Today, I'm tackling the opportunities that Gemini has listed as good business options, given that I want to work about 20 hours per week on the effort and make around $75K in total income per year. I've listed my unique skillset as "software engineer," and am asking the cold, vapid machine serving as my "friend" and my "business coach" to make suggestions with these facts in mind.
To no one's surprise, it wholeheartedly agreed with me that delving into a boring, niche business is the way to go. It's not a surprise because (a) it's a good general business plan to solve problems others have no desire to solve because they're perceived as not interesting or not financially feasible (read not gonna get me a yacht anytime soon, if ever), and (b) because an LLM will never consistently tell you something you don't want to hear. Just for the lulz, I asked Gemini for help with this query:
I'm thinking about starting an artisan cottage cheese business. Help me get started on the idea and develop a business plan.
Its response started off, thusly...
An artisan cottage cheese business is a unique and potentially rewarding venture. Here's a breakdown to help you get started on the idea and develop a business plan.
In all fairness, though, it did refuse to help me with a safe-T-poke eye mechanism to keep people awake, so I suppose the AI ethicists are doing their jobs.
Even though LLMs are probably bigger cheerleaders for any of your hairbrained schemes than your mom ever was (You go ask that sweet Giselle to the prom, I'm sure she'll say 'yes.' How can she turn down a sugar bear like you?), I was able to get some useful structured advice.
With the parameters I gave it - and revising my income requirements down to $24K per year - Gemini gave me 3 potential leads for a category of business:
- Micro-SaaS options, which included a tool for home brewers to track recipe and fermentation data.
- Digital products, like a set of 50 animated, accessible UI components for React, for a one-time fee.
- Productized micro-services like a "Landing Page in a Weekend" that allows clients up to 5 pages built using a modern framework.
But woe is me! All three of these niches are technically "dead" according to the latest clarion calls from the AI ivory tower, because vibe coding can do no wrong. Even with that dire warning planted in my head, I'll move ahead undaunted, because these aren't shabby starting ideas.
Gemini also hinted heavily that I should haunt subreddits like /plumber or /anarchist and focus on pain points those groups mention frequently. Terms like "spreadsheet" or "template" are great starters for more targeted software.
Of course, the problem with consulting an LLM for business ideas like this is that it sources its content from previous discussions. It's not generating new ideas. It's still possible that some of the regurgitated ideas it spits out are in underdeveloped markets, but just remember that nothing it's providing you is a "new" idea. The novelty will need to be a synthesis of its suggestions and your own creativity.
I decided to do some very preliminary research on the above topics, and here's what I found with no more than a straightforward Google search:
A subreddit for home brewers indicated that most of them use a combination of standard Google tools like Calendar, Keep, and Sheets to track their data. This would indicate that it's a good candidate for further exploration, because no premier organized solution exists. However, there's at least one well-established provider that seems to fit the niche well (a second glance at the research shows three). Given that there are businesses covering this niche at a price point I wouldn't beat (Gemini suggested $9.99 a month vs. an existing $30 per year from one of those providers), and the fact that I know absolutely nothing about homebrewing, that makes this a less desirable niche for me to pursue.
A search for "animated UI components" surfaces
this as the top result. It's open source and fairly comprehensive. My target audience would be software developers, so, unless I'm providing a solution that's insanely easy to use or adds novelty vs. the existing solution, this probably isn't a good fit either. Developers don't typically mind doing a little grunt work to get a free solution up and running, and the amount of effort to do so probably isn't worth the cost trade-off that would be more worthwhile for more complex integrations.
Strangely enough, landing pages - the most commoditized offering here - may be my best initial bet. Loads of high-profile companies offer drag-and-drop interfaces that can create effective landing pages. However, as with No Code solutions, the code resides with the provider. The price points fall in the range of $30-$40 per month, vs. the one-time $1000-$1500 Gemini-suggested fee for a 5-page solution. This isn't a wide discrepancy, and there's some value to having access to the code, so the owner can make changes or hire someone else in the future to make adjustments if needed.
I need to make sure the product I'm offering is comparable to existing ones - the existing sites seem to offer base analytics packages -while limiting scope creep to keep myself from going crazy. Even if this ultimately isn't a good opportunity, putting together 3-5 prototype sites in Astro.js helps me establish a stronger portfolio and helps me hone the chops needed to create templates that I may be able to market if more bespoke landing pages aren't the way to go.
Landing pages it is. At least for now.
Until next time, my human and robot friends.
Comments
Post a Comment