Bringing MVP Culture to Every Product Feature
What “lean” means for us in product development
I very firmly include myself in the group of people converted to the “lean” product philosophy, but what does this word really mean? Well for the broader context, lean derives from post-war Japanese culture, where limited resources were the standard. Working in these conditions – a situation experienced by many product teams today – a huge value was placed on cutting waste, efficiency, and doing more with less.
How do we apply this important concept in the day-to-day development of our products at Resultados Digitais? First of all, it’s so integral to our development approach that it’s officially written into our Culture Code. But to mobilize these ideas in real product management processes, we use a Kanban Continuous Flow to feed epics and projects into product releases and team sprints.
The flow we use takes every new development – including the smallest of feature ideas – through from Study & Benchmarking, Minimum Viable Product, and Design (solution). This way we’re certain that all of our efforts are channelled into lean, efficient processes to determine our next solutions for development.
How does the MVP stage fit in to product development?
The MVP phase, Minimum Viable Product, is one of the core elements of the Lean StartUp movement. We keep the name MVP in reference to this well-founded methodology, but in practice it might be better understood as MVF (Minimum Viable Feature). The culture of lean product development doesn’t stop after launch, but remains relevant throughout the evolution of your product, feature by feature.
To understand MVP or MVF it’s important to understand where that fits into a wider process of building, measurement and learning. In the very first stage – Study & Benchmarking – we collect and analyze data on hypotheses for solutions to our problems; this includes existing solutions to these problems (the benchmarking aspect), as precious resources can’t be wasted making the same mistakes others have already learned from. This research component into what we consider committing resources to testing is very important. Resultados Digitais has a complex product with hundreds of different possibilities for its evolution and development. We have to be very conscious in our choices for what to add to our roadmap. Any MVP must start with a list of problems that the project needs to address, and a list of success criteria the project should achieve by resolving them. In sum; a business case.
When we’ve identified a reasonable hypothesis in the Study & Benchmarking phase, this is passed to the MVP (or MVF phase). Here, the project or epic owner must – with very few resources – develop a cheap and feasible solution to test the value of that proposal. In order to make this a real test, this minimum viable product or feature has to be measured with data in order to present defensible, quantitative arguments for its full development. Either success criteria are retained, and we progress forward to the Design phase, or we simply fail early and devise a different minimum solution to solve the same problem. Lean product development is all about looping. If the first hypothesis won’t work, it doesn’t mean you have to despair and start all over. Review, improve and test again.
What does an MVP look like?
It’s all in the word minimal. And MVP or MVF is a quick and cheap development and the most basic of tests that we can build whilst still proving that a proposed and final solution has the potential (or doesn’t) to achieve its success criteria. You see, the MVF doesn’t need to prove the success criteria of the final solution definitively – and it doesn’t have to use the exact same success criteria as your final proposed solution – but just enough to demonstrate that your final results are attainable. The indicators to measure the final solution and the MVF may differ amongst themselves.
Many different product experts use the cupcake analogy to explain this process. While the resources ploughed into a 10kg birthday cake are much too high for a test, a much simpler (minimal) version of the birthday cake – the cupcake – allows me to test whether my son and his friend enjoy it. Although not a definitive proof of the success of the birthday cake, the cupcake has proved to me that my strawberry cake solution has a strong chance for success at my son’s party.
How can you develop something equally quick, cheap and measurable? It’s certainly not easy, and here’s where the creative and highly Lean thinking part comes in. Your MVP might not even be a software. Use your imagination to create something that proves the value of your solution. Anything goes as long as it’s measurable; but if you can’t measure it, don’t even start it.
A good analogy to keep in mind, that’s perhaps a bit more intuitive than MVP is KISS – “Keep It Simple Stupid!” Although timeframes will vary from project to project good MVP can probably be built in just one day.
How has the MVP really helped our products?
This phase in our product flow has clear effects on the quality of what we prioritize in our roadmap. Our entire product growth is focused on clear and assertive metrics. By testing and validating solutions in their very genesis, we are becoming ever more sure that we will achieve our expected success at the end of a project, which is only reinforced by discarding solutions that have been proved ineffective before ever developing them. And the lean approach only becomes more effective with time; we can now be much more assertive with metrics like product engagement and medium ticket, and we have a much better idea of what to expect in the future from what is planned in our roadmap.
Another very important aspect of this phase is the “spread” of the Lean culture to the team. Lean principles may still seem counterintuitive for many, but by experiencing the consequential results of a Lean process directly, people end up better incorporating such a culture. We have been able to establish a virtuous cycle where the product process develops a better product culture, which in turn is essential to the continued improvement of this process.