8 types of MVP Experiments
When you run MVP experiments, you will be presenting something that implies that your new feature or product is real or coming soon because you need validated learning from real feedback.
The extent to which you need to fake it will determine which type of MVP experiment you need to use.
The Email MVP
For this experiment, you need an email client and an email list. By pitching users on a new product, you can see how they react in a simulated scenario.
This technique is better for smaller organizations that don’t have as much brand anxiety.
-It can come off sloppy.
-It might dent your brand if the tone and production value differs from your company’s typical email.
Pro tip: Try to pair this experiment with a landing page or other MVP technique, like Concierge.
Shadow Button MVP
Instead of building a new feature, you show a button that supposedly links people to that specific feature. The link could take them to another page saying the feature is coming soon, or it might just look broken. The number of clicks you get will signal the amount of interest.
Shadow buttons are sneaky — they generate great data but create a universally negative response from your users because they think your site is broken.
-Acknowledge and thank users. An intentional bug is way better than one they can’t fix.
-Limit the amount of users who see this.
404 and Coming Soon Page
For this experiment, you act like you’re adding a new feature but when the user navigates to the page, it either displays a 404 error message or a page that says the product is coming soon and will ask you to sign up.
In deciding between the two ask yourself: If I’m the user, would I prefer to think the page is broken or that it’s misleading?
Bigger companies can get away with 404 errors because they have a lot of other pages users can jump to (i.e., Amazon). Smaller companies can get away with ‘coming soon’ pages because the audience is aware they are new.
Think about user expectations and which option will result in the least negative outcome.
Videos are used to explain what new features will do and converts much better than text pages (unless that’s not your demographic). There are two types of videos: tutorial style and sales style. In tutorials, someone explains how to use a specific feature, even if it’s not yet real, by adding video effects that make it appear real. In sales related videos, you make a fake promo by pitching the product and explaining its benefits.
For startups: If you haven’t set the bar for quality yet, you can get away with whatever quality you can afford at the moment.
The Fake Landing Page Pitch Experiment
You create a single webpage that pitches the benefits of a new product/feature and has a call to action. You then drive traffic to the page and see what users do. This is a natural fit for running pitch experiments.
Concierge MVPs are where you manually help your users accomplish their goals as a means of validating whether or not they have a need for what you’re doing. No buildout is required and it’s a transparent process. This is basically a hybrid between customer development and MVP testing.
Can be run discreetly
No fake buttons
You get to see a customer’s internal reasoning (i.e., asking them why)
Management and time intensive
A lot of resources per customer
For this experiment, instead of building your product, you take existing out-of-the-box software, piece it together, and get the functionality you need to test a basic version.
This is a good option for adding common functionalities. One con is if you have to implement multiple functionalities as it is usually a pain to get off-the-shelf software to cooperate with each other.
Wizard of Oz MVP
Remember the movie “The Wizard of Oz” where the wizard turned out to be an old white guy hiding behind a curtain pulling levers and pretending to be a big scary green head? That’s exactly what this experiment is. From the outside, this MVP looks completely functional, but all the tasks that computers and automated systems should be doing are being completed by a human.
This experiment is the most resource intensive, but the safest for your brand image (remember, people don’t know it’s not real).