How do you test your ideas?

Slava Korolev
8 replies
The way I see it is: 1. Create small landing page/FB page/something similar 2. Run small ad company 3. Start again Do you do something more or different? How many testimonials is enough for you? If you run ads, how do they look like? Thanks!

Replies

Louis Grenard
@korolvs publishing a teasing video or something similar (screenshots, PoC, etc.) on Reddit or other targeted communities that is in your domain might help.
Wyatt McBain
@korolvs I always start with throwaway MVPs, literally sketches on paper and I get out there and interview people. This almost always changes direction, gives good insights, and helps me find themes within my idea. Make sure to ask 5 whys for each question! When I've gotten a bit farther along and I have a variety of ideas to test I create fake landing pages for the product with a sign up field. I typically use Airtable to capture these emails for follow up. I'll also run ads on facebook to measure CAC and click through rate. If none of the variations stand out it's time to go back to the drawing board with more interviews. But if one stands out you're typically onto something. Avoid surveys early on. Get out there and talk to random people. Do shop alongs with the people you're trying to impact.
Luu Dao
@korolvs I prefer some mockup / prototype -> landing page, then connect with your target community for qualitative comments before going with ads :D
Fabian Dahlke
@korolvs I found that you will get a lot of wrong data and answers by simply using a little landing page or doing customer interviews. For now I think there are two good ways: 1. Build a landing page that looks like the final version and like a real company. Then let people pay for your product and refund them later. 2. Really ship your first MVP - a product that people can buy and test and has minimal but VIABLE features.
Fabian Dahlke
@keeev Yes this can be a little bit shady and depends on how you want to approach this. But like you said you can modify this approach in some other ways. You could also just track the clicking on the button for example. I like your version the most though.
Kevin
@dahfab isn't number 1 a bit shady? I mean if you really let them pay for a thing that doesn't exist, even for validation. Or is it more like the approach that you get to the step where people filled out the billing information and let them know that it doesn't exist but they can enter their email address for further information?
Kevin
@dahfab phew, okay. Yeah maybe this works good if you already have a product up and running and try it with a paywall. Thanks. You can also do this on a feature page where you link your, let's say three most important features, and open an overlay or 404, signup page or whatever. Just so that you can see what gets clicked most. Although it's just quantitative data you can see which sound most interesting for your target audience.
Astha Sharma
@korolvs Focus groups helps a lot! If you have some money to spare, you can get a bunch of people that would do it.