For those of you who haven’t read the Lean Startup by Eric Ries, the basic concept behind it is that we have been going about the startup process all wrong. Instead of writing long business plans, raising early funding, and building initial prototypes over the course of a year, we need to foster startups that constantly learn and iterate around a minimal viable product. By constantly tweaking, or occasionally pivoting the product, service, and business model in response to early customer feedback, startups can improve their likelihood of success or at least fail faster and cheaper.
I have spent the past few years immersed in sustainable and smart cities. So naturally, I recently began thinking about how to merge the concept of lean startups with smart cities.
To me, the idea of a Lean City is completely complementary to the idea of smart cities. Lean startup principles suggest that innovators should develop a hypothesis about likely reactions to a minimum viable product and be prepared to rigorously measure the results. Smart city solutions frequently involve the use of sensors and real-time data to enable city staff to monitor key metrics and modify systems to improve performance. For example, I recently wrote about a new city development in Portugal that will make use of over 100 million sensors for a planned population of only 225,000.
While I prefer a broader definition of smart cities than just "the ample use of sensors and data," the decreasing cost and increasing availability of sensors suggests that these devices will grow in importance in smart cities around the globe. This trend is perfectly aligned with lean startup principles.
Another key component of lean startups is the creation and iteration of minimum viable products. A minimum viable product (MVP) is the most efficient, minimum product or service that can be developed to test a hypothesis about how users will interact with the innovation.
Cities are increasingly making use of MVPs, although they are commonly referred to as pilot projects. Several months ago, I wrote about San Jose’s framework for demonstration projects, which seeks to streamline the use of pilot projects (MVPs) to help the city innovate and support local sustainable economic development.
Let’s quickly apply these two key tenets of lean startups--hypothesis testing and measurement, and the use of MVPs--to the context of smart cities.
Until March of this year, I lived in Mount Pleasant, a residential neighborhood with commercial corridors only a few minutes by bike ride to downtown Vancouver. Led by its visionary mayor, Gregor Robertson, Vancouver seeks to become the greenest city in the world by 2020.
Mayor Robertson’s team has made a concerted effort to prioritize pedestrian and cycling use over vehicle use. This policy of course has its detractors, although I am not one of them. Aside from removing parking spaces in downtown Vancouver and using a lane on a major bridge to support increased cycling infrastructure, the city has begun to experiment with other MVPs, as well. In Mount Pleasant, for instance, the city decided to experiment with the replacement of two parking stalls on one of the commercial corridors with park benches, in the hopes of animating the sidewalk and the community.
If Vancouver were to implement lean startup methodologies in this process, it might look something like this.
- Develop a hypothesis: By systematically and strategically removing parking spaces throughout the city and replacing them with green spaces and/or community spaces, we will increase the amount of residents interacting with each other on the same city block.
- Determine a set of metrics to test the hypothesis: Measure at different times of the day, on weekdays and weekends, the number of residents on the street before and after the test project. They could also measure the amount of time residents stay in the area before and after.
- Develop an MVP: In this case, I think the city did a good job with a very low-cost, low-impact test project (instead of trying to install 1,000 benches throughout the city all at once in the hopes that it would work).
- Measure the results: Using low-cost sensors and perhaps observers or even using a mobile app, apply the metrics in step two.
- Iterate: Leveraging the analysis in step four, experiment with similar models. For example, what would happen in the same location if they converted the space to a few stationary bikes or a mini art exhibit?
- Measure the results and create another MVP.
Once the process has achieved the target improvement in community interaction, it may be time to expand the program to other neighborhoods.
Smart cities aspire to be efficient with taxpayer dollars and resources. Applying lean startup thinking to cities could be a useful tool to achieve increased efficiency and improve the quality of life of city residents. It could also support innovation in procurement practices, which have the potential to encourage local innovation by reducing the bureaucracy that small companies have to deal with to participate in city innovation. This of course is another hypothesis to be tested and iterated.