We have an idea, and we choose to go to an MVP. What's next? Use NASA's example and processes to successfully go through all phases of creating a new software or hardware product.
Creating a product, regardless of whether it is software or hardware, is complicated and complex. It is an ideal environment for creating countless communication, technological, and quality errors. The only remedy can be an efficient process, thanks to which we can be sure that when we start the next phase of work on the product, we have the maximum knowledge and confidence that we are going in the right direction.
Through a series of articles, I want to show our process from practitioners at Untitled Kingdom. Today I will present it briefly from A to Z, or from (R)easearch to (M)VP, while the next time, I will focus on each phase separately.
Desperate times call for desperate measures
Let's leave Earth for a moment and return to the title where the Moon is missing. Maybe you associate the famous podcast 13 Minutes to the Moon, in which stories from the NASA mission to the Moon are told. If you treat it as a regular project, we get an interesting explosive mixture, and at the same time an example of working in an uncertain, changeable, and unrecognized environment:
- in the mission control center sits an incredibly young team, with an average age of 27,
- the team is not limited by schematic thinking, does not use beaten paths, has a fresh approach and faith that nothing is impossible,
- the team is facing extremely ambitious and unprecedented tasks,
- they are all divided into subgroups as distributed teams, operating independently,
- the essence is many iterations, and subsequent attempts are made very quickly and verified by errors in a short time,
- what matters is the speed of response to problems and searching for possible solutions,
- perfect organization prevails, refined launch status check (aka “go / no go poll”) allows minimizing the risk of failure.
Some believe in comic superheroes, but for most of us at Untitled Kingdom, NASA people were and are the real idols. Why? We implement their approach to work at many levels in software development projects, but it is crucial when creating hardware products from scratch to MVP. Sometimes it really is rocket science.
“Desperate times call for desperate measures”. It is worth remembering this every day, but currently, in the face of rapid changes on the market and the vision of the crisis, it is especially important to minimize losses and iterate quickly.
For this reason, at Untitled Kingdom we have developed our own project status check, with subsequent phases of software and/or hardware prototype creation:
- Proof of concept
- Feasibility study
- Design & development
- EVT, DVT, PVT (hardware only)
Let's go through them briefly. However, we'll start with the fuel that allows our spaceship to take off. This fuel is the approach.
Rapid prototyping approach. You’ve got the idea? We’ve got a process to prove it
Often, in theory, an idea looks good, but when we start working with it, it turns out that we have not anticipated the enormity of factors. Hence, we need to get back to the starting point, and as a result, we lose time, money and energy.
We are proponents of software & hardware rapid prototyping and our most important goal here is the greatest idea of exploration and providing an experience, not a document (and we save trees, right?).
When creating solutions, especially those that are unprecedented, we must assume high risk, so in order not to invest huge financial resources in an idea that will be a failure, it is best to test it:
- very fast, iteration after iteration,
- with the lowest budget possible,
- with as many tests and errors as possible,
- with the most frequent feedback from users.
What allows you to go through the entire process of rapid prototyping from idea to MVP successfully are three factors:
What's the most important is extensive collaboration and coordination between specialists in various fields, including design, mobile, software, hardware engineers, QA, product development experts.
We have 12 years of experience on the market as a software and product development company. During software projects, we often had to create simulators, test hardware solutions, which is why we organically developed the competences of electronic, mechanical, and industrial design at Untitled Kingdom under the banner of the Untitled Lab.
Prototyping involves testing solutions that we are not sure are possible. Therefore, we cannot assume that we have all the necessary equipment in the Untitled Lab.
Therefore, to create specific components and prototypes, we use previously built relationships with several partners from San Francisco, London, Krakow and Shenzhen, so we can manage the entire process and commission models and components to create if needed.
So what happens right after starting cooperation with Untitled Kingdom?
From research to MVP in a nutshell. The process of creating software & hardware prototypes
We can start work from the very beginning, but if you are already in the prototyping phase or just before the release of MVP - we can examine the needs and adapt the process so as not to start from scratch.
Phase 1. Research
We choose to go to the moon in this decade and do the other things (...)*
We start working on a specific idea for software or hardware at the stage where:
- we conduct competitor research on the market,
- we conduct scientific and technological research,
- we acquire knowledge and consult with domain experts,
- we generate as many ideas as possible during brainstorming,
- we build initial theses to test and prove or refute in subsequent phases of software or hardware prototyping,
- we talk with potential users.
Phase 2. Proof of concept
(...) not because they are easy (...)
After the research phase, we begin to develop Proof of Concept (PoC), which in the case of hardware may be the innovative use of a component, the first version of the algorithm, a study of accuracy or sensor measurement. This is one of the smaller parts of the project, but the key one that shows us whether further work will make sense:
- we verify theses
- we decide if the chosen direction is possible to continue,
- we investigate what technology COULD work.
Phase 3. Prototyping
(...) but because they are hard (...)
After the initial verification of theses and choosing a possible architecture, we go to software or hardware rapid prototyping, i.e.
- we check in reality whether a given technology and architecture work,
- in the case of hardware, a "bettered", working, "roughly" functional version of the product is created, tailored to customer requirements, which is not even close to being market-ready (box of wires),
- some parts might be simulated/mocked - we want to understand general architecture,
- it is a base for trying other functionalities,
- we conduct the first tests with users, collect feedback.
Phase 4. Feasibility study
(...) because that goal will serve to organize and measure the best of our energies and skills (...)
After checking how the prototype works, we proceed to the second, complementary research, based on specific solutions:
- we research the final available parts and solutions,
- we check technical feasibility, i.e., "evaluation of the hardware and software and how it meets the need of the proposed system",
- we check legal feasibility, i.e., required certifications or data management,
- calculation of approximate costs of certification, software or hardware development, early BOM (bill of materials),
- we develop time feasibility, i.e., a roadmap,
- we develop a risk analysis.
Phase 5. Design & development
(...) because that challenge is one that we are willing to accept (...)
In the next phase, we iteratively approach the finished product:
- we carry out product work on the design and the combination of software and hardware,
- we build design prototypes,
- if we lack any skills, at this stage, we use an extensive network of Partners responsible for manufacturing, electrical engineering, firmware, industrial design etc.
Phase 6. EVT, DVT, PVT
(...) one we are unwilling to postpone (...)
In the penultimate phase, which concerns only hardware, we go through three seemingly difficult to decipher abbreviations, i.e.
- Engineering Validation Test is the stage where we make a batch of 20-50 product units that are intended for testing. These are looks-like and works-like prototypes; this is a critical moment in which the last shortcomings in the operation of hardware are caught.
- Design Validation Test is a stage in which, after checking the operation of the prototype, 50-200 units are created for broader testing and the last cosmetic corrections. It is also the moment of performing functional tests in various conditions - on fire, water, after falling from a height.
- Production Validation Testing is the first stage of official production. 5-10% of the first units are made, which undergo the last tests but should be ready for release on the market.
Phase 7. MVP
(...) and one which we intend to win (...).
In the last phase, right after Production Validation Testing, we are ready to celebrate and first public release our Minimum Viable Product into the world (and space).
At this stage, we are finishing the development of MVP, starting with the idea, but the fun with product & software development is just beginning here, just as setting foot on the Moon was only the beginning of space conquests.
In the article I wanted to briefly present our own project status check at Untitled Kingdom, inspired by NASA, along with the subsequent phases of creating a software and/or hardware prototype from a bird's eye view:
- Proof of concept
- Feasibility study
- Design & development
- EVT, DVT, PVT (hardware only)
In subsequent blogposts I will deepen each of the phases, present our approach, and share our experience.
Thanks for reading so far and if you have any feedback or question, I will gladly answer them: email@example.com
* All quotes are from President John Kennedy’s Rice Stadium moon speech: https://er.jsc.nasa.gov/seh/ricetalk.htm