[PROGRESS THREAD] Finding "product market fit"

[uhhfeef]

I was building an education tool for a specific niche, with a problem that I currently struggle with.

I've always wanted to create a progress thread since I started working on this but never got around to it.
Thanks @Andy Black for the advice on LinkedIn! : )

Progress so far:

• Built 3 MVP's so far, all to test if the idea builds interest in people and if they'd want to use it:
  • Started with an MVP that had only 1 functionality. Reached out to friends at first. I wanted to see if they were receptive to the idea
  • They were (Adobe is now implementing this feature for their own website btw).
  • Reached out to a bunch of people and tried getting them to use the MVP
  • They liked it as well and I instantly started building better versions
  • Second MVP, added a bunch of core features that I wanted to test, they had a bunch of feedback and I've been iterating ever since.
  • Third MVP, made the UI better and added a few more features for them that they were happy about. All this with GPT4 because I don't know how to code nor do I like it.

• Learned how to market it myself:
  • I used cold outreach methods through Instagram, Discord to get people to use it.
  • The crazy thing is, and I'm not sure if the people in my niche are just "nice", but i only have about 15-20% rejection/ ignored rate so far.
  • Here's how i cold DM someone:
    1. Compliment them (genuinely, fake feels very obvious!), if you cant genuinely compliment, relate with them.
    2. Show them how you can provide value to their life and what you are working on in the most concise way possible. Most people are busy / lazy and don't want to read 5 paras about what you're up to even if it provides value to them.
    3. Lastly you have 2 options:
       1. If they're someone well known in that niche, ask them for their opinion on it.
       2. If they're anyone else, ask them if they know anyone who may be interested in using it. If they want it, they'll ask, if they don't they'll point you to someone or some group
  • This method has worked wonders for me. I've reached out to well known people within my niche and app developers as well, who have been more than willing to give their insights on it.
  • Just reach out! Most people are willing to help you as long as you provide value through the conversation.
  • Make sure you don't "over-sell" and push them a lot which will leave you on read.

• Figuring out GPT4:
  • I have a feeling most people are still sleeping on this. This is THE BEST tool out there and will help you solve any problem you have
  • How? All you have to do is provide as much context as you can. I mentioned in my other post how I get GPT to answer exactly what I need in this comment. I've used GPT4 entirely to create this app and if you're wondering it wasn't as easy as "build it for me looool"
  • Don't think of it as a machine that does xyz thing, think of it as a clueless robot with unlimited information.
  • Best way I've found to use GPT is to divide and create multiple chats each with context of their own so that it doesn't end up confusing itself (which it does) or forget the context (which it also does and is very annoying), with the added benefit of having it organized.
  • That's how i was able to create multiple features with GPT and then connect it all using python

• Challenges I'm currently facing:
  • Although people have showed interest in the app I realized its just a "nice to have" and doesn't solve a need as such. And if I want product market fit, I have to solve a need.
  • Most people have other means to solve that problem and a user told me its replaceable even if other methods are annoying or awkward (ouch)
  • Haven't got a sale yet so my users don't have "skin in the game"
  • Alot of users say they want something but it just isn't important enough to waste time building and implementing it in the product.
  • Feedback I receive, although very positive, isn't enough for users to keep using it on a repetitive basis.

• Moving forward:
  • I'm tracking down teachers that may find use to this tool and get their insight on it, and maybe even use it for teaching.
  • If its genuinely not useful for anyone else (although it is for me) I will pivot, keep certain features and target a different market that may have more of a use for it.

I won't have weekly updates but I will update whenever I figure out something or have value to provide.
Hope this stuff resonates with you, and I hope you join me on this journey!

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[MJ DeMarco]

Congrats on getting out there and creating something that people want!

 

[uhhfeef]

Congrats on getting out there and creating something that people want!

Thanks MJ! Appreciate it.

 

[FreakyThomas]

A lot of value in your processes/execution.Thanks a lot for sharing!
It is true that to validate your need you need to obtain some skin in the game.
I'm very impressed that you build your app with chat gpt. Probably we totally underestimate the possible uses of this amazing tool.
Could you have build it on our own, with your own coding knowledge?

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[RightyTighty]

Excellent work! With all you've learned, it won't be long before you hit one over the fence.
Question: Have you found chatgpt4 a ton better than chatgpt? I've read comments that gpt is getting dumber and I believe it. It did some marketing copy for me a few months ago that was good stuff, but I tried it recently and the copy sounded like a circus poster for eight year olds.

 

[uhhfeef]

A lot of value in your processes/execution.Thanks a lot for sharing!
It is true that to validate your need you need to obtain some skin in the game.
I'm very impressed that you build your app with chat gpt. Probably we totally underestimate the possible uses of this amazing tool.
Could you have build it on our own, with your own coding knowledge?

Thank you! Yea probably.

But to be honest, I wasn't looking to get better at code, I wanted to build a good product and iterate as fast as I can. GPT helped me do things quickly and in an optimized way, instead of hiring someone (and potentially wasting money).

I'd like to be able to delegate this to a coder in the future because I find that naturally, I tend towards ideation for products and that is what I'm strong at.

 

[uhhfeef]

Excellent work! With all you've learned, it won't be long before you hit one over the fence.
Question: Have you found chatgpt4 a ton better than chatgpt? I've read comments that gpt is getting dumber and I believe it. It did some marketing copy for me a few months ago that was good stuff, but I tried it recently and the copy sounded like a circus poster for eight year olds.

Thank you, I really appreciate that!

Yes GPT4 in general got pretty "dumb" recently but they fixed it. (Apparently, they were using 3.5 for less "complex queries" and it would mix up and produce bad results...but not sure). It was really frustrating because I tried everything I could to get an answer from it but it just refused to understand my query.

But i did learn a lot in that period, that I condensed and wrote in my comment.

It seems to be working fine now and it gets the job done. When it doesn't all you have to do is follow those steps and build even more context until it does. Not sure about just GPT because I don't use it.

If anyone needs it, here's how I built my MVP's with GPT4. You dont need to use all, just what is necessary:

• Roleplay as [___]
• Here is my assumption
• Here is my problem
• Here is my solution to the problem
• Here is what I want to do with this solution
• Here is who I think the target audience is
• What can I do to solve so and so problem I'm facing currently (within this context)
• Before you send anything, ask me if you have any questions (personal favourite)

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[RightyTighty]

I understand that your query becomes part of the database used for the next guy's answer. For this reason I've avoided queries that include specifics on my innovations. It appears that you have done so; does this concern you?

 

[uhhfeef]

I understand that your query becomes part of the database used for the next guy's answer. For this reason I've avoided queries that include specifics on my innovations. It appears that you have done so; does this concern you?

I really dont care.

I'm using it as I need similar to how you would use google based on what you need. Do you think Google isn't keeping track? Would you stop using Google as well?

If you're concerned about ideas being stolen, don't worry you're not the only one thinking of them but you could be the first who actually follows through with it.

 

[RightyTighty]

Thanks for your response. I agree that execution usually trumps an NDA. And I’m sure that Google is looking over my shoulder, but I don’t think it is pushing my queries to others like gpt is. From what I’ve read the risk associated with your IP actually getting disseminated is substantially greater with gpt.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[journeyman]

Sounds like great execution, wishing you some paid users soon!

 

[uhhfeef]

Sounds like great execution, wishing you some paid users soon!

Thank you!

 

[RightyTighty]

I really dont care.

Man, you got me to thinking. I realized that my perspective on AI security is probably akin to the typical IP paranoia where people take their innovations to grave because of the fear someone will steal it. I’m going to start loosening the reins on most of my inventions (but not that cancer cure!). Thanks for your insights.

FYI, there is a FLF inventor’s video meeting tomorrow. Should be good stuff.

MEETUPS - Inventor's Virtual Meetup

Today I had a great call with @RightyTighty where we talked about all things inventing. We both thought it would be worthwhile to have a biweekly or monthly virtual meetup of people who like to invent products, so we can share ideas, brainstorm, network, etc. If you're interested or thinking...

www.thefastlaneforum.com

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[uhhfeef]

Man, you got me to thinking. I realized that my perspective on AI security is probably akin to the typical IP paranoia where people take their innovations to grave because of the fear someone will steal it. I’m going to start loosening the reins on most of my inventions (but not that cancer cure!). Thanks for your insights.

FYI, there is a FLF inventor’s video meeting tomorrow. Should be good stuff.

MEETUPS - Inventor's Virtual Meetup

Today I had a great call with @RightyTighty where we talked about all things inventing. We both thought it would be worthwhile to have a biweekly or monthly virtual meetup of people who like to invent products, so we can share ideas, brainstorm, network, etc. If you're interested or thinking...

www.thefastlaneforum.com

Best of luck! You'd be surprised how much it knows and understands.

I'll join when i get some free time

 

[uhhfeef]

Progress​

A few users complained about my UI so I revamped it which took a lil bit of time but got it done.

Sent it out to my users. I feel like with each iteration its getting more interest. But I need to validate that properly.

How I built a simple desktop application:​

• I have a tiny bit of experience with Python and since my app is based on image processing (currently), Python is pretty good for that.
• You can create applications within Py in many way but I chose PyQT because it has its own GUI designer that you can use to easily build a GUI for the app. You still need to code out the backend and make sure everything works and every button is functional. For more info:
  1. Qt Designer and Python: Build Your GUI Applications Faster
  2. Tech With Tim - Playlist for QT Designer + coding it
  3. Codemy.com on youtube - Focuses mostly on QT Designers GUI to build all kinds of applications
• As I mentioned above, I used the guidelines to get GPT4 to code out my breakdowns and how I wanted the app to behave.
• I created separate files each containing the different features, one file containing the gui configuration and initialization which then connected with all the other files and a main file to run the app. I could just run the GUI file but I feel its more organized this way.
• Again, this is my first time doing this so if you have any suggestions, do let me know!

Current Challenges​

• Asking the right kind of questions and getting accurate feedback
• Expanding my user base, reaching out to more people
• Iterating quickly even with lack of coding knowledge
• General excitement =/= user retention

Current direction​

• I want to change the direction from a tool that assists people with their work, to an education app that helps answer the "Why's". I feel like there's untapped potential there.
• I reached out to a bunch of users and asked them questions that targets the gap between where they want to be and where they currently are. This helps me build a "need" base which can help problem solve and build the right features.
• I've realized certain problems exist not just for beginners, but also for well known already established people within the space!
• The reason why my users don't need the app is because I haven't created a large enough skew over existing apps or it just doesn't solve the right problems.
• I have 1 more core feature that will take a lil bit of time to build. Feedback from that will be the deciding factor if I should continue with the current idea or pivot.

Some recent motivating messages​

Worked on Spider-verse :

Popular mentor in the space :

Popular artist:

Others:




As always, appreciate any insights and suggestions!

 

[buscus.stan]

Progress​

A few users complained about my UI so I revamped it which took a lil bit of time but got it done.

Sent it out to my users. I feel like with each iteration its getting more interest. But I need to validate that properly.

How I built a simple desktop application:​

• I have a tiny bit of experience with Python and since my app is based on image processing (currently), Python is pretty good for that.
• You can create applications within Py in many way but I chose PyQT because it has its own GUI designer that you can use to easily build a GUI for the app. You still need to code out the backend and make sure everything works and every button is functional. For more info:
  1. Qt Designer and Python: Build Your GUI Applications Faster
  2. Tech With Tim - Playlist for QT Designer + coding it
  3. Codemy.com on youtube - Focuses mostly on QT Designers GUI to build all kinds of applications
• As I mentioned above, I used the guidelines to get GPT4 to code out my breakdowns and how I wanted the app to behave.
• I created separate files each containing the different features, one file containing the gui configuration and initialization which then connected with all the other files and a main file to run the app. I could just run the GUI file but I feel its more organized this way.
• Again, this is my first time doing this so if you have any suggestions, do let me know!

Current Challenges​

• Asking the right kind of questions and getting accurate feedback
• Expanding my user base, reaching out to more people
• Iterating quickly even with lack of coding knowledge
• General excitement =/= user retention

Current direction​

• I want to change the direction from a tool that assists people with their work, to an education app that helps answer the "Why's". I feel like there's untapped potential there.
• I reached out to a bunch of users and asked them questions that targets the gap between where they want to be and where they currently are. This helps me build a "need" base which can help problem solve and build the right features.
• I've realized certain problems exist not just for beginners, but also for well known already established people within the space!
• The reason why my users don't need the app is because I haven't created a large enough skew over existing apps or it just doesn't solve the right problems.
• I have 1 more core feature that will take a lil bit of time to build. Feedback from that will be the deciding factor if I should continue with the current idea or pivot.

Some recent motivating messages​

Worked on Spider-verse :
View attachment 51201
Popular mentor in the space :
View attachment 51202
Popular artist:
View attachment 51203
Others:
View attachment 51192
View attachment 51194
View attachment 51196

As always, appreciate any insights and suggestions!

Great update man! Well done!
This is a fantastic example that even with a tiny bit of coding experience you can ship a product and then learn on the job. Excellent mindset!

Asking the right kind of questions and getting accurate feedback

Can't say I have great experience here, but what helped me get a bit better is Rob Fitzpatrick's "The Mom Test" book.

Expanding my user base, reaching out to more people

From what I read above, you're reaching out to people one by one. That can be slow. Can you figure out some places where your potential users get together? I'm thinking facebook groups, subreddits, discord servers, linkedin groups, wherever a community exists. You could then join those groups then try to participate in discussions and plug your product here and there.

Another idea, advertise on product hunt and similar websites. You could get early adopters, and early adopters would usually give you feedback. Then you can grandfather them in or something when you add billing.

Looking forward to seeing the next update. Keep going!

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[uhhfeef]

Great update man! Well done!
This is a fantastic example that even with a tiny bit of coding experience you can ship a product and then learn on the job. Excellent mindset!

Thanks!

Can't say I have great experience here, but what helped me get a bit better is Rob Fitzpatrick's "The Mom Test" book.

Looks like a really cool book, thanks for the suggestion!

From what I read above, you're reaching out to people one by one. That can be slow. Can you figure out some places where your potential users get together? I'm thinking facebook groups, subreddits, discord servers, linkedin groups, wherever a community exists. You could then join those groups then try to participate in discussions and plug your product here and there.

Another idea, advertise on product hunt and similar websites. You could get early adopters, and early adopters would usually give you feedback. Then you can grandfather them in or something when you add billing.

I've done both, reaching out individually and in servers and groups. The issue with the second is I need to first build rapport with the moderators so that I can "self advertise" in their servers. But I reached out to 3 server owners and all approved to upload any updates in their groups, which was super nice of them.

This has been my primary way to reach out to more people. Learned a lot of people skills this past 2 months.

The other idea you mentioned, I don't want to upload to product hunt just yet, because I need to solidify my core features and the "Why" of my app. But you're right to reach a wider audience I should probably launch it on there, and hopefully once I'm done with this new feature I'll launch and share a link with you guys.

At the moment its a very niche solution to a niche problem, and I'd like to build it further so that i can broaden the scope a bit more.

Thanks for the kind words and insights!

 

[uhhfeef]

UPDATE​

GOT MY FIRST SALE!​

Not gonna lie, a lil bit emotional guys, coz I kept pushing it off and finally got the courage to ask one of my users. But it went way better than I thought it would!

Thanks @buscus.stan for recommending me the book (The Mom Test), it truly changed my perception on good/bad questions and how to use feedback. Wish I read this book sooner.

 

[buscus.stan]

Amazing! Congratulations! You're one step closer to freedom. I'm so happy for you
Celebrate a little, and back to work, you have this

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[uhhfeef]

Amazing! Congratulations! You're one step closer to freedom. I'm so happy for you
Celebrate a little, and back to work, you have this

thank you man! i really appreciate it!!

Good luck to you as well

 

[Spenny]

GOT MY FIRST SALE!​

We love to see it. Keep going, my friend, I'll be following.

 

[uhhfeef]

We love to see it. Keep going, my friend, I'll be following.

Appreciate the kind words

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[uhhfeef]

Progress​

Built a feature to test out what kind of users I have.

If they use this often, or find that this feature is promising, I'll keep the direction as an education app. If its not an important feature, I'll shift it to being a utility app since my users already use it as one.

How I've validate my ideas so far​

I first look into already existing resources out there, find common features, look into what others find useful (reddit, facebook groups, discord, etc )

Find out where they lack and polish what they have based on the niche (which requires a lil bit of knowledge of your niche).

Decide which features are most important and which ones you can let go of.

Simplification is key. I kept simplifying until I find the best way that i can think of to use said feature. This involves letting go of alot of ideas that seem like important functionality but its more important to question why they exist.

Even then it depends on how users use it, the goal is to make sure its easy for them not for you.

Simplifying has the added benefit of allowing users to get creative with the least amount of features. (Don't know how to back this up but this is a thing)

Your users are your babies. They get confused easily, so don't make them think.

How I speak to users and get what I'm looking for even if they're lying​

This has been by far the hardest but most rewarding part which I will keep iterating on and learning (and hopefully adding more insights to this).

What I thought was validation was simply people being nice to an idea I presented because of the way I phrased my questions.

And recently, just before I knew this was a thing I was going to give up on this idea because I thought I just wasn't solving the right problems.

I wasn't asking the right questions.

"Whats your opinion on it?" Wrong, begs for fake validation

"What do you think?" Seeks approval

"Here's what it does.." Again, high chance of fake validation

"Would you like any specific feature" That's my job

"Wondering if you had any insights" Too broad

Here's what I started doing instead:

"Whats the one thing you're trying to learn and improve on" To find a need

"When was the last time you used <specific feature>" Checks if they found it useful

"What is your process like" Finds out how they already think, gives ideas on what to build and to simplify it

My product idea got validated! Probably not​

"Wow that's amazing!"

"I'll be using this for quite a while!"

"I would definitely pay for that"

"Cant stress how much i adore the app"

These are real messages I've received by the way. And it definitely is a huge motivator to push through and build.

But the same users will leave you if you haven't built a solution that solves their specific needs.

Compliments, hype, excitement, etc is not indicative of validation. Very un-intuitive but it is what it is.

Get them to validate their words by asking for either of the 3 types of commitment:

1. Money (pay for the product)
2. Time (use the product more so we can get better feedback)
3. Reputation (ask for referral in a group or to a friend)

Ask the hard questions, get out of comfort zone and find the truth about the product.

This has helped me learn how to listen and find out why users feel the way they do, how they think, what problems they have that I can solve and how I can make life easier for them.

Again, take all of this with a grain of salt because I still have a lot to learn but this has been my general insight as of late.

Books I'm reading at the moment​

1. Hooked: How To Build Habit-Forming Products
2. The Mom Test: How to talk to customers & learn if your business is a good idea when everyone is lying to you (Highly recommend)

 

[uhhfeef]

Progress​

Currently trying to simplify it even further and making the User experience as smooth as possible.

I've created a chrome extension to test as an MVP. Works much faster than my desktop app and definitely feels like an upgrade.

Some recent W's​

• Cold emailed a popular YouTube mentor in the space who's now shown interest in the app and whos insight would be invaluable to build better versions.
• A user recommended the app to his colleagues and his studio now uses the app as part of their workflow.
• 2 people reached out to me first without me having to pitch to them.

Current challenges​

• Getting the most out of user conversations
• Focusing on what I need out of each conversation while digging for ideas and possible bottlenecks
• Converting each user into a paid user
• Not a big fan of coding but I don't want to hire at the moment

Moving forward​

• Going to focus on simplifying instead of building features since I've established that my users have a need for the app and i really want my users to have the best possible experience, better than already existing services.
• Will try to reach out to more group owners to get more leads. That's the only way that has worked well for me.
• Found out my app actually takes a while for users to find value in, so I'll need to shorten the time it takes.
• Figure out what kind of business model makes sense for this kind of app

Books I'm reading at the moment​

• The Go Giver (Very wholesome, definitely recommend it)
• Inspired: How to Create Tech Products Customers Love (Too corporate, not very helpful at the moment)
• Hooked: How To Build Habit-Forming Products (Amazing insights about human psychology and why users get addicted)

Appreciate your thoughts and suggestions.

 

[uhhfeef]

Update​

Cold emailed a popular YouTube mentor in the space who's now shown interest in the app and whos insight would be invaluable to build better versions.

Showed him a demo and he got back to me.

---

---

I'm not really sure what to reply with because I don't have a website yet nor do I know how the business model for this would work.

But I'm thinking of charging him a low fee to see if hes genuinely interested and would want to help me out with it.

How would you go about this?

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[uhhfeef]

Update​

But I'm thinking of charging him a low fee to see if hes genuinely interested and would want to help me out with it.

He's in and is very much interested in testing out and helping me build this thing, which I'm super excited for.

Moving forward I'm building a chrome extension as this helps me with the below:
1. Simplifies the app even further
2. Eliminates OS compatibility issues
3. Lightweight
4. Broader audience
5. Easier installation and updates

I'm really confused as to how to go about it as documentation is pretty limited. Built a simple MVP though, thinking I should just upload this to the store for now and keep iterating as I go.

Probably may have to hire. Depends on if it's validated.

The question is not "Can this product be built?" In the modern economy, almost any product that can be imagined can be built. The more pertinent questions are "Should this product be built?" and "Can we build a sustainable business around this set of products and services?" - The Lean Startup

The goal now is to get up to date with Javascript using The Odin Project's free courses and find easy ways I could build MVP's.

 

[uhhfeef]

Finallt fixed all my issues!!

Created the extension but ran into way too many issues which just kept snowballing with each iteration.

It works amazingly now, only thing I have to do is upload this officially to the Web Store.

What I did the past few days (with a heavy focus on GPT4):

• Learned a lil bit of JS, started building the extension with the help of gpt.
• Found out what CORS is and why it exists (from my understanding, its a security measure which doesn't allow you to request data from one website into another). Since my extension involves a lil bit of scraping, this wasn't allowing me to take data from websites and display it on mine.
• To solve the issue, I went the "easy" route which was creating a python API and deploying it to a cloud (thanks @circleme for your insight and suggestions!). That way, it's going through an api and not the website.
• No idea how to do that so I looked up cloud providers and with a quick google search, saw there were already a few Flask to GCP tutorials so I chose that.
  • How to Deploy Flask app to Google Cloud Platform(GCP) | In 3 Steps
  • Google Cloud Platform Flask Python App Deployment
• Did all the prerequisites, created a Flask script and deployed it.
• After a really long 20 hours figured out what issues I was running into while deploying it to the cloud and why my extension wasn't running.
• Anyway, I don't consider myself to be a technical guy so I'm glad this phase is over.

What I aim to do in the next few days:

• I'm still on my vacation so productivity is gona suffer for a bit.
• Need to replicate the features from my desktop app into the extension.
• Need to upload the current version to web store which will take another few days. The last time I tried this (for another idea), it didn't get approved so I'll have to figure out Chromes security policies as well.
• Try to get my current users to shift to the extension, as paying members.
• 4$/month is what I'm thinking at the moment.

F*ck CORS

 

[amp0193]

My favorite kind of progress thread. Focused action, and no fluff.

Keep at it @uhhfeef and let's see some paying customers!

Anyway, I don't consider myself to be a technical guy so I'm glad this phase is over.

What do you feel your strengths are instead?

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[uhhfeef]

My favorite kind of progress thread. Focused action, and no fluff.

Keep at it @uhhfeef and let's see some paying customers!

Thank you for the kind words and I hope I do as well!

What do you feel your strengths are instead?

Without being too vague, as this is something I'm figuring out myself, is that I find my core strengths to be to simplify processes and to find different ways I could to achieve it. I just like being able to spot annoying bottlenecks and to get rid of them, increase productivity, reduce time, etc.

 

[amp0193]

I just like being able to spot annoying bottlenecks and to get rid of them, increase productivity, reduce time, etc.

This skill will serve you well, especially as you scale. It can be a force, if you also have the capability of technical + product creation (even if they are not your favorites).