9 different MVP techniques explained

MVP's are probably one of the most important thing on how to proceed with idea validation. That's the key point where you can save a lot of time and effort early in the process and gain insight on how things could be done. Also it might differ from people to people on how you understand the MVP/MVE/MVS/MMF. In this article I'll try to explain how I think about the MVP and the types of MVP's I've used in the history.

Photo by Nikita Kachanovsky on Unsplash

Dear child has many names

So that you wouldn't get confused, there are a lot of terminology involved. What are Minimum Viable Solution/Experiment/Product or minimum marketable feature? Do they differ? Usually people talk about the same thing in many ways and you really have to ask about who is talking, what it means. For example I've met software engineers talking about MVP as a ready built feature, that is release-ready and polished for customer use. In this case it was more of a definition of done rather than a minimal experiment.

What MVP should be? It's a tool for validating the core idea or core hypothesis. Yes, we have assumptions like "We can target customer group x for a lot of sales!" or even "Our customers want to buy towels, because they are swimming a lot!". As a PM you keep every one of these as assumptions an ideas until proven to be true. I think there's no product manager who hasn't run straight into assumption, done a LOT of work to make it happen and then realize that the assumption was wrong -  the customers wanted towels, but they didn't want to buy them - they wanted it to be included in the spa price. Otherwise they can just bring their own towels with them. Changes things a bit, would you say?

Different types of MVP's

So MVP is for failing early and learning from the real, authentic reactions in real environments. Its an art of it's own to do it subtle, fast and recurring. You can do a lot of MVP's in the time you do the full implementation and improve it over time really fast to provide the best results early.  The most important thing in running MVP is to define a measurable goal to see if it worked as it should. I'm not going too deep in running MVP's in this blog post, so let's go over the different types first, just remember that you aren't restricted to these. It's always about the CHEAPEST and EASIEST way to PROVE your assumptions! Use your imagination!

Micro surveys

Micro surveys are one great way to test new features in existing app/webpage. You've probably have seen these over the internet "Did this feature help you to tackle your problems? <enter text here>". These can be implemented in applications quite easily and there are even a lot of tools to use it. Surveys shouldn't contain more than one or max two questions - and keep them open ended to get feedback.

There are a few ways to implement short surveys: chatbots, survey tools, just a button.. even error reporting/feedback pages can be used, though they are usually a bit biased.

Pre-order

When you have an idea that it would work, but the price validation is still missing, a pre-order MVP is a really good tool for that. This is used many times to prove even the idea in crowdsourcing websites or through landing pages that have "campaing" of pre-ordering. Pre-order MVP is quite nice in a fact that you can limit the needed input and get the orders in line even before the product itself is ready. That's even how Oculus Rift started, a pre-order before anything even was in production!

Also a lot of e-commerce sites uses a variation of this by adding new products to websites as "coming soon" to measure the feedback and evaluate the product for interested audience.

Concierge MVP

This is one of my favorites! It's all about dropping off all the shitload-of-work automation components and replacing it with human touch. Really useful when doing a digital service for customers that you don't know are interested about your product. Let's say you want to provide weight loss programs to customers. Provide a landing page that asks the needed questions for the customer and so that they can place an order. You receive the questions in your e-mail and start crafting that diet manually. When it's complete, send the diet and material to the customer. If it starts to fly, the automation has to be there, but for the idea validation you can do it in some extent by yourself. Online book store Zappos started with this kind of test, by taking pictures from the local store and buying them as soon an order from real customer came in. So, almost anything can be replaced with a human - it's expensive in the long run, but easy to start with.

Wizard of Oz MVP 

This is really close to concierge. The difference is that in some cases there are a lot of background software, mechanics etc. already existing and they just aren't connected to the product itself. In Wizard of Oz MVP it seems for the customer that the website is doing all the work, when instead there is a human behind the scenes using the background software to provide the service. For example if you're doing a hotel booking site, you could add a text box on website that allows the customer to tell you what kind of hotel they want: "A queen sized bed in a hotel with at least 4 start rating between 2.-3.9.2019 from Joensuu, Finland that has breakfast, sauna and allows two large dogs. Wouldn't mind if it had nice places to eat nearby." - It's quite hard for computer to calculate, but I bet that I could find one with existing solutions and my networks quite nicely.

Video or presentation

Videos and presentations tend to always be a good starting point for any MVP. Provide a short video of a non-functional prototype or even a mock-up for your customers and spread the word around to your targeted assumed customer base. Get feedback through comments and iterate. Presentations can also be used if you're building a professional service like training, just provide the material in advance. Videos might be a bit costly, but can be done. One famous example for video without perfected implementation is Dropbox, where the creator Drew Houston created a video about what they had - a basic synchronization idea. With that video they could explode their beta-waiting list to 75000 "literally overnight".

Ghost/404/coming soon/fake door MVP

This is one of the most creative and reusable MVP in many different situations. I'd like to call it a ghost button MVP, but it has more names than I can say. The basic idea of this MVP is to provide an actionable item that leads to - well, nowhere. The idea is to measure if the users are interested in clicking the actionable item. If they are, it might be viable to implement. Usually the ending page is something between nice "404" so that it looks like an error on website and a nice message that says "Sorry, we're still doing this and if you're interested to find out when it's available, leave your e-mail address in the field below". Sounds familiar? I bet you have bumped into this earlier. One famous example used who used this was Buffer, who tested their pricing options using fake doors by providing a page including their payment options and when users clicked any of them it ended to a page describing "it's not ready yet". But they could prove, that people were willing to pay for their product!

Piecemeal MVP

Internet today is full of free tools. And combining them to give users experience of the final product in. For example Groupon used piecemeal MVP. They offered physical coupons for local shops by creating their pages on Wordpress and providing their PDF coupons to local shops by using FileMaker. Sending of the coupon was made automatically with AppleMail. The product itself wasn't full fledged, but it provided the value for customers right away. It's all about combining different software components together and add value on top of them. Especially nice when trying out totally new business ideas with low or zero budget.

Digital prototypes to help

I personally try to use these quite a lot. They're just no-code, "zero-functioning" prototypes that can show interactively how an application or website would work. These are pretty good on situations where you want to see how a feature would work and get feedback out of it. Paired with an interview about the customers problems and then allowing customers or stakeholders to use the prototype can really help up verifying the assumptions before doing anything. Prototypes don't give value to customers, so they're just for the show - pair it with video or presentation and you can use smoke and mirrors to prove your MVP. In manufacturing industry the digital prototype might be 3d model that could vision the product itself, and as I know, they use it a lot all the time!

Don't forget the physical prototypes 

I'm all in digital product management so I usually forget the physical aspects. Today there are a lot of ways to create physical prototypes: 3d-printing, casting, wood prototypes, foam modelling, you name it. These all can at best be applied to digital products, especially if you're working with IoT or other industry that has both physical and digital aspect. I have personally used a projector and whiteboard combination to create a prototype with a customer to prove if it works. Just project the software over a whiteboard and draw on top of it. It ended up being a combination of physical and digital world, and the prototype was really easy to comprehend and to communicate the needs from the customer to the team.

There are a lot more of these, but the text is getting so long, that I'd better stop! More of MVP techniques or supporting techniques to idea validation I like to use are interviews/social media feedback, service design workshops, blogs/articles, landing page and of course split testing. Let's talk about those later on. The aim of this post was to give you a good impression of different kinds of MVP's, hopefully it served it's purpose!