Ideas and Validation November 7, 2019

How you validate an idea?

Darius Mann @RedJack

Hi, I just joined IH and I am from Germany, Berlin. I wanna ask you guys how do you validate ideas before you start building an App or Service for them?

  1. 9

    I'd recommend talking to people. You can learn a ton before even building anything. Talk to people that could be potential customers. Avoid bias's by not letting your pride out on the table. Don't tell them about your idea. Instead ask questions that give you real data around your idea. Like what they have done in the past. Or how they have solved the problem in the past. Know what info to listen to and what to throw out. The book The Mom Test is a great book to learn how to do this!

    1. 2

      I live and breathe by the mom test. cannot recommend this book enough!

    2. 2

      I was about recommending The Mom Test. Great book!

    3. 2

      Great book recommendation, everyone should read this. I've found that it just helps me be a better conversationalist.

  2. 7

    Hi RedJack, from what was taught to me on a product development course, there are a few ways to do that.

    One is to create a video demo and get feedback on that. For instance that's what DropBox did when they were just starting out, they had a video demo of the product with files being transferred from one device to another and someone explaining what was going on. They didn't have the tech at that time, it was all an illusion but it was enough to get feedback.

    Another method is called 'Wizard of Oz'. That's when you setup a site that looks very sophisticated, but it's actually people in the background doing the work. For instance that's how Zappos the shoe site started. It looked like they had hundreds of different shoes on the site but in reality nothing was kept in stock, if you ordered something from the site, the founders would literally run out and buy it from another store and ship it to the customer. All they had in reality was the pictures of the shoes which made them look sophisticated.

    The Concierge method also works with people doing the work, but this time the idea doesn't try to look more sophisticated than it is. For instance take a look at It's a pretty cool idea where you can "hire a boss" to kick your butt if you fall behind on a your task list. The boss is just someone who checks up that you're completing your Todo list, and they send an email every so often checking up on your progress. It sounds very simple but if you tried to automate that it would probably take a long time.

    Those are a few ways but I'm sure there's more I've missed. What's your use case?

    1. 1

      Hey Mark, very interesting.

      That are good ideas. I really like the second one "Wizard of Oz".

      I wanna make a search service for online courses. It's like a little google only for this topic. I am thinking about a little input where you can search for keywords to find the courses of you interest available in the internet.

      Actually I wanna aggregate all the data from different online course platforms, put them in my own database and optimize search on this data. What do you think about?

      1. 2

        One way you could do it, is have a landing page with a form that people can fill out with the type of course they are looking for, then you as the 'Wizard' would receive those details (you could use Google Forms for this) and manually search courses that you think might be relevant to them, and email the user back with their results.

        You'd save a lot of Dev time instead of trying to integrate with lots of different course sites and writing search algorithms etc. Plus you'd be collecting email addresses so you can get back in contact with your original customers if you want to release new features.

        1. 1

          This approach ist ingenious!

      2. 1

        "I wanna aggregate all the data from different online course platforms, put them in my own database and optimize search on this data."

        I think you won't beat Google at Google.

        You could, however, set up a simple directory with ads and/or affiliate links and build it into the go-to place for anyone looking for online classes.

        1. 1

          I get you point.

          But when I don't miss understand it does google crawl this e-learning-platforms.
          I wanna fetch data direct from they API. So I have much specific information about the coursers and can optimize much for what people are looking for. As well as develop a comparison algorithm to evaluate courses.

          What do you think?

      3. 3

        This comment was deleted a month ago.

        1. 3

          No way, are you serious?

  3. 5

    I am posting almost the same answer that I used for other post:

    1- Google Keyword Planner, check the monthly search volume for the problem you are solving, how many people are actually looking for solutions. (If there are 1k to 10k per month searches, then it's a big problem affecting many people).

    2- How many competitors are there? Are they well established (i.e 5+ years being leaders in the market). Can you compete with them? Can you offer better prices/features and be able to convert customers profitably? If not then it's not a great idea!

    Also did you check on Capterra, GetApp, TrustRadius for alternatives or not?

    Note: In Peter Thiel 0 to 1 book, he has the 10x rule if you are going against well established competitor. Build something 10x better (i.e Slack or Stripe), now I know this might not be good rule for indie SaaS and is more focused on actual startups with funding etc, but still its very relevant, otherwise why would customers pay you vs others.

    3- Check Reddit, forums, Facebook groups, or other places for people talking about this problem? If there isn't anyone talking about this problem, then maybe it doesn't exist and you are wasting your team in building a solution.

    4- How would you acquire customers? Now that your problem/idea has passed all the previous checks, you need to consider how would you acquire customers. How much would it cost? If the CAC is higher than LTV of your customer, then it's pretty bad idea for sure!

    1. 1

      This is an awesome method! I really like the points you write, thank you!

  4. 4

    I'm in the process of validating 3 business ideas so I can share with you some quick thoughts.

    Idea 1: Specialized software for law firms
    I created a landing page with a sign-up and driving some traffic to it through Google Ads. This hasn't been super helpful as it's really hard to target this specific niche simply with Google. But either way, I needed to have a website up.
    I scraped a law firm directory to get a list of law firms. Then I narrowed down that list to a specific niche and then simply went to law firm websites and got contact info for specific lawyers. This took about two weeks, relatively manual process. I then created a drip campaign to send an e-mail to those lawyers to schedule a demo (I have a clickable prototype). I offered a $25 Amazon gift card for the first 25 people to sign-up. This way I was able to schedule 27 demos. E-mailed 4k people, 27 agreed to a demo. I know it's an expensive way to validate but for me it was worth it. This helped narrow down the target market, learn about the market, remove some features, etc. Currently I'm running another e-mail marketing campaign but to a more targeted market which I was able to learn about from the first campaign. This time I'm not offering any gift cards and the responses I'm getting are infinitely better.

    Idea 2: Online legal matching system (website)
    Created a landing page which outlines the product. There's actually no product, I'm doing everything manually for now (Zappos story). I'm driving traffic to it through Google Ads, costing me about $0.60 per click. Learning a lot about the search terms that work and getting ideas on how to better advertise the website. Frankly not too much luck validating the product yet but by site visitor actions, I can tell where the journey stops and what I can improve. So far I've spend about $60 on traffic and it's money well spend I think. Plus I didn't know anything about Google Ads before this so I consider it a paid lesson.

    Idea 3: Better restaurant reservation experience
    Created a Google Form with about 13 questions and posted it on LinkedIn and Facebook. The answers are very interesting and are already helping me think about the product in a more informed way. I only have 11 answers so far but hoping to get more.

    So a couple of thoughts...
    A lot of people say "Just talk to potential customers". This is lovely advice but in reality super hard to do. I'm probably not the smartest tool in the shed but I have not been able to figure out how to actually follow this advice at scale.
    On the flip-side here, I've learned that you really only need to have conversations with 4 to 8 potential clients to learn a lot. In my first example above, after the 5th demo I really understood the problems, the solutions, and what I needed to change.

    Hope this helps. A lot of great comments in this thread already. Feel free to ping me if you'd like more detail on anything.

    1. 1

      Thank you also, very useful information!

  5. 3

    I recommend you take on pre-selling your app or SaaS idea for validation with a MEVO (Minimum Economical Viable Offer), which is basically the promise of the app or Service you’re pre-selling to the prospective customer. If you gain enough attraction during the presale, that means you're on the right track.

    1. 2

      This is also a very good approach!

  6. 3

    From what I've heard, the simplest thing you can do is to create a landing page, write some copy about your intended product, and put out a signup box. Distribute to friends, families, and online communities. If the response rate is high, you're good to go. If not, try to create an MVP where users can actually get a feel of what you're planning to do.

    You might be asking how long should I wait for? Nobody knows. Companies like Airbnb and Airtable took some time before they became who they are today. Just start building what you want, hustle hard and believe in yourself!

    1. 2

      Yes good words! thank you too.

      1. 1

        All the best, @RedJack!

    2. 2

      For what it's worth, I think this approach works much better for B2C businesses than B2B.

      1. 1

        Most definitely :)

  7. 2

    Some thoughts:

    1. You'll never get to 100% certainty. You won't know if an idea is going to work until it's in peoples' hands and they are paying you for it.

    2. The idea is to remove as much uncertainty as possible by doing successive rounds of validation.

    3. Start with a problem you have, or a problem someone else has. There are many ways to discover problems, some of which have been suggested in this thread.

    4. Continue with conversations with people who have this problem. The Mom Test (mentioned here) is a great resources for asking the right questions.

    5. Ask if people are willing to pay $x/month for it.

    6. Some people ask for purchase commitments or take deposits. Personally, I ask for purchase commitments, but I'm not sure it matters much either way.

    7. (below)

    One of the big problems I see is people finding 10 people willing to pay for something. Launching it. And winding up with 10 customers with no ability to find more. That's where I come back to the approach I've used dozens of times, building pre-launch email lists for things ranging from a conference to a book to an info product to a SaaS app.

    Setup a landing page with your USP in the headline and figure out if you can drive traffic to it in some kind of sustainable fashion, whether it's Facebook ads, mentions on podcasts or forums, Tweeting, etc. The idea is not to find an infinite source of leads, but to at least validate you can find someone beyond your first X purchase commitments.

    This last step is one a lot of people miss. It's not whether people need what you've built - it's also whether you can reach them in some kind of scalable fashion.

    If you do everything above you still won't be at 100% certainty. It's pretty much always going to be scary, and a bit of a grind to launch and then find p/m fit.

    But each step above can get you one notch closer to confidence that an idea will work. It's far from an exact science, but an interesting thought experiment to think that each of the above might get you 10% closer to confidence. So even if you have great results with all of them you might be at 60% or 70%.

    I did a talk about the process I went through validation Drip a few years back. It holds up and is worth the watch if you have the time:

    1. 1

      I really like you respond and I also have seen you video. Thanks for the support!

  8. 2

    I start by building something that solves a real problem for myself. For e.g. I am making an email designer cos I have a weekly newsletter for, and couldn't find a tool that is easy to build newsletters in the way I really wanted