10
12 Comments

Where is your first stop to find problems to solve?

Hey folks, I'm looking to leverage some of my recent experience with combining more traditional machine learning techniques with LLMs to interpret data nicely.

I'm wondering where people go first to try to find problems to solve as inspiration for their product. I know mine is Reddit.

Ultimately I'm hoping to build a framework that can distill some of the important info from these sources using ML, then use LLMs to make the results understandable and pretty.

Thanks in advance for your feedback!

on September 15, 2023
  1. 2

    So here are step by step guide I use personally. The mind going to this step is DO FAST FAIL FAST

    1. Go to reddit and pick a subreddit that you are most familiar ( of course the one with good market also ) with as this will make it easier for you to communicate and provide value in the subreddit.

    2. Go trough the posts and see the one with most comment and engagement ( not jus upvotes, downvotes also good things as there is something people don't like )

    3. Go through the comment. Here you can see generally what make people tick. What they like what they don't.

    4. See where you can fit in. What problem you can solve or something you can sell.

    5. Now you should have a brief understand what business you want to make and who will buy the product. So go to creativeblogtopic.com insert the business and target customer and the tool will analyze google trend and search for you. It will then return you topic and keyword that best for blogging for your idea.

    6. From the topic and keyword write a simple blog and also a simple reddit posts. Try make it controversial so it will give a reason for people to engage.

    7. Now you will do it couple of time to test the water, and see how people are reacting to it.

    8. If you got engagement now you have an idea that somehow you know people are interested to.

    9. So go built MVP. Remember this MVP is for validating some more. Think the MVP as a glorified survey. Put if interested give email on the back of the product and do pop up asking them to rate the product.

    There you go. From there you can get idea and validate it and remember to do it fast. Don't spend so much time as you need to iterate things fast and move on top of your previous understanding of market.

    Here some more reading if you interested:

    1. Acquiring Your First Users Out of Thin Air

    2. How to build a MVP: Focus on what user want

    3. When Building MVP, Speed is the Focus

    4. Struggling for Idea? Here’s How to Extract Ideas from Reddit

    Good Luck

    1. 1

      Really like this perspective

  2. 2

    Hand crafted content at Micro SaaS Ideas - Works for 27K subs.

    1. 2

      Thanks. Looks like you're the author. How do you source these ideas?

  3. 1

    I’ve learned that you should see what problems you have yourself. If you have them, then it’s likely others do too.

    To validate if the problem is real, I’ll usually do a waitlist or super basic landing page and post to Twitter/reddit.

    If I get some interest, I’ll pursue the idea.

  4. 1

    Best problems to solve are those you have experienced first hand. Solving problems that you never came across and just read about online usually leads to failure as you cannot understand a problem that you have never faced. You will just think you know the solution to it .

  5. 1

    Hey Matt 👋 great question.

    Ultimately, there's probably many 'right' answers to this, but from a personal perspective, I've always started with me. What are my problems and pain points? How do I solve them? If I can solve them, is it optimal? Can it be made much better? And if I can't solve them, there's my path. Figure out a solution, and expand from there.

  6. 1

    Great question! I tried online but ultimately didn't find anything that I felt I knew the market well enough to build for.
    Instead, I built a product based on a problem I'd had to solve for my freelance clients. And hey presto https://loginllama.app was born!

    It's a great idea to do this because you

    1. Know the market
    2. Know the solution
    3. Know there is a need for the solution
  7. 1

    Dogfooding is the best way to find problems. Take a look at your own life, what problems do you face? Then plug in the gaps.

    The result? A business you're more passionate about.

  8. 1

    always reddit....

  9. 1

    I tink the most obvious answer would be forums like Reddit and Quora. However, looking to see what people are searching online when it comes to a certain topic can give you some guidance of where people are having issues. You can also look up a topic on youtube and see what tutorials are out there around a certain topic. Is there a problem that you can solve? Make it easier?

  10. 1

    Hi Matt! I think you're asking a great question. I've personally really struggled with coming up with ideas that I've found truly worth pursuing (at least ones that are financially viable). That's why for now I have decided to go with the agency route and create my own boutique firm. I don't recommend finding your ideas "online". If you are looking to build and properly scale a company, the next 5 to 10 years of your life will be dedicated to this task. Think about the problems that you've experienced in your life – things that "keep you awake at night". Can you use your skill set to solve any of those problems?

    I know this was vague but hope it helps!

    Parsa

    1. 1

      This comment was deleted 3 years ago