The second concept that software teams take pride in is the Minimum Viable Product (mvp), which seeks to answer the question: what is the cheapest and fastest product that we can build to help us to learn something of value? My observation of software teams is that we use mvp as a scope management paradigm. When we have to do many things in the backlog and time/budgets are limited, we identify the mvp. Once again, the intent of releasing a usable product under given constraints is commendable, but an mvp defined based on the two dimensions of time and budget isn't really an mvp.
In early 2016, coach and consultant Henrik Kniberg came up with this drawing (https://blog.crisp.se/2016/01/25/henrikkniberg/making-sense-of-mvp). It has caught on quite well and is popular among Agile practitioners. As Henrik writes in his blog, "The picture is a metaphor. It is not about actual car development, it is about product development in general, using a car as a metaphor."
The preceding diagram shows Agile and lean development and the idea of the mvp. It is a good representation of a shifting mindset in software teams. The "not like this" section essentially highlights the futility of the approach, where we release a login story. The equivalent here is that we released a wheel. It brings out the flaw in the iterative software development mindset. In Henrik's own words (from his blog), "…. the first row represents 'bastardized Agile.' Technically, it might be incremental and iterative delivery, but the absence of an actual feedback loop makes it very risky—and definitely not agile." He then goes on to explain how this should actually be done and what real mvp thinking should be. This is the second half of the picture titled "like this!"
Each iteration in this picture is delivering a product. Henrik prefers to call it the "Earliest Testable Product." When the goal is to travel from point A to point B, then the first iteration is the skateboard version. It's quick to build. It meets the goal. It can be built fast. It can be built cheap(er). It is usable.
The skateboard fits the mvp paradigm well:
- It is minimal, requires minimal resources, and it doesn't have frills. It has only what is needed to meet the goal
- It is viable. The underlying technology of travel is viable. (This means that with two pairs of wheels, each connected by an axle that can turn the wheels. The product can carry a passenger from point A to point B, without harming the passenger.)
- It is a product. It is usable. Someone will pay for this product.
It sounds perfect. Well, almost. The skateboard answers only the "what is the cheapest and fastest product that we can build?" part of the mvp question. The second part about learning something of value is sort of lost.
Henrik's picture and its description of the mvp model is not really a practical representation of how a product can be launched in the market. I'm choosing this picture only because it succinctly captures the current mvp thinking. My critique is not of the picture, but of the mvp thinking itself in general. My first objection is with the problem statement. Business goals are not defined as "let's find the fastest way to travel from point A to point B." This statement is in the realm of research. Businesses have something more tangible. The goal statements are born out of something like: "I want to manufacture the most luxurious cars that will make people desire car travel, and have families look forward to long drives and road trips." For instance, the Tata Nano was meant to be an affordable family car that could be within the reach of Indian families who were then travelling on bikes. So, why would we build a skateboard to meet such a goal?
The questions that the mvp seeks to answer are as follows:
- Is there a problem worth solving? Will anyone pay for a solution to the problem?
- Is our technology viable? Does it solve the problem well for the target customer segment?
- Is our business idea viable? Does it garner sufficient interest from the target customer segment?
- Will our riskiest proposition nullify our biggest hypotheses? Will the failure of technology/business viability kill our business?
The mvp should help us decide between persisting or pivoting to a different target segment or a new business model. However, we still need to start with solving problems for a narrow target segment. mvp is not a proof of concept, nor is it a full-fledged product. It was meant to be the smallest thing we can put together to answer the preceding questions. In some cases, the mvp is not even a product. If technology viability is already proven, then let's not waste time building anything. We should rather spend all our effort validating the business idea and the riskiest proposition for the business model.
Can a skateboard help us to learn anything of value when we want to build a car? Will a consumer who wants to enjoy a luxury trip with their family adopt and give us valuable feedback about a skateboard? If you were a car manufacturer, would you launch a skateboard to your consumers? You might argue that Tatas wouldn't build a skateboard because there already are cars in the market. Well, exactly, but did the first car makers build a skateboard? What did car makers Benz and Daimler build? Was there a skateboard in the product development for the first cars? Not really.