My team and I have been busy with our MVP product, Gemoo Beta, for the past six months. Recently, we are about to launch the official version of Gemoo. Therefore, I think it’s time to share some of my simple insights on the development and validation process of MVP based on our journey. I hope it can be helpful to your startup team.
The process of MVP development is all about the cycle of “Hypothesize-Validate-Feedback-Iterate”. And commonly the process can be split into the following five actionable steps:
Step 1. Focus on the Problem You’re Trying to Solve🔍
We only want to solve a problem if we believe it’s valuable, and that’s the hypothesis we need to validate. There are two points that need to be validated:
✅If the problem itself is valuable?
✅Is our solution valuable?
More specifically, what needs to be defined clearly is:
Step 2. Create a User Flow ⏳
Think clearly about what users need to do to solve this problem, and what the core behavior is.
Step 3. Build a List of Features 📝📋
To support the problem users need to solve, what related features does the product need to provide? We can brainstorm first and list all the relevant features.
Step 4. Prioritize 🚩
Prioritize the defined product features, combined with the coverage of user scope, frequency of use, value delivered, problems or risks if not implemented, and implementation costs.
Step 5. Start validation – Build, Measure, Iterate⛵
Once the MVP is ready, it's time to ship it out to your early adopters. These early adopters will have plenty of highly valuable feedback regarding your app – feedback is the fuel for the next iterations of your product, no matter if your MVP is internal or external.
Also, start collecting data and analyzing it from the moment your first user launches your app. Data analytics together with user feedback from tickets and reviews will be the foundation for further development.
Once your main hypothesis is validated and your product gains traction, don’t throw whatever money comes into rapid feature expansion. Make new feature assumptions based on the feedback and data — before working on developing new features for the next product iterations, validating them first.
There are two approaches: one is to build the MVP and the other is to "do nothing at all".
That is to say, we could build a version that contains only the minimum necessary features and cut out other features. (Just like what Gemoo did)
Alternatively, we could try a small event to test the waters, validate it through a simple webpage, and then improve it for a permanent feature. Some products also do crowdfunding or presales. Regardless, all of these methods are about finding ways to validate assumptions with the smallest possible cost.
Although MVP is a great way to develop a product for startups, it also has its limitations.
👉Firstly, MVP solutions may be rough and may not receive accurate feedback. Additionally, a small sample size during testing can affect the results’ accuracy, which can then affect subsequent decisions.
Assuming there is a need for further optimization and iteration, we may not know if the problem lies with the problem itself or with the current solution to the problem, and this will require continued experimentation.
👉Secondly, MVP is generally suitable for short iteration cycles, low iteration costs, and a repeated transaction model; if the cycle is too long and it is difficult to see the test results or if the cost is too high, MVP may not be worthwhile.
I understand MVP as a change in mindset from “I am right” to “How do I know I am right?”. Before verifying an idea, keep changes as much as possible and find ways to verify the idea with low costs. Our goal is to try as much as possible at low costs without being eliminated.
Hope my sharing could bring you some inspiration, and you can also check out the full story with more examples here.
What are your thoughts on MVP? Any experience to share?👀
So when is a MVP too simple or complex. For some products some MVP's definitely need to be more involved vs others. For example, I'm building an app to track credit card points but it feels like without handling all the rules of each credit card, a lot of value is lost and you lose a bunch of potential users.
I tried for a "beta" launch but I think one of the reasons I'm not getting as much traction as I would like is due to the fact the data was incomplete in some ways. When's the line to actually just release the MVP?
From what I learned, MVP is not just a one-round thing, you may need to make multiple MVPs for different key assumptions you want to validate. Each time, focus on one thing to validate. So the main question is not about the MVP, but about breaking down the final idea into several assumptions and proving them one by one using different "MVPs", just like the Groupon way. (BTW, the MVP does not have to be an actual product, it can be as simple as a slide or prototype, or combine a simple landing page with "the manual backend". )
Your product is very similar to Loom.
They are a giant in this space and there are other players that have branded differently(e.g screencastify that targets schools in the US).
It seems you will be going head-to-head with Loom. How do you intend to gain PMF?
This is something I have wanted to build specifically for the African tech market, but my research showed that customers were not ready for async operations in this market.
Thank you for checking out Gemoo! Yes, it's similar to Loom in screen recording part. But Gemoo is more than just a screen recorder, it also has the feature of screenshot capturing and online documents, and what differentiates Gemoo from others is its ability to organize all these files in a custom view, mindmap view, calendar view, slide view, and more views to come. But I have to say, it turns out the UVP is a bit too post-positioned to be felt by users (a big lesson I learned from my user interviews during the MVP journey), and now I'm trying to figure out a solution to this.
This is helpful thanks for sharing. How long did it take to go from idea to mvp? And how were your steps afterwards in finding PMF?
Glad you find it helpful. It took me around 8 months to go from a rough idea to the MVP release. I've shared my lessons learned on the journey to finding PMF in this post before, you can chekt it out here: https://www.indiehackers.com/post/lessons-i-learned-on-my-journey-to-finding-product-market-fit-hope-you-can-avoid-them-89a4ece9f1 (But I'm still on the way to find PMF, and will keep sharing my lessons and experience during the journey)