Many years ago, I came up with an idea for a simple software application that I could sell online. Basically, an open source app I created for modifying your voice in Skype became incredibly popular and had millions of downloads, and it struck me that I could probably sell a "Pro" version with some additional features.
Of course what actually happened was that the code for the "Pro" version stagnated for years as my enthusiasm for this project came in fits and starts, and I couldn't make up my mind exactly how I wanted to build the thing, and wasn't even sure anyone would buy it. So the idea was going nowhere.
But about a year ago I started listening to the entreprogrammers podcast, and it inspired me to resurrect this idea. The only trouble was, would I be able to find the time to actually complete the thing, given that I was already spending a lot of my free time making courses for Pluralsight?
I soon realised that if I were to ever get something out the door I would need to pick a realistic and achievable feature set for the first version, and focus exclusively on completing that. In other words I needed to stop trying to make the perfect app, and create what's known as a "Minimum Viable Product".
Minimum Feature Set
For my application there were three main features I wanted to deliver:
- You can change your voice in a Skype conversation
- You can play back pre-recorded sounds for the other person to hear
- You can record your Skype conversations
For each of these features, I decided on what the bare minimum capabilities were that I would be happy launching with. For example, I wanted several cool voice effects, but I didn't really need the effects parameters to be adjustable. I wanted the ability to replay sounds, but I didn't really need looping and repositioning within those sounds for version 1. And I wanted to offer recording to MP3, but for version 1, recording to WAV would probably suffice.
By deciding on a minimum feature set for my application, I finally had an achievable goal and was able to focus on getting the job done.
Minimum Lovable Product
I've heard some criticism of the "minimum viable product" concept from people saying that what you really need to create is a "minimum lovable product". In other words, if your application is too stripped down and basic, you may find that no one is willing to buy it, and your potential customers go off in search of alternative solutions. So while I did go for a fairly basic feature set, I did attempt to ensure that I had just enough in each of my three key features to offer customers something genuinely fun and useful.
Minimum Deliverable Product
Unfortunately though, completing the minimum viable feature set of my application was the easy part. There is a whole host of additional stuff that needs to be done just to sell a single piece of software. I needed an installer, a logo, a website, a domain name, some help documentation and tutorials, an email address to handle sales and support enquiries, a way of selling the software that complies with the new and arcane EU VAT rules, a way of refunding people, a way of generating software licenses, and a way to handle errors gracefully within the application.
It turned out that it was these tasks rather than the core feature set that were preventing me from completing the project. So I decided I needed to take the "minimum viable product" approach with each of them too.
What's the "minimum viable website"? I created something very simple (and ugly) in bootstrap and ASP.NET web-pages. A few basic help pages, and the ability to install and buy were all that was really needed. It doesn't look pretty but I can improve it gradually or even hire a designer at a later date.
What's the "minimum viable installer"? I opted for ClickOnce, since it's quick to make an installer and easy to keep customers updated with new features. Even that turned into a bigger task than I wanted as the hassle I went through to purchase a code signing certificate was much greater than I anticipated.
What's the "minimum viable licensing mechanism"? Here I may have made a mistake. I opted for portable.licensing,which itself is great, but that meant I had to create an automated license creation and emailing system. Maybe I should have just decided not to worry about pirating, and save myself time by not using licenses.
What's the "minimum viable checkout"? Well a "buy now" PayPal button was my plan, but thanks to the new EU laws on VAT collection I needed a complicated system of verifying the locations of my customers and charging them the correct VAT. So I decided to find a company who would collect and report VAT for me. After a false start with Digital River's MyCommerce, who disabled my account and refused to respond to any emails, I found paddle.com whose customer support was much better, and offered straightforward integration for my own license generation code (turns out I could have used their built-in license generation capabilities but I was too far down the road with my own by this point).
So actually going from a working prototype to a deliverable product is quite a journey. I realised that if I was ever going to complete the task, I'd need a way to motivate myself. So what I did was put an advert in my open source software for the "Pro" version. It would simply tell users that if they wanted to record their calls they could upgrade. The advert took them to a "coming soon" page hosted on my blog, where they could register their interest to be notified when the pro version was released, and they could purchase a pre-order license for a discounted price.
My email list grew quite rapidly - I collected 1400 email addresses in just over a year. And I had six people pre-order the product for $30 (actually seven pre-orders, but one wanted a refund when he realised it wasn't released yet). The very fact that I had some customers who had paid for the product helped motivate me to actually complete the task.
So finally, years later than I should have done it, I went live with my site a couple of days ago. I've still got to email all the people who registered pre-order interest, and offer them a discount. I'm not sure what my "conversion rate" will be, but with 1,400 emails I'm hoping at least a few people will buy a license.
Obviously going live is only the beginning. There are lots of tweaks and enhancements to my app and website that I deliberately held off doing, just so I could get my first version out of the door. But that's the point of the minimum viable product. Just get something released, and then you can gauge how much further time and effort to invest, depending on whether there is actually any interest.
At the time of writing I'm still waiting for my first customer from the new site (it's early days yet, but I suspect my marketing and sales ineptitude isn't helping). But if you have a use for a Skype Voice Changer or a Skype call recorder, then why not try it out? And if you'd like to be my first customer, then I'm offering readers of my blog a 25% discount. Just use the SOUNDCODE coupon code when you purchase.