Abundant Systems For Entrepreneurs

Examples of how I'm using abundant systems in my business

I’ve gotten a lot of feedback lately on my ideas. They are too abstract, people think they sound good, but they don’t quite see how these systems work.

So I’ve been on a mission for the past week to put these systems into action, and come up with some concrete examples.

It starts with identifying the people you want to work with, people who have put work out into the world that you believe is valuable. In this case, I’ve taken ideas from Jay Clouse and his Creator Science media empire, Foster, and Rob Hardy, the Manifesto King.

Identifying Quality Information

I’ve been listening to Jay’s podcast for a long time now, and I’ve gotten a ton of value out of it. I also know he’s about to have his first kid soon, and he’s not going to want to work on something new with me. There’s just not enough time or attention available there. So I decided to do something that might help him out. A few of his recent interviews have been with people who are incredibly good at YouTube. There’s Aprilynne Alter, who I’ve watched blow up her new YouTube channel. She talked about a lot of areas, but focused on scriptwriting. Then there is Paddy Galloway, who has worked as a consultant on a number of huge YouTube channels, including one with some guy named Mr Beast. There was a lot of talk around how to package videos, coming up with titles and thumbnails that worked together in tandem and created a compelling video. Finally, Jay interviewed Jake Thomas, who is known as an expert in titles, because he’s done a ton of analysis on what titles work for a YouTube video.

There are a couple of things that came into play for me here. First, I know Jay’s content. It’s something I trust, because I’ve followed him for a long time. I know his interviews end up getting high quality, actionable advice out of his guests. And I know his guests are quality, because they’ve put in the work and gotten the results.

When building AI tools, I’ve been working from the basis that it’s not about how much data you have to train a model, it’s about how well you can filter the data. Large language models (LLMs) are great at duplicating examples they’ve seen and converting information from one form into another. But there’s a reason that I don’t believe LLMs at their core won’t be most useful tools: in order to evaluate the quality of the given output, you have to have a higher level understanding of the information you are asking about. In other words, if you aren’t knowledgeable about the topic you are asking the AI about, you don’t know whether or not the answer is any good.

So in this case, I’m outsourcing the expertise to other people that I trust. And I want to take that knowledge and use it as guidelines in the AI tools I’m building. So I decided to launch a bit of an experiment:

I took the information in those three interviews, and extracted a bunch of guidelines for creating YouTube scripts, titles, and thumbnails. And I designed a generous offer: I’d give Jay 50% of whatever this tool made. I set up a pre-sale link to determine demand, because I wanted to see if anyone was interested in this tool. I want to validate whether or not anyone actually wants this tool, but I did prototype the main pieces: extracting info from the relevant podcast episodes, compiling it into prompts, and creating an evaluation process.

Then I did a little bit of testing and shared the results:

I actually like this economic model a lot, and feel like it demonstrates a very powerful concept. I’m starting out by creating something valuable, and I’m trying to do so for someone with a solid audience and community. In this case, I’m offering 50% of the money I make, because I see a lot of potential here. At worst case, it doesn’t go anywhere, people don’t buy the presale, that’s fine. I didn’t spend a lot of time on the prototype, so no harm, no foul. A slightly better case is that something does come of it. People like the idea, and I end up giving Jay a portion of the proceeds, and he can invest that in his company to make his content better or do whatever he wants. That’s a win to me, because it helps me reward him for something I’ve gotten a ton of value out of. Or, there’s the best case scenario: he decides he loves the idea and wants to work with me on a whole suite of tools like this. He tells me to keep the money as payment for future work. So I’ve managed to raise money for someone to pay me in the future without actually involving them. It’s money he sees as “free”, so it’s easier for him to “invest”, in something that has already shown promise. So I’ve taken the work I’m doing and having it lead to a working relationship with someone I admire and letting them pay me effortlessly.

It’s a different way of viewing the world. An abundant one. It showcases the power of abundant systems.

Building Tools For Yourself

The next example I wanted to share is something I’ve decided to build for myself, primarily. As part of Foster, one of the biggest benefits I’ve gotten is from our weekly writing circles. In those, we start with a somatic check-in. Then we do a short meditation. Then we write for about an hour. And finally, we have an opportunity to re-join the group and share some of our writing.

The check-ins are powerful for me. They really help me tap into what I’m feeling in the moment and transfer those feelings to the page in front of me. So obviously, for any writing I do outside of those writing circles, I don’t do them. It still doesn’t occur to me. That’s part of the fun of ADHD. My writing comes in frantic spurts, and I’m often bouncing over to write something because it popped into my head. I haven’t figured out the whole “writing at the same time and with the same procedure” thing.

So I decided to build a new writing platform that prompted me on how I’m feeling, and then once I finish that, will give me access to the blank page I need. It’s so simple, but powerful. Also, as I was thinking about it, I realized that I can also use this to track how I’m feeling over time, make all of my writing searchable, and do a lot of other cool things too.

And, since Foster is the inspiration, I also reached out to them to see if they might be interested in a partnership. Maybe they will, maybe they won’t, but I put the offer out there. I’m no longer trying to force specific future states in the things I’m doing, I’m only doing what I feel called to do and (this part is new for me) telling people about the things I’m doing and seeing if they want to be part of it.

That’s the hard part with abundant systems. The outcomes aren’t set in stone, they aren’t as predictable. You’ve got to have faith that things will work out. But they also require less work and planning upfront. You can just do things to see how the universe reacts to them. It’s scary sometimes. You might not get the outcomes you were hoping for. There’s a lot of uncertainty, and we all know how uncertainty feels.

But when you are building something for yourself and it will benefit you no matter what, anything else that comes from it ends up being a bonus. There aren’t really any bad outcomes.

If you want to join the preorder for this tool, you can either join the paid tier of this newsletter, or there’s a standalone presale option here:

Building Tools For Your Friends

I had a great chat with the Manifesto King himself, Rob Hardy. I’ve been following Rob’s work for a long time now, and when he posted about wanting more people to write manifestos, I decided to give my own a shot. I wrote the first version of the Optimist’s Manifesto, and sent it out to a bunch of people for feedback. It was pretty well received, but I also heard a lot about how I wasn’t being specific enough. I wasn’t being personal enough. It wasn’t as powerful as it could be.

I ended up rewriting it, and I’m really excited for what it’s turning into. I wanted to share that because I felt first-hand how great a manifesto can be. So when Rob was telling me about starting to write manifestos for companies and doing some B2B work as a way to grow his business, it got my mind turning. I saw a way to take what he was doing and potentially scale it up, making it easier for him to get 90% of the way there with AI so he could focus on the last 10%. That’s where mastery comes into play, so I wanted to give him a way to focus the majority of his energy there.

After our chat, I worked to analyze his writing on manifesto.quest and extract some of the principles there. I was able to create a framework based on that, scan a website, answer some questions from the site, and then generate a first draft of a manifesto.

That’s the AI-enabled future I see. Using knowledge to train an AI that can scale up the first 75-90% of your knowledge so you can focus on applying taste, experience, and mastery to the remaining pieces.

Most people don’t value the last 10-25%. They don’t have the taste or knowledge to see the differences that exist between 75% and 100%. That’s why it makes sense to scale up the 75% offering and then apply your mastery to the projects that either you value or for people that value that level of mastery appropriately.

An Experiment In Podcasting

When I shared the idea for Engineering Generosity to the other Buildspacers on Twitter, I got some feedback that I should think beyond the newsletter in terms of content. Granted, I create content everywhere, so I’m not that worried about it, but one person got me thinking about a podcast. Then that ended up leading me to a conversation with a musician friend of mine who was asking if I’d be interested in interviewing him in a couple of months when his new album is ready. I told him about the project, and he asked if I’d be interested in him writing a song for it, because he loved the idea and wanted to support it.

So I’m starting a podcast now, because I feel somewhat compelled to by the universe 😂.

But I wanted to do it a bit different (have you noticed a trend yet? I like to play with ideas and make them my own). I don’t like the typical sponsorship model for podcasts, because I don’t think they are particularly well aligned with podcasters and their audiences. But I’m interested in monetizing this podcast somehow so I can have more resources to apply to the bigger picture.

That’s why I designed an experiment. I want to test a piece of my hypothesis, that people are the world’s most undervalued asset. I’m looking for people who know somebody who is currently being overlooked. Maybe it’s someone who can’t find a job, even though they are incredible at what they do. Maybe it’s someone who is doing something amazing and just hasn’t been noticed yet. Whatever it is, I want to shine a spotlight on them. So I’m only going to accept sponsorships that ask me to highlight someone else. It’s up to you what you want to/are able to pay, but I’ll gladly spotlight people who could use it.

I’ve experienced what it’s like to be overlooked. So I wanted to create a space for those people to be featured.

The Idea Lab

I’ve mentioned this before, but it’s starting to come together and I’ll be opening up access to the initial users soon. It’s a lab I’m putting together where I’ll have the tools I use to quickly prototype things, extract and transform information, and new products I’m working on. Right now, I’m setting up the writing tool there based on what I’ve learned as part of Foster. There are a couple of other things there, and the next piece to add is probably a webscraper you can configure to learn something from a website. It’s a collection of tools I use to build abundant systems, that way, you can play with the ideas I’m sharing here, riff on them, and make them your own. Then, if you figure out something that works and you want some help, we can work on scaling it up a bit if that’s what you want. Or you can keep it small. That’s the beauty of abundant systems, they don’t dictate the needed scale. By joining the premium membership here, you’ll get bonus credits to use on the tools. And free use of tools like the writing one above that don’t cost much to use.

It’s a place where things aren’t polished and marketed, they simply exist in their state. They will be useful. They will be the things I’m using all the time to play with ideas and see where I can go with them. Once they prove themselves to be useful enough, and with high enough demand, I’ll consider productizing them. Or I’ll help members of the lab do so.

It’s where we can accelerate everything we are doing in the world.

One Last Thing

The main thing I’ve been working on lately is a pitch deck analysis tool for founders and a similar tool for VC firms. The founder-focused tool is going live on Product Hunt this week, and I’d love your support. We are officially launching on Product Hunt on Thursday. Olu and I have been working hard to package up his knowledge and experience in a way founders can get a ton of use out of. It’s been awesome to build and the feedback we’ve gotten so far is amazing. You can find more info here:

Beneath the paywall are screenshots of the lab as a WIP and how I’m using the tools to create/repurpose content. Nothing major so far, just wanted to give subscribers a peek at what’s coming soon.

First of all, I love you. 🥰 Thank you for being a supporter of Engineering Generosity.

Let’s start with a couple of screenshots of the tool based on Foster’s writing circles.

Current state of the tool: I’ve got it to the point where it stores documents both as a fulltext document in Mongo and as vector embeddings in a Chroma instance I was able to deploy. That enabled me to get search up and running, so I’m able to do semantic search across all my documents really easily. I’ve also started the process of integrating Google Doc access so I can pull all of my existing documents in. I ran into some issues processing all of them, so that will get knocked back a little bit. Will probably deploy without that piece initially.

Next, I wanted to share the Youtube Video to blog post tool. I’ve been using this more and more to generate blogs based on the videos I’ve been working on.

Here’s the video I used:

And here’s the published blog post:

Reflections on the tool: it’s not too bad right now. That’s a pretty long video and I tried to get it to key in on a very specific piece of the transcript with my prompt. It didn’t do a great job of limiting the context to that piece though. So while it captured it in essence, I don’t think it featured the content I wanted while excluding pieces that weren’t as relevant. So I need to do some better prompting around that (or maybe try switching to Claude 3.5 Sonnet… 🤔) I might actually make it possible to select different models because that would let me compare outputs more easily, or even run things side by side. I’ll have to ponder that and see which ends up with better results.