A concierge MVP is when you do the thing for people by hand instead of building software that does it. Wizard of Oz testing is when a person does the thing behind the scenes and the users think the software is doing it.

Both exist so you can find out whether your solution actually helps before you pay to build it.

Concierge: doing the job by hand

A concierge MVP is thinking about whatever your product will be and coming up with a way to just do the thing for people instead, to test it.

Groupon started this way. Rather than having an app where people found deals, they did a concierge thing: they talked to people about what kind of deals they would want, they went and found the deals, they brought the deals to people, and people bought them. That proved their concept and it proved their solution.

Note what got proven. Not that people liked the idea of a deals app. That people bought the deals.

What this looks like for a normal app idea

For my nanny app, the concept is that parents get to know about their child's day without being interrupted at work.

I could prototype that without an app at all. I could have nannies record what they are doing during the day, I could make a PDF from it, and I could give that to the parents. If the parents read it, and it changes their day, the solution concept holds up. If they skim it once and stop opening it, I just learned that for the price of some PDFs.

You are building exactly what they want, by hand, for a few people.

Wizard of Oz

Wizard of Oz testing is when there is some wizard off doing the thing your program would be doing, except you are having a person do it right now, and the users do not know a person is doing it.

This is the one to reach for when the experience itself is the thing being tested, particularly anything where the pitch is that software is smart enough to handle it. You can find out whether the output is useful, and what people ask for, long before you can build a system that produces it.

Be straight about the ethics: do not collect anything you would not collect in the real product, and tell people afterwards. You are testing a concept, not running a deception.

Some of you are already doing this

If your business currently runs on someone filling out a form and a person doing something with it, that is a concierge MVP. The concept is already proven. What is left to find out is whether it works in app form, which is a much narrower and cheaper question than whether anyone wants it at all.

This is also the prototype route for non-visual people

If you are not visual, and you are not planning to be the one who does the design, you do not need to make a clickable prototype. There are other ways to prototype, and these are two of them.

That matters because the standard advice sends everyone to a design tool, and for a lot of people that turns into three weeks of learning software instead of one afternoon of learning whether the idea works.

Keep reading

How Do I Validate an App Idea Without Building It?

MVP vs. MLP: What's the Difference and Which One Should You Launch?

Validation Before Building vs. Building Then Pivoting: Which Is Less Expensive?

Learn this with me

Testing a solution before it is built is class 6 of From Passion to Product, my free six-week live course, the same class where you brainstorm the solutions in the first place. Concierge and Wizard of Oz are in it, alongside proof of concept and the low and high fidelity prototypes.

You run a test on your own idea during the course, and the class covers who to test with, how to run the session, and what to watch for as well as what to listen for.

Join the waitlist for the next free cohort, or get the App Strategy Workbook and run the test this month.