visit — it also turns out that most people who visit your museum don’t actually know about those things. They’re not really aware, especially audiences who maybe are not experienced museumgoers. They’ve no idea of the scope of service and program that you’re already providing. So, DMA Friends acts as a platform for that, essentially a menu like you would have in a restaurant that lists out all of the dishes that you are offering. Friends will list packages in a way that’s nice and appealing, the kinds of things that you can do in the museum today when you’re here.
Some of the idea with the Friends platform is that by issuing a simple little challenge, you can actually get people to participate more. When you bundle activities together we call it a badge. You can call it a goal or a challenge or anything else. In our language, we call it a badge. An example might be we have an encyclopedic art collection with galleries from around the world. We might create a badge or a challenge that says, Hey, you can visit the world in just an afternoon at the museum, so try checking out our Asian galleries or European galleries and our Pacific Island galleries. That would be the challenge and what we found is that even just telling people the smallest snippet of the story and issuing them a challenge that is interesting and achievable, they actually go ahead and do it. And, you know our Pacific Island’s gallery or Indonesian galleries are not often our most popular spaces but, just that little bit of a challenge causes people to seek them out. It drives people into the permanent collection.
To what extent is the design of the system drawn from game design?
There is a game element there. It’s not heavy or overt but definitely the experience economy of giving credit for actions you’ve accomplished, recognition, issuing challenges and some sort of economic forces around benefit and reward — those are all typical gamification approaches that are used or leveraged in the Friends program. We are attempting to make it feel very relational, less like a game or a path that you’re on, and more like an ongoing relationship. We found that while museums talk a lot about lifelong learning in the relationships they build with their audiences over time, we have almost no datasets that track longitudinally arts participation over time. That’s kind of what we were designing Friends to do.
Let’s talk about the analytics that you’re using. I thought maybe together we can go to the webpage that shows data through December.
Alright.
What are some types of information you can now learn about your visitors that you didn’t have access to before?
We have, as of today, a little more than a 100,000 Friends who signed up. These are folks who are here in the building. We don’t really allow or facilitate online sign up of Friends. So, the dataset is fairly clean for us of people who have visited the museum. That lets us have the ability to do all sorts of things and we’re playing around with the ability to look at one particular user’s participation as well as whole groups of users. We’re trying to think about segmentation by experience or by geography and how those things drive repeat attendance or more diversity in the kinds of activities that people do.
Why don’t we look at one of the charts that you share with the public. Please help us understand something you’ve been able to learn about your visitors as a result.
This one has some of the most insightful data we’ve had. Every little squiggly shape is one of the zip codes in Dallas and we ask people, through their zip code, when they come because it’s a convenient way of hooking into other demographic data that we have in our statistical database. Because we’re dealing with very, very large numbers, we will tend to converge toward the census data and demographics in any particular zip code. In this map, the DMA is located at the center of the map. The darkest green zip codes are just to the north. What we’ve done here is we’ve looked at the population of the Friends program and divided that by the population of each zip code. Essentially normalizing for the population of the zip code and what that shows us is that white zip codes or the white shapes on this map are the ones that are in balance where we have the same number of Friends from that zip code proportionally as the zip code has as a part of the metro area. So, if 75214 has got 50% of the population, they should have 50% of the Friends and they’d be white. Green zip codes are relatively overbalanced, where we have a higher representation of Friends than the population says we might and red zip codes are underbalanced.
In this map, you’ll notice to the west are a bunch of red zip codes. For those who are not familiar with Dallas geography, the west is where Fort Worth is. Dallas and Fort Worth form one metroplex and Fort Worth has great art museums —The Kimbell Art Museum, the Amon Carter, the Modern Art Museum as well as other kinds of museums. If you wanted to visit an art museum, you could do so without driving all the way to Dallas. And, a lot of those museums have free admission as well. We’re not so worried about the west. You might notice there’s a green island in the northwest of the map, sort of stands all on its own in a sea of white; this little zip code is Denton, Texas. Denton is home to the University of North Texas which has a fantastic music performance program as well as an art school and so there are a lot of students that are in Denton who drive down the freeway to Dallas and become Friends with the program. We find that Friends as a program is really appealing to younger folks like millennials and students, in particular, and that’s why Denton is so high.
What is something you can share that you’ve done as a result of looking at this data that you wouldn’t have done beforehand?
There are two zip codes that are just to the south of where the museum is; they are red and they’re just to the south of downtown. These are areas that we’ve been working on for quite some time as a museum. They are lower socioeconomic communities, where a higher percentage of folks don’t speak English at home, and both parents, generally, work in the home. While those zip codes are only about a 10-minute drive from the museum, if they needed to take public transportation, they would be about an hour on a bus or a train. So, those are areas we focused on for a long time in terms of our outreach program but we’ve never really had a way or sample that was big enough that could tell us why when we do an outreach to one of those two zip codes it didn’t actually make a difference. This map is one that we’re using so we can look at the participation of those zip codes over time, correlated to our actual program outreach efforts and/or things that we’re trying — you know, certain kinds of events or exhibitions — to figure out which are the ones that are really drawing most effectively so then we can plough into those kinds of activities more often. That’s how we’re using this data.
This chart above looks like it’s a chart of activities that garner the most participation. Can you help us understand what type of things museums learn by being able to access data at this level of detail?
This chart mostly demonstrates that we have the ability to see the forest and the trees, so I can know what the whole audience is doing. I know granularly what kind of things they’re participating in, so I can get a view of the balance of activities by category, to say that we’re being most successful in driving the following kinds of experiences. This one is looking at the very basic level of those items that would go on the menu, you know, contributing to selfies or an exhibition in one of our galleries, attending exhibitions, going to the permanent collection spaces. What we do with this and this chart doesn’t necessarily show but we do group these by type or category and we can trap over time what are we doing in our studio programs that is the most effective across the Friends audience.
Is there anything that you can point to as an example of something you’ve changed or done as a result of data like that?
This chart is a good example. Even in describing the program to you, it takes a couple of minutes to get into the details so, we’ve worked really hard for in-house staff to create an experience that is welcoming to talk about the program in a way that’s compelling and concise. Our visitor services staff play a really key role in making sure the public feels like the museum is a place for them; this chart is showing us how well they’re doing in describing the program and encouraging people to sign up. The blue vertical bars are the number of Friends who joined in a particular week and then the black horizontal bars are the 8- or 12-week average; the black bars are showing you the trends.
In the earliest part of the program, in January through the end of the spring, we were floundering around a little bit trying to learn for ourselves the best way to describe this to the public. Then, you can kind of see in the summer of 2013, it really took off. That trend line goes up significantly. And, then, it sort of hovered around and, in the winter of 2014, we got a little bit better. Some of that is experimenting with how we staff our entrances so we have taken an approach of actually staffing a more senior and a more junior gallery attendant at the entrances. They can mentor each other and actually proved to be a good balance.
We also try to ensure that we have at least one Spanish speaker at all the audiences. We find that made a big difference for us — not surprising but we hadn’t anticipated that. Now, we can begin to make decisions on how we actually approach and greet Friends, so around the winter of 2014, we decided that just continuing to sign up more and more people was great but we already had an audience that was large enough to inform about what we were doing. We wanted to try to challenge those staff to really spend more time talking with existing Friends about how they can continue to use the program and come back and reengage. You see that it starts to fade off and you might see even December starts to pop around a little bit more. It becomes a point where you’ve never really thought analytically about what the greeting experience is when somebody enters your museum for the first time but it is that first 5 minutes that sets the tone for the visit that makes people either feel welcome or not and either they’re feeling confused and don’t quite know where to go or they’re feeling plugged in and ready to go.
We think that entrance experience is tremendously important and we’re trying to use data to help us make it better.
What are the other resources that are required to design, launch, maintain and update such a system?
In Dallas, we’ve done a lot of the work on designing software and experiences that go into this. We’re trying to do a good job of documenting that. We’ve written a number of papers that describe that process. They’re all available online if you just search for DMA Friends. The software that we created is available to anybody under an open source license so it’s free for you to use; still does probably require a web developer to help you install it and configure it for your team but all of the work that we did to develop that is available to anyone for free. It’s as simple as purchasing a couple of iPads to set up in your lobby and the most significant investment is in the time and brainpower to create the content that fills up the bucket and then the brains to watch it and learn from the data that comes up the other side. The DMA has invested about a quarter of a million dollars in terms of software development, design, the concept and then all our work to test it out and refine it.
We’d love lots of people to join us. We’re actually funded additionally by the Institute for Museum and Library Services, with a National Leadership grant, to pilot this program in three other museums. This summer, LACMA (the Los Angeles County Museum of Art) will launch a pilot based on the software and they’ll use it with their NexGen program, which includes about 200,000 student participants. The Denver Art Museum also will launch a pilot this summer for their public audiences to learn about all different ways to plug in with the Denver Art Museum, in the museum and around Dallas. This fall, it looks like the Minneapolis Institute of Arts will launch a program integrated with its larger membership initiative in Minneapolis for a free membership tier there at the MIA. And, we recently announced that one of the national museums in Seoul, Korea — the Museum of Modern & Contemporary Art with about 1.4 million visitors a year — will also rollout a platform based on the Friends software.
What would you want museums to take away as a key lesson learned here?
It is often considered too hard to measure the impact that your museum is making on audiences, whether that’s learning the galleries or appreciation of art or history. Most folks consider it to be really hard. The traditional approaches of intercept surveys or focus groups to determine this are very slow, it takes a long time to do the analysis correctly and the sample size is very small compared to the size of your audience. This sort of approach is a great companion to those efforts. It doesn’t replace or get rid of the need to do focus surveys and face-to-face interviews but, it gives you a very large sample with real time data and the ability to have it be actionable. So, you can use this technique whether you use this software or not, you can use this technique and give your staff insight into the intangible things that are happening in and around the galleries.
