10 shades of MVP (or: how to develop a product without developing a product...)



10 shades of MVP (or: how to develop a product without developing a product...)

A minimum viable product (MVP) is often perceived as a subset of the real product: a minimized version, lacking some features or missing some UI fine-tunes, but in fact, an MVP is more of a tool to test the core idea of what’s intended to be a product someday.

While the name MVP suggests that it has to be both a product and a viable thing - the software industry has proven it doesn’t have to be the case. 

In fact, an MVP doesn’t have to be a product at all: it can be an email, a Facebook group, a service, or a bunch of processes performed manually.

An MVP is there to help you test your business riskiest assumptions, see if your product can provide enough value to attract customers, and collect some feedback that can guide you through the product development.
And last, people should be willing to pay for it (with real money or some level of commitment) - otherwise, it’s just theoretical exercise that cannot prove that the idea is commercially viable.

An incomplete list of MVP types:  

Here are 10 types of MVP followed by some successful examples of how they were deployed by famous companies.
Although the examples talk about new products - these methods can be used to test new features in existing products without getting sucked into costly development cycles.

1. Concierge MVP

A concierge MVP can be a good way to begin if you understand the business needs well enough and have a clear vision of what the product should do, but don’t have the resources (time, money or budget) to develop it yet.

Concierge MVP - 10 shades of MVP (or: how to develop a product without developing a product...)

With a concierge MVP, you validate your product idea by providing the main service(s) completely manually (but not necessarily for free) to a small group of early adopters while delivering a VIP experience, observing their actions, and collecting their authentic inputs.

This process requires the involvement of the product people (founders or others), but it can pay off because it saves most of the development effort, and more importantly: it helps the product people better understand their potential users: what are their pains, how do they act in real-life scenarios, and how valid are their business assumptions.

Examples:

Food on the table started as a food butler (kind of…)

Food on the table was a mobile app that collected users’ food preferences and then suggested recipes and grocery stores with the best deals for those recipes.
This “little" product required some serious development: mobile apps, logic, integration with stores and much more, so the founders started with a concierge MVP instead: they found a few early adopters and personally visited their homes once a week to learn and collect their food preferences.
Then, they manually created the recipes, purchased the relevant groceries and delivered them to the early adopters in person.
“Quite an effort, but if you think that’s costly, try software development… “
Over time, more customers joined the service and pieces of technology were added to the puzzle: online payments, recipe lists over email, integration with store lists and more.
This example demonstrates how the product value can be delivered long before the product technology exists. This MVP was presented as a VIP service and the founders charged money from day one, which is another important way to demonstrate a product/market fit.

Yipit MVP was developed in 3 days

Yipit used to aggregate daily deals being sold by different companies and put them in one place.
Building the deals database required developing sophisticated crawlers and a classification algorithm, so the founders skip the technology part and took a manual shortcut: instead of developing a crawler, the co-founders would crawl out of bed at 3 am, search for attractive deals, and manually enter them into the database.
The whole “algorithm” was done manually…

[OK, seems like both examples came from companies that no longer exist. While that might look like a bad sign, I guarantee you that there's nothing wrong with this type of MVP. Nevertheless, let's move on to the next types...]


Before you continue, check out these 5 basic mistakes product managers still make and have nothing to do with features.

2. Piecemeal MVP

A piecemeal MVP is built out of existing components from multiple sources which are put together to create the foundation for a product.
In today’s world, it’s fairly easy to connect a few tools like a chat, files collaboration, online registration forms, photo sharing, and others to create some sort of product experience. Sometimes it's enough to deliver the product’s core value without involving a lot of development.

Piecemeal MVP - 10 shades of MVP (or: how to develop a product without developing a product...)

Examples:

Groupon started as a Wordpress blog

The Groupon’s MVP was powered by Wordpress Blog since its founders didn’t expect to get any revenue from their project.
The website had daily deals, published as simple blog posts with the relevant details of each deal.
The coupons for the deals were generated manually in PDF files (using FileMaker) and were sent to the clients as email attachments.
Groupon focused on promoting the blog, delivering the core value of the product without using any special technology.
Once the founders felt there’s a strong product/market fit, they started developing their real technology.

Product Hunt started as an Email list

Product Hunt's 20-minute MVP started as an email list.
The concept was simple: to build a community for product people to share, discover, and discuss new and interesting products.
To check if anyone cares about such a community, Ryan Hoover used a service called Linkydink to invite people to his group. Once he had over 150 subscribers, he started collecting product recommendations from 30 hand-picked thought leaders such as startup founders, VCs, and bloggers.
You can more about it in here.

AngelList started as Email introductions

AngelList is a marketplace connecting startups and investors.
The MVP included Email intros to investors (sent by the founders) and basically provided the core value of the service without requiring any development at all.
Once the potential was obvious - the team started developing the marketplace.

Check out the YOUR most popular posts in here: 
The best of the Mobile Spoon 

3. Wizard of Oz

A wizard of Oz MVP is quite similar to the piecemeal MVP, except for the fact that the users don’t know the product is not yet ready.
This type of an MVP usually requires some development, at least on the front-end level, so users can interact with it, thinking that it’s a well-established product.
While the user experience should be close-to-genuine, the rest can be handled manually by people (i.e concierge) or using a set of tools (i.e piecemeal), but it’s done behind the scenes.

Wizard of Oz MVP - 10 shades of MVP (or: how to develop a product without developing a product...)

Examples:

Zappos started selling shoes without having shoes in stock

To test his basic assumption (that people will want to buy shoes online) without investing a great deal of money in shoes, Nick Swinmurn uploaded pictures of shoes (manually taken in different shoe stores) to a simple website called Zappos. Once someone clicked on the button to buy a pair, he went to the store, purchased them, shipped them, and handled payments.
While the website provided a close-to-genuine UX, the rest was done without any technology involved.

Amazon started with books only

Amazon.com was born with a huge collection of books.
Similarly to Zappos, Amazon didn’t keep any books in stock: when someone ordered a certain book, it was manually bought straight from the distributor and shipped.
The MVP focused on books because books are relatively easy to buy, easy to ship, and were not considered to be a big investment for the consumers. A great niche to begin with.

Aardvark - a search engine operated by humans

Aardvark (now a Google company) enabled users to ask questions, mainly subjective, that were then distributed to the social graph for users for answers.
Aardvark tested their product assumptions with a series of MVPs with no algorithms.

While users believed the service had a fancy technology and sophisticated algorithms, it was in fact done by a team of people who searched for the answers manually and sent them back to the relevant users.


Looking to improve your product management resume? Here's a tip: treat your CV like a product to increase your job search conversion rate.


4. A Single-Feature MVP

This MVP follows the 80:20 rule by focusing on the most important feature of your product.
While the definition of what’s included in that single feature might be blurred, there are some nice examples out there that demonstrate the benefits of this approach:

A single page Yahoo!

Yahoo! started as a single-page website with a list of links to other sites.

Virgin Airline (without the ’s’) 

The Virgin Airlines MVP was just a single plane flying back and forth between two cities.
After testing the concept and improving the offering they expanded their fleet.

Foursquare check-in app

Foursquare was launched as a simple mobile app allowing people to check-in at certain places and win points. No special design, no special functionality, just a single-feature minimum viable product.


5. The Experimental MVP

This MVP approach leverages a special situation or a one-time event to perform an experiment.
You can call it a pilot, a POC, or a light version of the concierge MVP, but I took the courtesy to call it “experimental” because this MVP is typically performed as a one time experiment and continues to roll only if the initial results show some positive signs.

Examples:

Airbnb air mattress experiment

Airbnb started as a one-time experiment during a large conference in San Francisco back in 2017.
The founders put an air mattress in their living room, effectively turning their rented apartment into a cheap bed and breakfast alternative. They wanted to check if anyone would even be interested in such an alternative.
Surprisingly, this one-time experiment worked (3 guests were interested) so the team continued with the idea.

AppSumo started as a Reddit ad

AppSumo is a daily deals website for digitally distributed goods and online services. To validate the assumption that people will be willing to purchase apps and services through limited-time offers, Noah Kagan placed an ad on Reddit promoting a special deal for Imgur.com pro licenses. The campaign ended with 200 accounts sold and confirmed that the product had potential.
You can read more about this fascinating process in here.


Here's another great one: 10 lessons learned from asking our customers to pay.


6. The Accidental MVP

Not a formal definition, but I thought it fits some scenarios where a certain activity, such as a side project, or a last-minute pivot, leads to a successful MVP and the birth of a  product. There’s a long list of products that started as side projects, or a step of despair and ended up to be extremely successful.

The accidental MVP - 10 shades of MVP (or: how to develop a product without developing a product...)

Examples:

Unsplash was born out of despair 

This product was born as a side project when the founders of a failing product called Crew decided to give away (for free) all the extra photos they photographed while redesigning their website.
Using a Tumblr theme they launched Unsplash  - a side project that turned into a standalone product that generates a mind-blowing 11 million unique visitors/month.

Twitter was developed as an internal tool for a company called Odeo

Twitter was originally created as an internal tool called Twttr, designed to allow individuals send short statuses to a group of users via SMS.
The MVP was a hit, it was extremely addictive but caused Odeo employees to spend a lot of money on text messages so the next version included a web interface.
Although the MVP was missing some important capabilities and was only used internally - it was enough to prove that the product has a strong potential.

Slack was created as a game chat 

Slack was created as an internal chat for a game called Glitch. The game did not succeed as expected, and many of the developers have left the company. However, from the ashes of those days, grew one of the most successful products in recent years.


Recommended: 11 lessons learned while trying to become a data-driven company.


Other types of MVPs

There’s a list of additional types of MVPs that do not fall into the definition of ‘viable’ or ‘product’.
These are techniques to promote (or experiment) a product before it exists and gain valuable insights based on the way people are reacting to these experiments.
When done right - these techniques can save a lot of time and money:

7. A Landing Page

Why spending months developing a product, just to find out nobody wants it when you can spend 30 minutes designing a landing page that will provide the same answer to the question: does anyone even care about your product idea?

A landing page is one of the fastest ways to validate an idea and try to create some traction without developing anything.

You can create few landing pages, craft your value proposition, play around with the design, the copy, even the name of the product, add an early registration form and start promoting everything as if you are marketing a real product.

The Landing Page MVP - 10 shades of MVP (or: how to develop a product without developing a product...)
How do you like my Scoober idea? Think about it: a ridesharing app for scooters. Who wouldn't want that?!
All we need to do is develop the app, optimize the routes, manufacture 3-people-long scooters and we're done!
Or, we can design a landing page...

Many startups test their initial opportunity hypothesis by running some ads in the relevant channels (Facebook, Instagram, Google, Postcards, pigeons) and bringing some traffic over to the landing pages in order to convince them to click on the CTA button (dummy install or sign-up).
You can basically measure and learn a lot just by analyzing one or more landing pages. It’s a great way to start.

This technique can easily end up with a great list of early adopters willing to give your future product a try.

Examples:

Buffer started as a landing page (and then evolved into a more sophisticated landing page)

Buffer allows you to queue up your social media posts/tweets to be posted later according to an optimized schedule.
Before developing this app, the founder created a landing page that described the product and included one CTA button labeled “Plans and Pricing”. If users clicked on this button, users would be taken to a data capture form (saying: “You caught us before we’re ready”) where they could leave their email address.

The aim of this two-page MVP was to check whether people would even consider using the app.
The next thing to validate was whether people were comfortable with paying for such a product, so the landing page was enhanced to include a middle step with 3 plans (talk about Decoy Effect) so users could choose one of them. The measurements proved that people are still very interested in the product, while some were willing to pay for it.

Buffer MVP - 2 landing pages

Read more about Buffer’s MVP in here: Idea to paying customers in 7 weeks: how we did it.


8. Product Mockups

This technique can be executed in different levels of details: wireframes, mockups, demo videos or anything else that demonstrates how the product will look like.
The idea is simple:
You sell first, and build later
In B2B products it’s called a slideware (or vaporware): a software that exists only in the slides of the product manager and is used to convince customers to buy something that isn't developed yet (and will be developed once someone pays for it).
In B2C products it can be a set of static wireframes or an interactive mockup used to get some seed money.

Product Mockups and designs - 10 shades of MVP

Either way - it’s not a working software yet, but instead, a high-level UI concept that is used to present the vision of the product to investors or potential customers.
To succeed with this strategy, you need to create powerful mockups that will convince that you and your team can deliver. This is why this option might work better with existing customers that already have a solid relationship with you and your team.


9. Crowdfunding MVP 

This is a special case of “sell it before you build it”.
You launch a crowdfunding campaign on platforms such as Kickstarter or IndieGoGo and people pay for your product before it even exists.
Not only will you validate if customers want to buy your product, but you will also raise money.
The only drawback is that if you’re experiment works - you have to deliver.
There are plenty of crowdfunding success stories. The ones I can think of are LiveCode and Pebble, although I doubt if the Pebble investors were satisfied with the sudden death of the smartwatch pioneer.


10. Explainer Video

An explainer video is a short video that explains what the product does and why people should buy it.
An explainer video can be designed in different styles. One of the styles is a demo, but of course, you don’t have to use a working software to make an edited video look and feel like the real deal.

Dropbox Explainer Video generated thousands of registrations in a few days
While developing the popular service for which it is famous today, Dropbox first had to figure out whether or not customers would even be interested in their product. They first made a simple explanation video showing what it was like to use Dropbox before spending time building the application. The video generated hundreds of thousands of views and convinced the team to release the product.


A summary (and a bonus)

There are 2 additional types of MVPs that I didn't mention in this exhaustive list: MWP and MLP.
MWP - Minimum Wowable Product.
MLP - Minimum Lovable Product.
Both types are gaining momentum among startups and product groups, but I chose not to include them in the list because I believe they should be part of any type of MVP that you create.

My view is that regardless of the way to create your MVP, be it the Wizard of Oz, or a Landin page - it should be awesome. It's the freaking iPhone generation, people expect nothing less than a great experience, so regardless of the logic behind the scenes, the backend puzzle you've built, or the manual service you personally provide - make your MVP lovable and "wowable" from the very first moment. I think it sums up pretty nicely this whole post, right?


Before you leave, check out my crazy collection of 84 cognitive biases you should use to improve your product's conversion rate.


That's it for this one! Follow me on twitter @gilbouhnick, or subscribe to my newsletter to get some occasional posts directly to your inbox.


Comments

Gil Bouhnick The Mobile Spoon
Anonymous said…
Appreciate this post. Let me try it out.
Gil Bouhnick The Mobile Spoon
Anonymous said…
Outstanding post, I think blog owners should
acquire a lot from this website its really user pleasant.
So much good information on here :D.
Gil Bouhnick The Mobile Spoon
Entony said…
Very interesting especially about Food on the table. Recently I read article about similar topic: https://www.cleveroad.com/blog/mvp-creation. Here is story about Airbnb, Buffer and Twitter MVP. You can check, it's very interesting.