How can you test the viability of the business idea you intend to bring to life? The answer is by creating a preliminary version of your software product (the so-called MVP or Minimum Viable Product). But this is only half the battle. You still need to make an effort to be sure your hypothesis has the potential as a profitable investment. And the task is more difficult than it might seem at first glance!
There are different ways to test MVPs, and our team is well versed in all of them. Moreover, we've prepared an article on the topic, in which we’re willingly sharing our experience.
We're describing how to test your minimum viable product and what kinds of MVPs exist. After reading our piece, you'll learn how to find the most effective value proposition and what to provide the user with in order to interest him in your offer. In short, you'll be able to focus on building your MVP to create software that has every chance to be in high demand and join the number of market leaders.
Why build and test an MVP(s)?
Reasons to create a minimal viable product are beyond the topic of our article, but we'd like to briefly touch upon the issue anyway. We want to be sure we're on the same page and understand each other.
An MVP is the earliest version of your idea. It has limited functionality, which includes only the most necessary, must-have features. So it's a sort of “rough sketch” of the future mobile app or website (if we talk about software; in fact, the technique is applicable to any business field).
Why start the MVP development process?
There are several reasons to start your project with a well-designed MVP:
- hypothesis-testing without extra time and money. The MVP model allows you to get feedback from users in the shortest possible time and with minimal financial investment.
- the decrease in time-to-market. The factor of a reduced time-to-market indicator belongs to the main advantages of the MVP for startups. Surely, at the initial stage, you’ll launch a rough, imperfect version of your idea, but it's better than nothing.
- functionality improvement. The MVP may also suggest a way of improving the product. After testing the model on the first users, you’ll be able to find out their opinion (and their wishes) about possible changes in software functionality (or its design). So you'll have a chance to plan future updates based on the data received.
- development cost reduction. Using the MVP strategy, you reduce the risks of financial failure. Besides, the MVP approach provides you with an opportunity to focus on key product features and build only those of them that will be in demand by the user (which means the development cost is lower either).
The process of building a successful MVP includes:
- the creation of a simplified version of the product (and making it public);
- choosing the best way of testing MVPs.
- getting user feedback.
- making the needed improvements to the project.
Of course, MVP creation involves extra work. But such efforts are justified because your final goal is to find out whether the product is worth the money invested in it. So these additional labor costs will save you from more significant expenses in the future. But, above all, you should take your time to choose a really good platform for running MVP tests.
What is the purpose of MVP testing techniques?
Tactics for testing minimum viable products are created in order to get data on:
- technical & functional product characteristics: we mean checking the effectiveness of your preliminary proposition;
- business viability in a specific market. The project may be promising, but market conditions sometimes don’t favor launching certain positions.
Money & time are too valuable resources to spend on developing a knowingly losing product, aren’t they?
Best MVP testing strategies
Now let's see how to test MVPs in practice. What techniques should you take advantage of?
Wizard of Oz
The name of the model is quite intriguing, isn't it? And it was given for a reason... Do you remember the Wizard of Oz, the hero of the famous fairy tale, who used special tricks to pretend to be some frightening characters (a giant green head, a fireball, and others)? So, in our case, we also use tricks to give the user the impression that MVP has the demanded features. In fact, it's all a facade...
At the interface level, an MVP is fully developed: it’s crucial because people have to interact with your product without any problem. The user experience should be as close as possible to the real one… the customer mustn’t suspect the truth, namely: the application or site has only the appearance of functionality! And most of the features are an illusion, their work is being performed either manually by the project staff (see the MVP concierge method), or using a set of available open-source tools (take a look at piecemeal MVPs below). Users have to be sure they're working with an existing full-fledged product until the moment the functionality is really get-at-able.
Yeah, you're a bit of a trickster but you mean well, and that's the only thing that matters.
Example
- Zappos is an online shoe store. The project has an impressive history: over 10 years, the company without funds has turned into a multi-billion dollar corporation. Its founder, Nick Swinmurn, used the Wizard of Oz strategy when starting his web resource. He posted on the website images of shoes, which he didn't really have (and he didn’t have the warehouse itself as well). If an order arrived, he went to a real store, bought the required pair of shoes, and sent them to the buyer. And only when Nick Swinmurn became convinced of the value of his idea, he updated the functionality of the online store. So, as we see, he decided to test the product, and it was a smart thing to do.
- Amazon.com, which is known to have started as an online bookstore, was also originally based on the Wizard of Oz. Its owners didn't keep books in stock, they ordered the required positions from distributors (upon receipt of the order). And it worked really well: today Amazon is more than just a successful project. It's on everyone’s lips!
Concierge MVPs
Here we see another example of a good MVP testing approach. It’s a bit like the previous item (with a key difference we'll talk about below).
The essence of the technique: you check the viability of your product by providing services to a small privileged user group (sometimes they even pay you a fee). To be precise, you, as in the earlier described case, do the work of features manually, your main task is to collect user reviews and impressions.
But, as we’ve mentioned, there is a difference with the Wizard of Oz method. The client knows: the real person is behind the service provided. You aren’t trying to pretend your site or application has already been operating.
Example
- Today, Food on the Table is a popular application. But in the past, it was primarily aimed at collecting information about users' food preferences and making recipes according to the data received (as a bonus, a list of stores was offered where a user could buy the necessary products without overpaying). However, the founders of the company weren't sure their idea would work, and they resorted to the MVP testing techniques (a Concierge model, to be more specific). They agreed to cooperate with several clients whom they personally visited several times a week: they interviewed those potential users to find out what they liked to eat, drink, and so on. Then they formed up suitable recipes, bought products, and even delivered the purchase to the сustomers. As it turned out, there were many people willing to use this kind of mobile app, so the owners of Food on the Table began to supplement their program with various features: integration with stores, online payments, recipes, and more.
- Wealthfront, financial planning, and investment service started as a Concierge model too. Wealthfront employees spoke directly with clients who needed help with money management and gave them expert advice. It was the first step to success, and gradually a simple minimum viable product grew into a fully functional mobile application.
Interview with customers
As Steve Blank says in his book ‘The Four Steps to Epiphany’, presenting a client’s problem is an important part of the process of exploring his needs. And this will also help you test the business hypothesis on real users. Of course, we’re talking about an interview with customers.
It seems too simple: no special software for MVP tests, nothing sophisticated... but the technique is very effective if you approach its implementation wisely.
A wise approach involves improvisation: no pre-written scripts are welcome! Instead, you have to:
- study the real needs of the consumer and collect as much information as possible about his problem (which he would presumably solve using your product);
- hint that your solution to his problem is almost ready;
- ask the interviewee to express his unbiased opinion about your product (its primal version). Have you taken everything into account during the development process? Have some key points been ignored?
As can be seen, the technique is much smarter than one might initially think! The main trick is to ask the right questions and draw the right conclusions from the answers received.
Landing pages
In describing different kinds of MVPs, we shouldn't ignore the landing pages either. They're very popular and effective.
A potential customer gets to this stand-alone web page after clicking on an ad or in any other way (it doesn’t really matter!). And since he, a future user, got here, he should become your client. It's your chief goal, overriding purpose. So do your best to interest him in your product.
However, the landing page is useful not only from this point of view (as a marketing opportunity to talk up the features of the site or application and acquire more subscribers). A landing page is a great MVP and perfect testing platform.
By the by, you can create several pages to experiment with the UI, online or mobile content, and more.
Example
Have you heard of Buffer? Yes, it’s a program allowing you to publish posts on social media according to your personal schedule. But did you know that Buffer had started its life as a simple landing page? It had a description of the application’s idea and a clear call to action: “Tariffs and prices”. When the user clicked on the button, he found himself on a page where it was said with a bit of humor: ‘you caught us unprepared, leave an email address, and soon you’ll receive the necessary information (something of the sort).
The goal was to check who and how much was willing to pay to get the services Buffer planned to offer. To find out answers to these questions, the Buffer owners provided users with 3 tariff plans of different price categories to choose from. Later they analyzed the testing data and revealed the strong demand. It was safe to continue working on the project! Which they did.
Obviously, among other ways to test MVPs, the LP technique is the most democratic and unobtrusive (and admit, not without a funny stroke).
Ad campaigns
You may be surprised, but ad campaigns are wonderful for market research. Platforms such as Google and Facebook would come in handy: using them, you can easily track all the indicators you’re interested in (e.g. click-through-rate and conversion metrics). As a result, you'll learn what features of your website or app are most demanded by target users.
Of course, in order to test your minimum viable product in this way, you need to have one in the first place. As you understand, there is no need to build a full-fledged application, just include the most important features in its functionality.
Videos
You may read long descriptions indefinitely, but the project benefits would remain vague anyway. So it's better to show your idea instead of using hundreds of words; then it'll all become clear. That's why a video demo is an excellent way of testing MVP.
There is no need to create a product, a simple video demonstrating the principle of working with your program would do just fine. And don't forget to ask viewers to leave their comments, in which they’re welcome to share their impressions.
Example
Dropbox is well-known to everyone, today it’s a modern and convenient workspace providing personal cloud storage. A huge number of people around the world enjoy using it since it means the opportunity to synchronize documents and files of different formats and thereby simplify the work routine. And the history of Dropbox began with a demo that attracted thousands of users in a matter of days.
However, the case deserves a more detailed story!
https://www.dropbox.com/login?cont=https%3A%2F%2Fwww.dropbox.com%2Fhome
The creators of Dropbox didn't want to waste time and money in vain and decided to test the product by exploring the possible demand beforehand. They published a concise video, which advertised the benefits of working with the Dropbox system. The result impressed them: literally in a single night the number of subscribers had reached the mark of 75,000 (starting from 5 thousand). Of course, they proceeded with their idea!
Fundraising & Crowdfunding campaigns
The idea to raise money for your app or website isn't bad, is it? In the end, you not only receive project financing but also make sure the game is worth the candle: if there are payments, then there will be users.
Does it sound too good to be true? But crowdfunding campaigns do exist, and they're aimed at fundraising (also, they help to test the viability of business).
Among the most popular crowdfunding platforms are Kickstarter and IndieGoGo. All you need to do is describe your idea with eloquence, convincingly, making sure potential customers would have a desire to ensure its implementation through financial contributions. Any means can be used: landing pages, video presentations, and much more.
As a bonus to money, you get a group of interested participants: if a person invests in your project, he'll be willing to follow its creation and share his opinion. And you'll get the opportunity of improving the product during the MVP development process.
Example
We can list endless success stories of crowdfunding campaigns. Just think of the Ouya game console and the Pebble smartwatch! Both brands managed to get financing before the start of the project. That's pretty cool, isn't it?
Pre-order
The situation is similar to the previous one, but here you don't take advantage of platforms such as Kickstarter. You just create a web page with a pre-order form (surely, it’s also worth describing the benefits of the product to captivate the user!).
The idea is really winning because the availability of orders confirms the demand. Especially if there is an advance payment. Nobody would pay without the intention of using your app (or site) in the future. Therefore, the method, too, belongs to efficient MVP testing strategies.
Example
A striking instance of demand testing through pre-orders is Oculus Rift, the VR helmet. Its creators started offering to purchase a development kit even before they began production of a virtual reality helmet itself.
Prototypes
Prototypes belong to easily testable and therefore good MVP models either. And you can choose both a digital and a paper version of the prototype (whichever is simpler to implement in your particular case).
Digital prototypes
They visualize the principle of the product in almost real conditions and come in several types:
-
low-level option (screenshots & sketches);
-
high-level option (detailed mobile or web layouts imitating the most realistic user experience).
Paper prototypes
Of course, they're easier to execute. They require no special equipment and usually consist of clippings. And sometimes it’s enough to draw a rough sketch.
Example
Tinder took advantage of the prototype approach to test the product, or more precisely, the swipe mechanics. Thus, app owners popularized this gesture.
A/B Tests
A/B Tests are especially good for checking the effectiveness and viability of software ideas. Furthermore, you have a chance to test two variants of product realization and find out which one would be in greater demand.
In other words, some users will see the app version #1 (A), while others will see version #2 (B). Then you may resort to analytics tools (Optimizely or the famous Google Analytics) to measure the key indicators of each option: frequency of use, bounce rate, conversion, etc.
Piecemeal MVPs
We briefly mentioned such a testing technique when we talked about the Wizard of Oz, remember? The essence of piecemeal MVPs is very simple: you're invited to create your product with minimal effort, simply by connecting together several ready-made tools and open-source services (those that are in the public domain). Shortly put, you demonstrate the project idea with almost no development costs.
Over time, when you realize that the product pays off, you may improve its functionality by abandoning ready-made patterns (which are often far from ideal) in favor of your own unique solutions.
Examples
- When Ryan Hoover conceived Product Hunt (today it’s the largest service with a powerful start-up base), he decided to start with a simple step and organized a special group on a popular social network. The task was to check whether the community project intended to discuss new products & ideas would be interesting and in demand. You know the result: Product Hunt successfully operates in the market, and its number of users is still growing.
- Once Groupon, an American collective discount service, was a combination of a few available solutions, including Apple Mail, WordPress, and AppleScript. Though, now that's ancient history, for Groupon became an advanced, well-designed platform.
Single-feature MVPs
Sometimes the best solution is to focus on a single feature of a minimum viable product: the one defining the entire functionality of your program. For Viber and Whatsapp, such a feature is obviously messaging.
In some cases, it's better to spend time and money to perfect 1 priority characteristic of the platform than to create a more advanced but ineffective functionality.
So it remains to choose your core feature and proceed with its development.
Examples
- Yahoo! started as a one-page site with a list of links to other web resources. And look at how powerful it has become nowadays!
- Another example of single-feature MVPs is Foursquare. Today it’s a popular location-based social network, a true city guide. The program has quite extended functionality and is therefore loved by many. However, in the past, the application focused on a simple feature and provided users with the possibility to “check in” in different places. And only when the app creators succeeded with such a kind of service, they decided it was time to improve it. And they didn't have to act at random, they were guided by the wishes of users.
Accidental MVP
Here we’re dealing with a rather unusual case when a great MVP was initially unplanned, it appeared by chance... one might say, by accident.
Examples
- Oddly enough, Twitter is the result of just such an accident. It was originally developed as an internal tool for Odeo and was called Twttr. Its purpose was to send short messages via SMS (intracorporate messages, between staff). Odeo employees took a real fancy to Twttr, which led to the idea of creating a whole online network of the same type.
- Slack was created as a chat for players of the game Glitch. Alas, the game failed, but Slack succeeded and turned into a corporate messenger.
Experimental MVP
In each of the above cases, you have to build and test an MVP(s). But here you just conduct a one-off experiment. And if the results don't satisfy you, you forget about the whole thing at all.
Example
It’s hard to believe but Airbnb was created thanks to an experiment too. It happened during a conference in San Francisco: the hotels were crowded, people had nowhere to stay... and several enterprising guys suggested that city guests rent a pair of mattresses at their house (as a place to sleep). And since the experiment was a success (meaning, the service was in demand), it was decided to develop the idea into something bigger.
Summary
Now you know how to test your minimum viable product and why you should do so in the first place. Moreover, hopefully, we managed to explain to you why the cost of developing an MVP is justified; of course, with the right approach to solving the problem.
