6 Ways To Validate A Product Idea
(aka. Learn about your future customers without spending a lot of money -> build something people actually want)
But an idea is nothing until it has been validated. In an amazing example of Quora doing what Quora does best, Keith Rabois shares why most startups do not succeed.
"90% of the time consumer Internet companies fail for one reason: Inability to acquire and retain a substantial number of users. In isolation the founders can articulate a reasonable value proposition, but in the real world cutting through the clutter of the 3,000 advertisements per day that the average American is bombarded by is extraordinarily difficult."
So how do we validate an idea and avoid the number-one problem facing entrepreneurs--building something that isn't compelling enough? Many of the methods of avoiding this startup pitfall are found in the product person's bible, The Lean Startup, where Eric Ries makes a science out of testing and validating ideas.
The focus of this post is to highlight some methods that I know of to validate a startup or product idea. The idea is to learn as fast as you can without using significant capital.
1.The Fake Landing Page
The most widely known method is the fake landing page. The fake landing page has some light marketing to draw in the potential customer, a value proposition statement, and a link to submit an email for product updates. Some startups take preorders of the product, but this is technically illegal and I discourage it.
A really important note: do not answer the question of how on the landing page. Do not highlight specific application features or the method of solving the stated problem. Just say you are going to solve the problem.
I recommend testing a few ideas at a time and using them to benchmark one another. Take three ideas that you have been developing and start building out a landing page for each. If you have the coding and design skills to standup a landing page quickly, go for it. If not, no worries. Unbounce, LauchRock, Lander are three excellent services that will have a hosted, beautiful landing page up in a matter of hours. The startups I have worked with in the past have used Launchrock with success, but I have heard great things about all three.
Once you have a landing page up and running, you need to start driving potential customers to the site. I recommend using Bing Ads for this. Rather than compete for attention on Facebook and Google, our valuable dollar goes way farther at Bing. But if your product is after a younger demographic, you might want to consider Google or Facebook.
Based on the number of product signups, click through rate, etc, the better ideas will begin to surface. Now it is time for your smoke and mirrors to get a bit more sophisticated with some A/B testing to allow you to pick the one idea to focus on.
Pick your two or three best ideas so far and take the landing pages to the next level. Make multiple versions of each page to further test value propositions, potential pricing, and product design. I use the tools provided by Teehan + Lax to mock up some basic mobile UI.
Rather than 100% rely on web and mobile ads to drive traffic to our site, we are able to message those that signed up for product updates in our pervious tests. This will help us measure the stickiness of the idea. I recommend using MailChimp to stay in contact with your signups (it's free under 2k subscribers).
Based on your refining round of testing, you will be able to identify that one idea to push ahead. This simple test will greatly increase your probability of success.
2.The Ebay Test
I discovered this method from Tim Ferris in The 4-Hour Work Week. This method only works for physical goods that can be sold on eBay (e.g. iPhone case, bike wheels, etc.) and ideally you have a prototype already built. If not, you can try mocking up the product using a free 3D modeling program or outsource the work.
Once you have a few high-quality product pictures, set up an eBay account and start an auction for the product. Make sure the auction includes a headline, value proposition, and a call to action. In this case, the call to action is to bid on the item. Set the auction length to 10 or so days and watch what happens. Before the last 24-hours of the auction, cancel the auction. This test will not only give you demand for the product, but helps set a price point.
Please note, I do not endorse this method.
Mark Cuban recently stated that "[k]ickstarter should be a requirement for every startup. It’s a way for you to create demand and sell the product without giving up any equity."
I couldn't agree more. Kickstarter is amazing. Personally I have backed 4 projects this year. Not sure how much I can write about this given I have never ran a kickstarter campaign, so here is a list of kickstarter best practices brought to you by Chris Kocek: 1. Keep the product design simple. Accessories to already widely used products have had the most success. 2. Keep the product description under 200 words 3. Build credibility through testimonials and associations 4. Clearly describe where the money is going 5. Spring for high-quality product photos or videos 6. Include an FAQ 7. Carefully define pledge levels 8. Give frequent updates
4.Talk to as many people as you can
I vividly remember watching the first UX/UI test of my first mobile product: "Shit, I didn't think about that . . ."
For 2-months I had been sprinting with the development team to get an MVP out for some user testing. I was so proud of the product and then I threw my baby into the dragon's den. They ripped it to shreds. Getting out and talking with people helps you get our of the weeds.
Whether they are your advisors, friends, co-works, or total rando's, talk to people about your idea assumptions. Do be aware of your audience and tailor questions as needed, but the more you listen, the more you are going to learn. Get a prototype into the hands of as many people as you can.
Even if you are in the B2B space, make a list of your highest valued potential customers, start knocking on doors, asking questions, and listening--over and over again.
Simply put, get out of the building and start talking to people.
Surveys go hand-in-hand with talking to people. If used properly, they can yield solid validation of an idea. If used incorrectly, they will only create more noise for you to deal with.
When creating a survey, I suggest that you start with a hypothesis and work backwards. What do you expect that user to answer and design your questions around these assumptions. If you are surprised by the results, you conducted a good survey. If not, you may or may not have learned something.
To drive users to my surveys I have used Amazons's Mechanical Turk, asked friends on Facebook, posted on Twitter, written a few guest blogs with links, and emailed those on my product update lists.
Remember that these are all tests. Tests are never perfect. It is important to tests your tests with your gut. No one could have predicted the success of Apple, Facebook, or Intel. You couldn't fully test the ideas. They were driven to existence by amazing entrepreneurs.
So get out there and start testing your ideas.
Big thanks to the following for their posts on idea validation.