This article was hilarious, in the best way. I’m genuinely impressed that you dove into Celery without any technical background. That’s amazing.
I think your product idea is awesome.
Imo, it didn't take off mainly because most users just aren’t mentally ready to invest the time and effort needed to set up a working system with AI. It takes time and rounds of education for them to realize the true value behind it.
An anecdote, other builders have reached out asking me to try similar products and give honest feedback. One of them was quite simple, I even used it to write a Medium article that gained a lot of traction. I didn’t stick with it because it can only write generic stuff, but building a system like this would be great.
Thank you…Celery was definitely a beast to sort out! 😊 Agree that the extensive set up (creative brief + style guide + writing sample + writer persona) may be a bit daunting, especially for newer AI users. And I have found it genuinely useful in my own work, which is why I still keep it running and maintained!
Ambitious and amazing you got it working and launched in 7 months. I say congrats! Now you can a bluff from any of the engineers you delegate such tasks to. I had to do this today with our dev team and their projected timeline and cost. The more you know, the better leader you become.
If you did this from zero (or limited) technical background then I am hugely impressed. Building and deploying a multi component system that people can actually use is HARD.
I didn’t see any mention of Docker or any other kind of containerisation though; and if you want to have a much easier time developing and deploying consistently - it’s worth looking into.
Yes, I just started researching Docker because I want to set up my local environment to be EXACTLY the same as my production environment. I believe that will save me quite a bit of pain. 😆
Straight up: building an AI-powered writing studio isn't about what you want to offer, it's about what the market needs. The cold truth is, if nobody wanted it, the demand signal was missed or ignored. Pivot, don't persevere.
The math doesn't lie: 90% of startups fail due to lack of market need. It’s not about your vision, it’s about their demand. In my experience, the most successful companies are those that listen, adapt, and deliver what the market is screaming for.
Wow amazing work. So I’ve actually bootstrapped something like this using the AI writing software Novel Crafter. AI produces great and consistent content with the right instructions. Novel Crafter is specifically designed for authors writing with AI. But you can repurpose it for newsletters and blogs. Recently I wondered if people would pay for a platform like this specifically created for newsletters and blogs. I guess I have the answer now.
Now I’m curious about Novel Crafter. My tool was really focused on long-form blogs and papers…a newsletter tool might be more interesting. My main takeaway was talk to potential customers before investing a lot of time and energy into development. 😄
Just out of interest I asked perplexity (I have a "pro" subscription) to find your website. It failed. This is what it told me, and I think it's instructive:
Despite searching recent and comprehensive sources for "good bloggy" as the name of a specific tool, platform, or software, there is no widely recognized tool or service called "good bloggy" within major blogging, tech, or software communities as of August 2025.
I'm going to try it.
Well done for getting your head round the morass of different tools.
Makes sense! 😄 I didn’t optimize the website for SEO, which can take months/years with a brand-new domain. I did test marketing primarily with paid ads (Google + Meta) and then stopped when I got no traction.
You need better marketing. "Good Bloggy" the name of your product literally at the very end of your entire post. I guess go sign up for one of nicolas cole's courses like premium ghostwriting academy or https://writewithai.substack.com
Probably your app is fine, certainly your marketing isn't customer centric and doesn't explain how your product solves their problem.
I've been through the same pains, trying to build something that nobody wanted. Not once, but multiple times - because I'm a slow learner 😄.
Most recent fumble was this spring, when I built this complex dashboard that had a postgreSQL database, Grafana app for visualization, and a mix of AWS lambda functions, some Python scripts, Nginx front-end proxy all neatly dockerized and hosted on Digital Ocean. It had a lot of data from posts and notes coming through a data pipeline ingestion, and crafted complex SQL queries to pull data for fancy dashboard panels on Grafana.
I got 20 beta users to sign up, but eventually all but one cancelled their subscription promptly.
In the meantime, my simple Chrome extension for Notes scheduling is selling like hot dogs during a Red Sox game.
I'm convinced that there is some sort of impedance mismatch between what the users say and what my brain hears. It's causing me to come up with all these fantasies, while I could just write a few lines of code to add on existing product to fix the most recent problems.
Seriously! I think sometimes less is more when it comes to features…although I do enjoy dreaming up elaborate functionality! 🤣 For Notes scheduling, I have an odd use case…I have two Substack accounts (this one and a poetry/fiction newsletter) but can’t automate Notes for both at the same time because the Chrome extensions require a live login.
I have a few customers with similar use case. They've been using two different browsers - Chrome and Brave - each one logged into separate accounts and they installed my Substack Notes Scheduler on both browsers. I've not tested this myself but they reported that it works.
I love this type of story. I have more projects that flopped than projects that succeeded, so I know what you mean. I haven't gotten to test it yet, but I love how clean the landing page is. The copy is also really good.
Thank you! It was really educational…I feel like I’ll be in a much better place the next time I get an idea. Ideally, I want to figure out a bit faster if something’s not a great market fit. 😆
I love the sequence, Karen. You had a problem, you built something to address it, you decided others might like it, you found out it wasn't quite a fit, you learned a lot. History is filled with amazingly productive and successful creators that followed the same sequence.
Great capture and well done for working through it all.
Timely reading this on my end as I’ve started a similar project.
Curious if you considered an app that has multiple tools within, so that a customer could have access to not only a tool that produces a blog, but also supports with research, outlining, or other relevant functions.
Genuine question whether customer would see more value in a suite of tools vs. a singular output.
My sense is that writers might be more interested in a tool focused on research, outlining and other writing-adjacent activities that does NOT include an AI-writing option. 😆 I highly recommend conducting some light market research sooner rather than later—interview some folks, see if people will sign up to be beta testers, etc.
Thanks so much for sharing your experience. Moving into building and shipping adds in a whole new dimension beyond just the technical building. Learning this myself.
I'm thinking this sounds like a FABULOUS product. AI is not all slop, and when you are the brains behind the prompts, and the style guides, and the everything, it's more human than most people realize. Now, that said, you can't publish AI right out of the gate. But if you are an overworked writer trying to serve way more clients than you should have to, a product like Good Bloggy could be a life saver at getting the first draft out of the way and in a form that can then be edited by you, the master human. Just sayin'.
That was my initial thought! And it's why I still use it for a few specific projects. 😄 But, yeah, the writers and content marketers I interviewed felt differently.
It's just that it takes a little bit more than 2 weeks to build something that no-one else has...... I'm into my 7th week now building mine and I've still got a few more things to go yet.....
I found this fascinating. The amount of work you did and all the learning along the way. I used to want to learn a lot of those applications, but more out of curiousity and so I never got there. Kudos for finishing it. That alone is pretty noble in my book.
Thank you! Working through all the bugs actually helped the learning stick. 😄 Now I use Python all the time and have AI tools running on my agency website!
That is definitely the downside to creating. It can take an enormous amount of time only to realize in the end it doesn't yield the results you hoped for.
Seriously! Although I think you can set yourself up for success by doing basic research up front around who your product is really for and what those people want. 😄
This article was hilarious, in the best way. I’m genuinely impressed that you dove into Celery without any technical background. That’s amazing.
I think your product idea is awesome.
Imo, it didn't take off mainly because most users just aren’t mentally ready to invest the time and effort needed to set up a working system with AI. It takes time and rounds of education for them to realize the true value behind it.
An anecdote, other builders have reached out asking me to try similar products and give honest feedback. One of them was quite simple, I even used it to write a Medium article that gained a lot of traction. I didn’t stick with it because it can only write generic stuff, but building a system like this would be great.
Thank you…Celery was definitely a beast to sort out! 😊 Agree that the extensive set up (creative brief + style guide + writing sample + writer persona) may be a bit daunting, especially for newer AI users. And I have found it genuinely useful in my own work, which is why I still keep it running and maintained!
Ambitious and amazing you got it working and launched in 7 months. I say congrats! Now you can a bluff from any of the engineers you delegate such tasks to. I had to do this today with our dev team and their projected timeline and cost. The more you know, the better leader you become.
Thank you! Yes, I’m definitely in a much better position to assess development timelines. 😆
If you did this from zero (or limited) technical background then I am hugely impressed. Building and deploying a multi component system that people can actually use is HARD.
I didn’t see any mention of Docker or any other kind of containerisation though; and if you want to have a much easier time developing and deploying consistently - it’s worth looking into.
Yes, I just started researching Docker because I want to set up my local environment to be EXACTLY the same as my production environment. I believe that will save me quite a bit of pain. 😆
Straight up: building an AI-powered writing studio isn't about what you want to offer, it's about what the market needs. The cold truth is, if nobody wanted it, the demand signal was missed or ignored. Pivot, don't persevere.
The math doesn't lie: 90% of startups fail due to lack of market need. It’s not about your vision, it’s about their demand. In my experience, the most successful companies are those that listen, adapt, and deliver what the market is screaming for.
Yep! That was the point of the post—do market research before committing a lot of time and effort. 🤣
Wow amazing work. So I’ve actually bootstrapped something like this using the AI writing software Novel Crafter. AI produces great and consistent content with the right instructions. Novel Crafter is specifically designed for authors writing with AI. But you can repurpose it for newsletters and blogs. Recently I wondered if people would pay for a platform like this specifically created for newsletters and blogs. I guess I have the answer now.
Now I’m curious about Novel Crafter. My tool was really focused on long-form blogs and papers…a newsletter tool might be more interesting. My main takeaway was talk to potential customers before investing a lot of time and energy into development. 😄
Yes, agree with your takeaway too.
Here is a good article on Novelcrafter:
https://kindlepreneur.com/novelcrafter-review/
Thank you!
Just out of interest I asked perplexity (I have a "pro" subscription) to find your website. It failed. This is what it told me, and I think it's instructive:
Despite searching recent and comprehensive sources for "good bloggy" as the name of a specific tool, platform, or software, there is no widely recognized tool or service called "good bloggy" within major blogging, tech, or software communities as of August 2025.
I'm going to try it.
Well done for getting your head round the morass of different tools.
Makes sense! 😄 I didn’t optimize the website for SEO, which can take months/years with a brand-new domain. I did test marketing primarily with paid ads (Google + Meta) and then stopped when I got no traction.
You need better marketing. "Good Bloggy" the name of your product literally at the very end of your entire post. I guess go sign up for one of nicolas cole's courses like premium ghostwriting academy or https://writewithai.substack.com
Probably your app is fine, certainly your marketing isn't customer centric and doesn't explain how your product solves their problem.
Amazing story, Karen.
I've been through the same pains, trying to build something that nobody wanted. Not once, but multiple times - because I'm a slow learner 😄.
Most recent fumble was this spring, when I built this complex dashboard that had a postgreSQL database, Grafana app for visualization, and a mix of AWS lambda functions, some Python scripts, Nginx front-end proxy all neatly dockerized and hosted on Digital Ocean. It had a lot of data from posts and notes coming through a data pipeline ingestion, and crafted complex SQL queries to pull data for fancy dashboard panels on Grafana.
I got 20 beta users to sign up, but eventually all but one cancelled their subscription promptly.
In the meantime, my simple Chrome extension for Notes scheduling is selling like hot dogs during a Red Sox game.
I'm convinced that there is some sort of impedance mismatch between what the users say and what my brain hears. It's causing me to come up with all these fantasies, while I could just write a few lines of code to add on existing product to fix the most recent problems.
Another day in paradise, as they say 🤣
Seriously! I think sometimes less is more when it comes to features…although I do enjoy dreaming up elaborate functionality! 🤣 For Notes scheduling, I have an odd use case…I have two Substack accounts (this one and a poetry/fiction newsletter) but can’t automate Notes for both at the same time because the Chrome extensions require a live login.
I have a few customers with similar use case. They've been using two different browsers - Chrome and Brave - each one logged into separate accounts and they installed my Substack Notes Scheduler on both browsers. I've not tested this myself but they reported that it works.
Thanks, I’m going to give that a shot!
I love this type of story. I have more projects that flopped than projects that succeeded, so I know what you mean. I haven't gotten to test it yet, but I love how clean the landing page is. The copy is also really good.
Thank you! It was really educational…I feel like I’ll be in a much better place the next time I get an idea. Ideally, I want to figure out a bit faster if something’s not a great market fit. 😆
I love the sequence, Karen. You had a problem, you built something to address it, you decided others might like it, you found out it wasn't quite a fit, you learned a lot. History is filled with amazingly productive and successful creators that followed the same sequence.
Thank you! It was a really rewarding project, even if it didn’t ultimately succeed as a product!
Great capture and well done for working through it all.
Timely reading this on my end as I’ve started a similar project.
Curious if you considered an app that has multiple tools within, so that a customer could have access to not only a tool that produces a blog, but also supports with research, outlining, or other relevant functions.
Genuine question whether customer would see more value in a suite of tools vs. a singular output.
My sense is that writers might be more interested in a tool focused on research, outlining and other writing-adjacent activities that does NOT include an AI-writing option. 😆 I highly recommend conducting some light market research sooner rather than later—interview some folks, see if people will sign up to be beta testers, etc.
Thanks so much for sharing your experience. Moving into building and shipping adds in a whole new dimension beyond just the technical building. Learning this myself.
Kudos to you for taking it on!
Very interesting write-up. Thank you!
You're very welcome!
I'm thinking this sounds like a FABULOUS product. AI is not all slop, and when you are the brains behind the prompts, and the style guides, and the everything, it's more human than most people realize. Now, that said, you can't publish AI right out of the gate. But if you are an overworked writer trying to serve way more clients than you should have to, a product like Good Bloggy could be a life saver at getting the first draft out of the way and in a form that can then be edited by you, the master human. Just sayin'.
That was my initial thought! And it's why I still use it for a few specific projects. 😄 But, yeah, the writers and content marketers I interviewed felt differently.
I think they doth protest too much.
You know why Karen?.... because there's a million AI writers out there.... you are all missing a very important ingredient.
SAme shit here. Spend aabout two weeks to build with LLM that no one needs
Next time, I'll be talking to people first before the LLM! 😂
Can I just ask you why you sound upset?
It's just that it takes a little bit more than 2 weeks to build something that no-one else has...... I'm into my 7th week now building mine and I've still got a few more things to go yet.....
I cannot say that I am upset. Just a bit bitter that it failed. But failure was expected.
If you build something just because AI told you that this is a great thing - want to disappoint you. No one yet succeeded that way.
I have learned obviously many things on the way, and that is good.
Jonathan, do you automatically believe every word a human says to you? Just curious, nothing more.
I don't believe people. I trust them. Some do, some don't.
I found this fascinating. The amount of work you did and all the learning along the way. I used to want to learn a lot of those applications, but more out of curiousity and so I never got there. Kudos for finishing it. That alone is pretty noble in my book.
Thank you! Working through all the bugs actually helped the learning stick. 😄 Now I use Python all the time and have AI tools running on my agency website!
That is definitely the downside to creating. It can take an enormous amount of time only to realize in the end it doesn't yield the results you hoped for.
Seriously! Although I think you can set yourself up for success by doing basic research up front around who your product is really for and what those people want. 😄