Steve Blank and Eric Ries helped popularize the concept of the minimum viable product (MVP) as a way to get a functional prototype into the hands of potential customers as quickly as possible. Previously, too many entrepreneurs built complex, elaborate products without incorporating customer feedback, resulting in failure.
Now, with some of the latest JavaScript frameworks, the minimum viable product is even simpler: a front-end only MVP. Here’s how it might work:
- Use AngularJS, Backbone.js, or EmberJS as the JavaScript framework
- Grab Bootstrap for the UI
- Store data in the framework’s model (e.g. Ember Data)
- Write tests (e.g. Karma)
- Profit!
Really, the key difference with traditional web-app MVPs is putting more logic and data storage on the client side (the browser). Functionally, to the end-user, it feels and operates like a fully-functional application, only long-term persistence and back-end processing isn’t present. Much like Parse provides a flexible backend with little work for mobile apps, these modern, MVC-based JavaScript frameworks make for a front-end only minimum viable product.
The next time the minimum viable product topic comes up, consider a front-end only MVP as a simpler starting point.
What else? What are some more thoughts on a front-end only minimum viable product?
David Cummings on Startups 
I love your posts David, but a front-end only MVP is total nonsense.
Let’s say it’s 2007, your name is Drew Houston and you created a front-end only file hosting service. Drew presents it to 1k potential customers. They all love it. Now you start with the backend ….18 months later you’re finished.
That’s like presenting a mockup.
If you have a technological edge, and I believe you really need to have one, you need to know how to create the backend.
F
But it all depends on the goal you are trying achieve from this prototype MVP: if the goal is check for market fit and answer the question, “will people use this and/or is it easy to understand” then you could have saved those 18 months if they hadn’t loved it.
The prototype MVP idea is great but don’t think about it strictly as a product has to be built to function. You could consider a landing page with an email form asking if people would use a product that doesn’t exist a prototype MVP if it meets your goal. Or, a landing page with a button that says “click here for this awesome new web hosting app” : if you receive 10,000 clicks and all the user gets in return is a dump page with text saying “coming soon” you might consider those missed opportunities had you had a product built. But if the product was built first, you run the risk that no one wanted to use that product and would have enjoyed knowing that before building.
Lots of use cases for a prototype MVP.
If you do front end only, then you skip the V in MVP. It is not a viable product or even a prototype, it’s a demo.
Why not just use meteor.js? Full stack covered using only css, html and js. There are packages for bootstrap and angular if you wanted to use that instead of blaze for front end templates.
That’s how we validated newstorycharity.org