RPG Campaign Starter - helping Dungeon Masters generate new quest ideas with AI
A few weeks ago, my brother was talking to me about the latest Dungeons and Dragons session he did with his friends. He talked about how he wanted to write his own quests, something he'd done in the past, but also how he struggles to come up with unique ideas for them.
Given the recent surge of ChatGPT-related content on IndieHackers Twitter, I considered how I could use it to generate "starter quests" for him, which could be used as a base to build off of. I could include a basic plot, rewards, enemies, NPCs, etc. - then all he has to do is flesh the rest out!
I'm a software engineer at my day job, and I thought that this could be a fun, simple project to hack on!
I'm kind of an anxious person by nature, so I immediately started asking the following questions:
- If I want to offer this as a paid product to others, do I need an LLC?
- Do I need a separate bank account for accepting payments?
- How do I handle taxes?
These are all questions that have prevented me from starting projects at all in the past; they all seem a bit daunting and it was easier to just not handle it. This time, I felt inspired and decided to just figure it out. The TL;DR of my decisions are: 1. No. 2. Probably a good idea. 3. I'll use a Merchant of Record.
After I figured the boring logistical stuff out, I got to work. I also did some minor market validation, which will likely come back to me as a mistake. I used the tools that I'm most familiar with (Rails/Postgres), worked on it late in the evenings while my newborn was sleeping, and shipped in about two weeks.
After I wrapped up the last feature, I announced the launch on Twitter (which I have practically no followers on) and posted in a few related subreddits. I managed to get 5 users in the first 24 hours, likely enticed by the free-credits-for-new-users promotion that I'm still running. Even though that's not a huge number compared to the more popular IndieHacker folks on Twitter, it feels nice to have real users even if they haven't paid me yet!
Here are some of my learnings from releasing my first SaaS:
- Tax is quite complicated, even when using tools like Stripe. I started with a Stripe Checkout integration and learned about tax remittance, which made me want to use a Merchant of Record instead. I don't want any part of that mess. I opted for Lemon Squeezy after getting rejected by Paddle for some reason that they wouldn't explain.
- I only released and announced the site after finishing all of the initial features that I wanted included. I probably would've had a better time getting users / feedback if I announced while it was still under development, offering features for free. I'll do this next time.
- Marketing is hard, especially doing it for free. I've bootstrapped this and would like to keep it going for as cheap as possible. I still have a lot to do here considering that I only have 5 users, none of which are paid.
If you've made it this far, thanks so much for reading; and if you're into tabletop RPGs like Dungeons and Dragons, Pathfinder, and Starfinder, consider checkout out https://rpg-campaigns.shop/
Thanks, all!
Hey Ryan, first off congratulations! While I'm not in the D&D space I definitely know some people who are and will send them this link. Had a look at your landing page and have some suggestions. Rather than focusing on features, really think about the benefits and the value proposition. This may help with your conversion of new users! Even include a quick walkthrough video at the start, so users can get a better understanding of what exactly your product is. And totally agree on the marketing, it is definitely time consuming and a massive learning curve!
I’m currently working on my MVP using generative AI as well, were trying to build an AI-assistant that assists early-stage & first-time founders with user-research and validating their idea. We’re currently offering some free AI tools. Would love to chat with you more and provide any assistance. I currently work as product designer as my day job. simpliux.com (:
I really appreciate your feedback!
Like I said in my post, my lack of up-front market validation may come back to bite me - I think it has. After talking with some other folks in the space, it sounds like one of my biggest challenges is dealing with the fact that there's so much free content already. People spend hours writing their own quests and release it for free. It seems to just be a really passionate community.
I don't want to use the "quit" word quite yet, but this does feel like an uphill battle with an audience that isn't accustomed to paying for stuff like this.
Interesting, Have you thought of pivoting or even reaching out to D&D communities to get some feedback? Im sure there is a problem in the space that can be solved ? or value you could provide.
Sorry i'm asking so many questions, what are your next steps ? I'm really curious and would love to help you.
Yeah, great question. I actually did get this feedback from D&D communities. A few users all gave rather thorough feedback (which was awesome, shoutout to them) explaining that this is a tough niche for paid content.
I wound up with 12 users in ~30 hours and the general consensus is that there's so much free content that it would make entering the space with a "starter" concept rather difficult.
The things that typically cost $10-$15 are very fleshed out, fully written campaigns, which this just isn't designed to do.
I've kind of settled on the idea that this won't work, and I'm fine with it.
Have you thought of pivoting, I still believe there could be value. Would you ever consider reaching to that community and ask more questions? Could you use what you have built and create something more.
What if you created generated-AI-image content? Help people invision their campaign ?
Congratulations on the release! Is it possible to 'onboard and extend' an existing campaign?
Curious about the pricing: what is the reason behind credit based pricing? What do the credits mean?
Hey, thanks for checking out the site! The credit-based system is probably something that could be worded better - each campaign generation (a few calls to ChatGPT) would end up costing me money, and is the monetizable thing in this project. The idea that I had was to use a credit/usage based system that limits the number of paid calls to ChatGPT and has people making payments on some regular cadence.
I said in my other reply above that this may be an uphill battle; there's loads of free content in this space already, so we'll see what happens!
Something I picked up from IH podcast: don't let existing competition discourage you from trying. Good luck!
Hi Ryan! This is a great idea! As a D&D master and player myself and a person in marketing I suggest to try bootcamps or accelerators that can answer all your questions regarding taxes, payments and so on. I can recommend the one I was personally involved with, if you're interested.
On the whole, I must support the idea - don't focus on features (especially on the early stage), focus on value and benefits your product can bring to your audience. Show it to them in a clear simple tone. If you need some hacks and tips here - feel free to reach out.
Very interesting using GPT for D&D!
Have you considered adding image generation to this as well? Think it could really help players build their own characters.
For img gen, Evoke has a pretty good backend