I have been following UNDP’s work on innovation, and I closely followed the Global Innovation Meeting that took place in November 2013 in Montenegro, including its outputs such the excellent Budva Declaration. Still, much of this to me was rather theoretical and I didn’t have a lot of practical experience on how to approach and manage an innovation initiative. So I contacted a number UNDP colleagues who work on innovation and asked them to participate in Peer Assists – a knowledge management methodology that brings together a group of peers to elicit feedback on a problem, project, or activity, and draw insights from the participants' knowledge and experience (to learn more about Peer Assists, watch this excellent 6 min video here).
I was lucky to win Arndt Husar from the UNDP Global Centre for Public Service Excellence, and George Hodge from the UNDP Country Office in Armenia for a Peer Assist session I conducted in Dec 2013 to tap their brains on implementing innovation initiatives, and I hope the following shortened transcript of the conversation can be of as much help to others as it was to me!
Johannes: I invited you to this conversation, because you both have been involved in practical innovation events and initiatives in UNDP in the past. My hope is to get to a better understanding of the conditions under which certain innovation initiatives make sense and how we would plan for something like this. Where would a team like ours here in Amman start? And what would be the conditions under which it would make sense to have e.g. an innovation camp in one of Syria’s neighbor countries to identify new solutions together with municipalities and local actors?
George: I like that you are trying to incorporate different approaches in addressing these challenges. The first thing I usually do is to get out of the office and talk to municipal officials and the other stakeholders in order to get a better sense of the problems. When running social innovation camps and innovation challenges, the better you can define the problem at the very beginning, the happier you will be about what happens later in the cycle.
If you run an open innovation challenge where you ask “tell us about your problems, and suggest solutions”, you get a better sense of where the problems lie. Whereas if you want to address more concrete problems of your stakeholders, then you should run an innovation camp or challenge around a specific question.
I recommend visiting your stakeholders to get a sense of which public services are under strain. Then you can run a series of concurrent challenges asking “Can you come up with ways in which we can overcome this particular issue”. But if you are at an early stage and you are trying to make sense of the environment, an open challenge may be best. You should expect over 100 responses, and from that you will get a sense of where people see the most pressing problems.
Arndt: That was great feedback from George. I think I need more clarity on the scale: Is this something you want to do at an inter-governmental level, or do you want to look for solutions in each specific country? This will define your immediate counterparts and your outreach. Municipalities are good, because they are the ones delivering your cutting-edge services. You could break your issues in the sub-region down into national challenges, and then go out and do the sensing with local partners. For me the big question is the connection between that massive scale of four countries, and the local services that you are looking at.
Johannes: To give you feedback on the scale, we are talking more or less about three countries: Jordan, Lebanon and Turkey, because those are the three countries that host the biggest population of refugees. They are also at a more advanced development stage with larger middle income population, so we may have a higher chance to get some of those technology solutions with private sector off the ground.
Arndt: Ok, I think that’s very good. This sounds similar to what we had in Singapore with our social innovation camps. We had activities in multiple countries and at the end convened a regional summit where we brought the various countries together to combine different country perspectives, which worked quite well.
Regarding the conditions that have to be in place to organize something like this, Mm experience in our regional initiative in Asia-Pacific was that the outputs depended hugely on the local organizing partner and also on the network that the local partner brought in. If you partner with some agency that has a very specific urban network, then you get that kind of result. If you partner with an agency that has a lot of techies in it, then you are going to get a lot of IT solutions. So it’s really important to pick the right partners (ideally a consortium of partners) so you can have a wider range of solutions.
Another lesson that we learned is that during the short duration of a three day camp people really just scratch the surface. I think if you want to get real solutions from municipalities to be innovative, you need to along the lines George suggested: First the sensing, then the definition of the problem, then the call, then probably some research (almost like the production of a case study), and then you go into prototyping and so on. That’s a really thorough preparation process, which we ourselves were not able to do through just a social innovation camp. The above requires a bit more work.
George: What Arndt has just identified is why we – after running a few Social Innovation Camps – decided to set up an Innovation Lab, because we realized we needed a bigger support structure around our events. You need to do the sense making first and then come up with a series of really specific challenges - this will give you results that are much better aligned with your mission.
After the first social innovation camp, we gave the teams grants and just said “good luck, come to us of you have any problems”, but this approach is too passive. By now we have a much bigger support structure where we get our hands dirty with the teams and actively invest in the initiatives that pass through the lab – much more like a incubator.
Johannes: So how many of these three-day social innovation camps in Armenia did you run? And have they all been successful?
George: We have run four of them, and they were definitely worth it. I would say if you run a three-day event, it is useful to look at what is already out there of which UNDP is not aware, and identify people who have a deep understanding of the social problems in question. Maybe there are teams that are already established, but don’t know how to scale their activities, in which case you could throw UNDP’s institutional weight behind an initiative and scale it quickly.
Johannes: And then you can go a step further and turn that to a ‘lab’ structure. What is a lab exactly?
George: It’s an incubator for social projects where we mimic business incubators: Identify ideas, conduct rapid prototyping, get a very basic product into the world, test it with users, and then look at the feedback. Is this working? Is this gaining traction? If yes, let’s invest more! And all the while you are giving the team access to mentors and design workshops, and develop their capacity. With a lab you can test multiple ideas and hypotheses at the same time, look for results, and then scale. This is very different to traditional programming approaches. The lab in Armenia attempts to solve big social challenges by harnessing citizens’ experiences, insights and ideas alongside public services. It applies approaches like horizon scanning, design challenges, user-innovation and service design.
Johannes: And what does it take? How much does it cost and how many people are involved?
George: Social Innovation Camps all-in-all cost about $25,000 here in Armenia. The event team should include an Event Coordinator working part-time, plus a full-time assistant, plus a full-time intern over the course of four months. And all that a Lab does is to extend this team people on an ongoing basis.
In terms of time you are looking at four months from the launch of the call for ideas until the event itself. During the preparation phase the team goes out running workshops, talking to people and developing ideas about the problems or challenges. Once the applications come in, they are also looking to find the experts that will complement the idea owners. For example, we had a doctor approach us who said she wanted to digitize Armenia’s blood registry. Of course she had no idea of how to build a database and how to build a web interface. She was just a doctor who understood the problem and had an idea how it could be fixed. So the organizing team then goes out and finds the extra skill sets and people needed to build a crude prototype of the initiative (in this case an online blood registry database) at the event.
Johannes: But even that is a lot of things to do in just four months.
Arndt: Well, in your case we may need to rethink the whole approach. For emergency response situations you need to get to solutions much more rapidly. It is almost like ignoring what Design Thinking says: You have to give it time through various iterations. But you are in a situation where you need quick solution which you can rapidly scale up. Maybe we need to design a process that can be more integrated into government and existing institutions, so we skip the part of building a team. Instead we do stakeholder engagement, and bring in NGOs, relief organizations and governments, host community and refugee representatives and come up with something that can be immediately picked up by the municipalities. Thereby you could crunch the time that you would usually need for the incubation.
George: But then, it is still an untested idea, and the whole point is to establish whether an idea works.
Arndt: Yes, you would still have to test that idea and develop it further.
George: Another idea that builds on Arndt’s could be to do an open call for existing initiatives that are already working on a pre-defined problem. If they have developed a prototype which has generated results we can help them scale up. You are basically looking to adopt alpha or beta products in the late stages of development. You can apply the same skill sets that you would need for a Social Innovation Camp, but you get to scale faster. The challenge is finding these existing initiatives.
Johannes: How do you identify who you can actually talk to? It’s not like you have in a Country Office a list ready with all the actors that you could potentially approach on specific issues.
George: Well, I would start with whatever list you have, and talk to them. First you talk about the problems they face and then you ask them who else they know in their network who is good at addressing this kind of issue, and contact them as well. Repeat this process until you find existing initiatives or citizens who are addressing the problems in creative ways.
Arndt: In your case in Amman, the approach is really a lot about user experiences, either from refugees that are coming from other countries, or the host communities. Think of who has the best information and can identify the crunch points for which you want to develop solutions. Then interview people and collect stories, so you do the sensing exercise in that particular environment that surrounds the issue you are looking at.
Johannes: In both the Armenia Social Innovation Camps and in the regional camps in Singapore, who did that sensing. Was it both of you, or did you hire consultants to do that?
Arndt: In our case it only happened in one camp, and I must say, that we didn’t go with the standard Social Innovation Camp approach because it didn’t do as much of sensing. It basically relied on the participants, the idea givers, to do that. Some of them had done a decent job, but a lot of them had not. The one camp that did this really quite well was the one in Singapore where we had a local counterpart institution that had the mandate to engage on this in the long term, and they did a fantastic job on this.
George: In the case of Armenia, we did it in a team of three: a consultant, an intern and me. The three of us travelled across the country delivering the sensing workshops. Because as Arndt rightly highlighted, you need to get a lot of different perspectives on the problem. I recommend keeping the team in-house instead of giving it to contractors, so you can develop and keep the knowledge, relationships and networks. Also, if you outsource there can be issues of quality assurance. It takes a little bit more work, but at the same time, it really supercharges your learning. We now have an entire list of organizations and people with whom we would never have developed relationships had we not run the event and the lab in-house.
Arndt: It sounds to me that if you want to do a three country project, it will be a quite elaborate process. If you want to create real value it make sense to invest some funds into it and have people do it full time.
Johannes: Do you think it is better to start off with only one country, rather than doing it in three countries at once?
George: It depends on how you are going to do it. If you do it in-house start with one country and make sure you build the skills within the team. But if you are going to outsource it, you might as well go big from the start.
Arndt: I would tend to say go big, because for the Sub-Regional Facility, that will prove your value.
Johannes: Well, it’s not like I am going to do this outreach myself, I am here on a detail assignment for another few weeks. But what I need to do is to put the train on the rails and a plan in place so the Facility can hire someone full time to take that plan and run with it.
Arndt: The other thing is: So far we only looked at the option of Social Innovation Camps at the grassroots level. The other possibility is to think of this rather as a public service initiative in support of an innovation-type exercise. You can look at examples of Policy Labs that were presented in at the Global Innovation Meeting in Montenegro, and now again at our event in Singapore, such as like the Mind Lab (Denmark) or designGov (Australia). The latter is more a governmental initiative which also applies a design thinking approach where it is government officials who do all the sensing. They have almost a purely government-driven lab, but they hire service providers which will accompany this process. I wouldn’t be surprised if some of Syria’s neighboring countries had a few companies providing design thinking consultancy services that you can hire to facilitate governments finding solutions. This is different than just coming up with a solution from a UNDP office and then developing a pilot, which is the traditional way UNDP does business. Instead, it means that people really go down to the ground level, get a sense of problems, go back and develop prototypes, test them out, and then scale up what works.
Johannes: In the case of Singapore, did the government already have something like that in place, or did they ask UNDP to help them establish the lab?
Arndt: Here in Singapore the Human Experience Lab (THE) Lab is a relatively recent initiative, while Denmark, Australia and UK have policy labs since while. The Singapore lab is only half a year old, but they have been practicing design thinking for a bit longer. Supporting such government labs is possibly an alternative option for UNDP, although these modalities don’t necessarily exclude each other. You could have a government-driven process going on, with social innovation camps as satellite events. That would be really nice because the innovation camps could feed into an official government-driven process and you could have a dialogue between the processes. That might actually be even better than each modality on its own. But of course you need more money for that.
Johannes: From your experience, how many of these solutions that come out of such modalities are IT-driven, like Web 2.0 or e-government solutions, and how many of them are non-tech, like processes or policies?
Arndt: The social innovation camps I’ve seen were almost all tech, but that is because we had specifically asked for tech. The output from the government lab was mostly non-tech, such as project re-engineering, cutting silos, cutting bureaucracy, finding new ways of communication with citizens, etc.
George: From its origin, the specific objective of social innovation camps is to apply Web 2.0 to social problems. So if you are looking beyond tech then you are looking at innovation challenges or design teams. Service design approaches do not involve crowd sourcing but can be a good way of generating lots of ideas for alternative approaches to service delivery. You certainly get non-tech solutions emerging from this. However, if you embrace an open process and just collect ideas, it actually doesn’t matter whether the solutions are IT-based or not. The point is: if you open up to many different inputs, you get good ideas.
Arndt: I second that. It is also easier to scale if the solution is non-tech.
Johannes: From all that you said, I understand that at one point we will have to make that national , regional or global call when you put the question out there who has a solution to a particular problem. How do you actually plan for that?
George: From our experience here in Armenia, once you have a sense of a problem, you can either do an open challenge where you say “tell us about problems and suggest solutions” or a slightly more specific one “tell us about problems in this sector and suggest solutions”, or you ask a really focused question like “tell us about solutions for this specific problem”. Once you have the question you build a small website and integrate social media into it. Whatever the social media platform of choice is in the country you are working in, focus on the top two. Then you have to go into different communities, both online and offline, and starting talking to people. You can’t just build a beautiful garden and expect people to come, you have to go out and engage with people where they are. Our intern worked very hard going into every single web forum in Armenia she could find, talking to people and looking for activists with whom we could work. Just talking to them at a personal level about the issues, and through that engagement process, a lot of hits were generated on our website, and then turned into applications for the event itself. You then complement that with a series of sensing workshops, where you are targeting specific groups of stakeholders. In your example you would talk to host communities, refugees, to state officials, municipal officials – especially in front line service delivery, and journalists. You invite them to a workshop where you ask them to define the problem more clearly. Or if you have already identified the question or challenge, you let them brainstorm on solutions. You also ask them “who do you know who is already working in this”, so you can meet them and see if you can support their work. From that engagement process you will get a lot of connections and ideas. Then you take the best 5 or 6 ideas into a prototyping event, much like a social innovation camp.
Johannes: Well, thank you so much guys, this all really helps our thinking here a lot. We will have to develop an action plan on this going forward. I will write this up, both in a blog but also in the form of some concept note for the Facility. And I’ll keep you updated on how I can inject those thoughts into what we are doing at the Facility, and we’ll be happy to share how far we will have come with it in a few months’ time!