Doing Things the Hard Way
Read the Full Newsletter Issue 2603

Do you ever realize you’ve been working really really hard, only to discover you’ve been doing it all the HARD way?
Yeah. Me too. And I am furious about it! Even my husband of four and a half years, Tim (shout out to my cutie-pie!), remarked that he rarely sees me angry. But this week, I was get-my-cape-because-I’m-super-angry kind of angry.
I have been up to my ears all week long, only to find that everything I did was either wrong, or much more difficult than it had to be – as evidenced by this week’s newsletter coming out on Sunday instead of two days ago on Friday, as is more usual.
Last week, I hinted at the AI project I have embarked upon. I shared with you that I was contacted and asked what AI I use to create this newsletter, and that I responded that I don’t – this is all written by me. In full transparency, as is always a priority, I do use AI daily to help me research, to answer my questions, to provide solutions, and to proofread copy that I write. This inquiry did get me thinking… I have years of newsletters, Youtube videos, workbooks, presentations, blog posts, and on and on. What if I uploaded all of that content into some really clever AI, and asked it to help me improve my business with that material? What would AI tell me to do? What if I could use this as an experiment to empower other entrepreneurs, and solopreneurs in particular, to improve their work lives, and their overall lives as a result? What if there are some really cool discoveries that could help ALLof us? AI is such an incredibly hot topic right now, and entrepreneurship in the US is currently at a historic high. There is a ton of opportunity for all of us to have greater prosperity and freedom, but how? Therein lies the question.
Not one to jump on a bandwagon, I know there are already loads of content creators giving detailed instructions on AI. I don’t see a void there for me to fill. However, mindset and the human experience are tremendous components of my work. This time, I happen to be the human failing forward on this specific journey. So back to my extraordinary (for me) anger…
My initial idea was to use AI to build better and better AI, and it worked for a while. I have been using Gemini from Google for about a year or so. I also regularly use whatever chatbots happen to be built into whatever platform or SaaS (Software as a Service) I happen to be utilizing. They clearly have their pros and cons. One of the most frustrating downsides I frequently encounter is their lack of long-term deep memory. I get annoyed that I can’t talk to Gemini this week about something we did last week because it’s no longer in Gemini’s memory. Given the volume of content with which I wanted my AI to be familiar, this was one of the first problems to solve.
Though I do not consider myself a tech-person at all, and one of the only things I miss about having a “real job” is the availability of a tech support desk to call, I am pretty determined about this project. “Hyperfocused” might be more accurate. So, Gemini and I worked at this deep memory problem experimenting with multiple systems, software integrations, new subscriptions, cancelled subscriptions, API keys, and even writing code for the last two weeks. On Thursday, ohhh that fateful Thursday that shall forever remain in my deep memory, we lost everything. Absolutely everything. Each little bit of hard-won incremental progress we’d made. Every two steps forward and one step back, shifting lanes, trying again, all of it. We got locked in a loop, which closed the gate on my work, and I had to do what was essentially a factory reset.
I was fuming. And exhausted. In this emotional state is where my darling Tim found me. He sweetly hugged me and said, “it will be okay.” My reply made him laugh out loud: “I know that! But I’m not finished being mad yet!” Tim’s laughing made me laugh and I got over my anger pretty quickly. Then, as I took a deep breath and sat down the next day to attempt to fix what Gemini and I had built and broken, I stumbled upon the previously mentioned loads of content on Youtube, Reddit forums, n8n forums, and so much more. I can now not just quickly recreate what I’d lost, but I can do it much better and faster by simply copying templates that have already been made by people exponentially more tech savvy than I am. In fact, at the gym last night as my workout buddy did fun things on her phone while we walked on the treadmill, I was watching AI tutorials on Youtube, stomping out my two miles before I went to the weight machines.
Here are my key takeaways for you:
Tech support 101 is to turn it off and turn it back on to see if that works.
Tech support 102 is to keep a backup. Keep copying and recopying so that if you “break it,” you can revert back to the last version. I know that we know this. Just don’t get in a rush or get overly excited and skip it, like I did. Please, just don't not do it.
Have someone in your life who can readily accept you at your worst. Cultivate this relationship and treat it as precious as it is, because it’s one of the most valuable investments you’ll ever make.
Laugh at yourself, even if you have to finish being mad first. You’re allowed to have emotions, even negative ones. But don’t sit in the metaphorical mud too long. Hose yourself off, have a laugh, and keep going.
Don’t be afraid to experiment. I effectively hit a factory reset button, but I’m choosing to call it failing forward. That deleted data wasn't a waste; it was the tuition I paid to learn a better way. Many studies indicate that experimentation, especially failed, is what builds success and confidence… eventually.
And last but certainly not least: Leverage other people's genius. For all that is good and holy, search YouTube first! Had I just done a little bit of research two weeks ago, I would have found people operating in their Zone of Genius who had already solved this problem. I could have saved myself this frustration and been much further along!
Or, as I've always told my kids,
"You don't have to make all of the mistakes yourself."
