MVP vs. prototype: choosing the best development approach

What is a product prototype?

A prototype is an interactive model of a future product, a draft that leaves space for detailed design and functionality. It is a demo to showcase the project’s potential before investors or a testing focus group.

The purposes of prototyping are to:

• Adjust the concept. At the prototyping stage errors in the concept can be discovered, the fundamental idea can be tweaked, expanded or contracted, the visual components and functions receive more care. Together these preparations lay the tracks for the working process to follow past technical problems that could derail the development early on. 
• Prepare a base for an MVP. With a prototype in place it becomes easier, quicker, and less taxing on the budget to draw up a user interface, layouts and visual elements. 
• Provide a detailed scope of work. If the software’s development is to be outsourced, a prototype will complement the work documentation very sensibly and ground the undertaking on the customer and the contractor side. It sets a frame and gives a reference point for the project that all parties can keep in sight. 
• Please investors. The hope, at any rate, is to make a favorable impression on the people bringing in money, perhaps even inspire the first investment round. However, a prototype is merely an interactive model of the software, not a working version. Still, it is better than nothing. 

Prototyping results in a raw, basic version of a product with many features missing, useful mostly to check if the concept can take off at all.

Varieties of prototypes

Prototypes are not going to be sold, they are in-house trial versions, so naturally they are rarely furnished with complete design and functionality. Let us consider the different varieties of prototypes and their advantages and disadvantages.

Graphical

A graphical prototype is a sketch of the application’s looks. Designers sometimes call these plain approximations wireframes, and they are useful to give some shape to the idea and a basic outline to the product. They are not going to impress investors, but the team itself will get a notion of what it is about to start making and in which direction to advance. Wireframes can also model future users’ interaction with a website and the degree of its convenience. For instance, if you plan to create a service through which painters would be able to sell their work, a design sketch will help distribute menus, galleries and so on around the page.

🤩	😰
Depending on the complexity of the project, wireframes take 10-30 hours to put together.	Does not convey the project’s full complexity.
Adjustments can be made quickly and in real time.	The team may not understand what the final product should look like from the demonstration. A graphical prototype must be accompanied by an extensive explanation.

Interactive

An interactive prototype is fully designed, polished and detailed. It may look, click and, to the user, feel completely like the final product. The only snag is that it is a dud. There is no code. The prototype can be leveraged as the interface of an MVP later on. 

🤩	😰
Gives a perspective on the concept’s viability, its potential in the course of real interaction.	Labor-intensive at ~200 hours. The screens have to be connected as well, which is laborious also.
Imitates use scenarios and routes and helps add precision to the UI/UX.	May nonetheless require turning and has no substance, i.e. code, to be sold.

Technical

Technical prototypes are a step up from interactives, upgrading an almost complete visual design with a portion of real functionality. This sort of prototype is needed when the whole project or a part of it is bound to be technically challenging, and it helps find out how feasible those elements are and what to change before advancing to MVP. If the technical aspects have been reined in and work as planned, the prototype is complete and ready for presenting to investors, and the team can move on to MVP.

🤩	😰
Gives an idea not only of the concept but of how it will work.	Takes a long time to complete, not possible to deliver on short notice.
Looks solid, functional and alluring to big-fish investors.	Requires more resources and a bigger staff.

The main difference between the types is the degree of development. The more thorough and functional versions are better suited for showing to the world, but they require time and resources to put together.

As you can see, prototypes and MVP are complimentary stages of product development, not alternatives. A startup needs both, if it wants to succeed. MVP is our next subject.

What is an MVP?

An interactive prototype conveys an idea of how the application will work to the people who get to play around with it. Putting out an MVP is the most economic and safe way to find out the degree of demand the product will encounter from actual end users. If the developers missed the mark on function or looks, they can steer the product in the right direction at this relatively early stage. 

The purpose of an MVP is to get feedback from end users and make adjustments to the technologies, the business model or any other unlikeable aspect at the testing stage.

Putting out such a trial version makes sense for startupers for several reasons:

• An MVP saves company money that it would spend on studying its target audience and harvests fresh statistics without expensive market probes.
• It creates an initial user base and starts mustering a loyal audience to support the project in its later development. 
• It reduces financial risks of development and lets the company discover mistakes in the method, approach or product before it is too late.
• It decreases the cost of development, because functions and properties not particularly embraced by the testing audience can be dropped.

The main advantage MVP have over prototypes is that they provide instant feedback from the target audience. Users get to try and test the product and give it an initial overall rating. An MVP brings only a minimum of functions, so the developer does not have to invest heavily into it, but it does gather a maximum of information about demand on the market and the target audience’s reactions. As for selecting which key features are to be included in such a trial, there are two schools of thought: User Story Mapping and MoSCoW. They differ a little in the details, but both approaches serve to highlight the features an MVP cannot do without. We have discussed these methods in another article. ⬇️

MVP development is less expensive than the whole thing, because ineffective features can be dropped, the strategy and project plan — corrected. It is a very important point for startups, because they can gauge how the market responds to the promise and whether the concept merits bringing in additional resources to develop the functions to the full.

The most popular and effective MVP types

An MVP must, of course, be a match for the scale and direction of your startup, so there are several types out there you can choose from. Here are the most common varieties and some details about them, with a few cases from our experience.

1. Landing page. A landing page is not exactly a product, but it can introduce potential clients to the startup and get them interested in buying; test the pricing model; check visitors’ spending ability and their interest overall.

2. Concierge. With this kind of hypothesis testing you do all of the operations personally to get the most feedback possible. MVP app development is not even required. You offer a service to a customer and take care of providing it; if there is demand, this process can be automated.

3. Wizard of Oz. Manual-mode like the method above, but through an interface of a website or application. This way you can quickly see if people are curious about the project and find it convenient. We suggested this kind of MVP for Cheflocal, a freelancer chef project. Before they came to us, all of the orders were processed by the staff over WhatsApp.

4. Frankenstein. Another MVP with an “imitation” of work, but with only a little manual handling; under the hood the operations are routed outside to be taken care of by other services. We chose this mode of testing for Grecha.pro, a chatroom service where restaurants’ owners would be able to link up with suppliers. Half of the business processes were sent out to be handled by the Integromat service. Purrweb’s programmers wrote up the other half of the code.

5. Single Feature. As the name says, the trial features one, most interesting and important function that works. This suffices to pique the interest of clients and investors.

Your startup may be better suited for some types than others. We discuss in greater detail here the different kinds of MVP and choosing between them. 

MVP app development in practice

Purrweb has encountered most often three types: Single Feature, Frankenstein and Wizard of Oz. They are the most effective at swabbing up users’ logic and their interaction with the product. Landing pages are limited in this regard, so they are not used as MVP so often. The types may be mixed, too, for better testing. Here are a couple of business cases.

Headcount payment system

Headcount is a global contractor payment system. The company came to us when the deadlines were already getting tight, and we had to get the service up and running in two months. The customer wanted to present the service at Y Combinator, for which the product had to be capable of something – as much as possible. We directly went for a no-prototype MVP of the Single Feature type. The feature that mattered was transfers, but this was not possible without service account creation and connecting bank accounts to those, so a single feature dragged a few others behind it. Focusing on the main functions, we were able to deliver the product on time and without exceeding the budget.

Powerbank renting service EnerGO

The concept behind EnerGO was to let users rent powerbanks for smartphone charging at one subway station and drop them off at another. It was our first Internet of Things project with communication between devices. The testing for the MVP was done using a Chinese-made device in our office, and we again chose Single Feature as a clearly workable demonstration method. When the service got up and running, we added a few more functions and a few months later uploaded a new version of the application. By now the startup has become successful, self-sufficient and developing.

Even though prototypes and MVP serve different purposes, some people still confuse them and think they must choose one or the other. Now that we know that these forms of demonstrating a product are different, let’s finalize the differences by 3 criteria: purpose, address and monetization.

Fundamental differences between MVP and prototypes

The main differences between prototypes and MVP are in:

• Purpose. Prototyping helps understand (better) what the product is all about and how it is supposed to be operated, while MVP helps estimate its competitive appeal and analyze market demand. 
• Address. Prototypes prove or explain a hypothesis to the developers themselves, QA engineers, investors, and MVPs unleash incarnated code on real end-users out there and reviewers in the media.
• Monetization. Prototypes are convincing, at most, to crowd-funded platforms and investors, but MVP has something to offer to the market, perhaps even globally. With an MVP a company has something to advertise, attract traffic to and stimulate interest in.

For your convenience we’ve summarized a brief comparison of two approaches:

	Prototype	MVP
Purpose of developing MVP and prototypes	See if the idea is practical and the product convenient in outline.	Develop a basic working product for speculative fire in real market conditions, at small cost.
Intended audience	Potential investors.	Potential users and current clients.
Reasons to make	Unknown risks, an economic justification for investors.	Risks are known, money for development and testing is in, but market response is needed.
When to make	No budget for an alpha version and the idea’s potential is difficult to estimate.	The target audience likes the concept, but more funding for further development and scaling (and advertising) is to be attracted.
Future fate	To be taken further to a prototype MVP. Design and functions may be used later on.	First version, to be tested and improved into a complete product; if the testing gives bad results, the same MVP can be reused in a restrart.
End goal	Proving the product will bring return on investment – to get some.	Selling the product, testing its chances for scaling and developing on this market.
Substitution	Not a substitute for the product.	Can be a substitute for a prototype on short notice or shoestring budget, but this is risky for the startup.

If the developer has enough resources and knows just what the market and the target audience will want, it makes sense to make a startup MVP without a prototype. We at Purrweb recommend making prototypes, though, because they give the company a chance to be flabbergasted by its product’s shortcomings rather than flabbergasting others. The damage control overhead will be lower and the MVP that results more competitive.

Hybrids: MVP and prototype together

MVP and prototypes are different stages of development, their purposes are not the same, but startups whose resources are limited sometimes combine them. Whether to make both or just one, or one that is both, is going to depend on the company’s particular situation. 

Ask Purrweb about yours, if you have doubts about prototyping or MVPing. We are a complete-lifecycle company, and we specialize in MVP development with an emphasis on end-user convenience. We can suggest the most appropriate model for you and make the plan a reality on short notice.