For two years I paid for a tool I used twice. This is the embarrassing math, and it's not even the worst one.
I finally did the thing I'd been avoiding: I opened every AI subscription I was paying for, lined it up against the time it actually saved me, and forced myself to look at the number. Some tools were quietly carrying my business. Some were just monthly guilt with a logo.
Here's the honest audit, the kind nobody posts because it makes you look a little foolish.
When I measured my AI subscriptions against real time saved, the ROI was wildly uneven. A small core of tools paid for themselves many times over by saving hours every week. The rest were "aspirational" subscriptions — things I bought imagining a version of me who'd use them, and never did. The lesson: ROI is about usage, not features.
What the audit revealed:
I wanted real numbers, not vibes. So I used one brutally simple formula.
For each tool: how many hours does it save me per month, and what's my hour worth? If a tool saves me four hours a month and my time is worth a meaningful amount, it has to cost less than those four hours to be worth keeping. Obvious. And yet I'd never once done the math.
I put every subscription through it. The results were not flattering.
If you can't name the hours a tool saves you, it isn't saving you any.
That sentence cost me several subscriptions on its own.
A few tools came out absurdly ahead, and they had one thing in common — they were attached to something I do constantly.
The biggest winner was my writing-and-content stack. I produce content all the time, so anything that shaves time off every single piece compounds fast. Same with the automation handling my repetitive admin — small per-task savings, multiplied across hundreds of tasks a month, add up to real hours.
The pattern is clear. The best ROI comes from tools bolted to your highest-frequency work. A tool that saves ten minutes on something you do daily destroys a tool that saves two hours on something you do twice a year. It's the practical edge of the argument I make in the honest truth about which AI tools actually earn their keep: value tracks usage, not feature lists.
The math here is genuinely counterintuitive, so it's worth sitting with. Ten minutes saved, twenty times a month, is over three hours. Two hours saved, twice a year, is four hours — annually. The daily tool wins by a landslide and it's not close. Yet when we evaluate a subscription, we instinctively get excited about the big one-time time-save and shrug at the small repeated one. We're optimizing for the wrong size of number. Frequency, not magnitude, is where the real ROI lives, and almost nobody prices their tools that way.
Now the painful part.
The worst-ROI tools weren't expensive. They were aspirational. I'd bought them imagining a disciplined future version of me — the one who'd do deep analytics every morning, or use that fancy research tool weekly. That person never showed up.
So I was paying, month after month, for a fantasy of who I wanted to be. The tool was fine. The problem was that the behavior it required was never going to happen — which is really the same point I make in why AI won't make you productive on its own: a tool can't supply a habit you haven't built. Behavioral research summarized by the American Psychological Association says as much — sustained behavior change comes from cues and friction, not from owning a better instrument.
Here's a rough shape of what I found:
| Tool type | Used how often | Verdict |
|---|---|---|
| Daily content + writing | Constantly | Keep, no question |
| Admin automation | Constantly | Keep |
| "Pro analytics" suite | Almost never | Cancelled |
| The one I used twice | Twice. In two years. | Deleted with shame |
The "used twice" one had auto-renewed quietly the whole time. I never even noticed the charge. That's the real trap — not the cost, but the invisibility.
Here's what surprised me most. Price had almost no relationship to ROI.
My single most expensive tool was also my most valuable, because I used it constantly. My cheapest tools included some of the biggest wastes, because I never touched them but never bothered to cancel either.
We instinctively audit the expensive subscriptions and ignore the cheap ones. That's backwards. A cheap tool you don't use is pure waste. An expensive tool you live in might be the best deal you have. Usage is the only variable that matters, and I'd been looking at the wrong column the whole time.
I keep coming back to why I held onto tools I never used, because the money waste is the small part. The real cost was psychological, and I think a lot of people share it.
Each aspirational subscription was a tiny monthly promise to myself that I'd become a more disciplined, more sophisticated version of me. The analytics suite said "you're the kind of operator who studies their data every morning." The fancy research tool said "you do deep work weekly." Cancelling them felt like admitting I wasn't that person — so I kept paying, month after month, to preserve a flattering self-image.
That's a ridiculous reason to spend money, and naming it out loud was what finally let me cut them. You are not the version of yourself your unused subscriptions imagine. You're the version that actually shows up, and that version has a much shorter tool list.
The freeing part is what happened after. Cancelling the aspirational tools didn't make me less capable. It made me honest about how I actually work, and that honesty redirected money and attention toward the handful of tools that genuinely fit my real behavior. I stopped paying rent on fantasies and started investing in reality. My stack got smaller, cheaper, and far more useful — all at once.
After the audit I made one rule, and it's saved me real money since:
It's not about being cheap. It's about refusing to pay monthly rent on a fantasy.
People always want the list, so here's roughly where I landed after the cull — not specific brands, but the categories that survived and the reason each one earned its place.
And here's what didn't make it: every tool that required me to build a new habit I'd never actually built. The "if I were more disciplined" tools. They all went, and I haven't missed a single one. Not one. That absence of regret is the clearest signal there is that they were never working.
The cull saved me real money, but the bigger win was clarity. My subscription list is now a fairly honest map of how I actually work. There's nothing on it that I'm paying for out of guilt, optimism, or the hope of becoming someone else. Everything left is load-bearing. And weirdly, that makes me more willing to pay for a great tool when I find one — because I trust that my stack reflects reality, not aspiration.
If your subscriptions have been quietly auto-renewing, try giving each one a job and a rough number of hours this week — the ones that can't answer aren't tools, they're guilt with a billing cycle.
Q: How do you estimate hours saved without obsessing? Roughly is fine. You don't need a stopwatch. "This saves me about an afternoon a month" is enough to make the decision. Precision isn't the point — honesty is.
Q: What's the most common waste? Aspirational tools — things bought for the person you wish you were, used by the person you actually are almost never. And quiet auto-renewals you've stopped noticing.
Q: Should I just cancel everything and re-add what I miss? Honestly, it's a great forcing function. The tools you genuinely miss within a week are your real stack. The ones you forget existed were never earning their keep.
Q: Does this mean AI subscriptions aren't worth it? The opposite. The good ones are wildly worth it — 10x and up. The audit isn't anti-AI. It's about finding which AI is actually working for you and firing the rest.
I added up every AI subscription against the time it saved, and the truth was uncomfortable: a few tools were quietly running my business, and the rest were quietly running off with my money.
A tool is only worth what you actually use, not what it could theoretically do.
So here's your homework, if you want it. Open your subscriptions tonight and give each one a job and a number of hours. The ones that can't answer aren't tools. They're guilt with a billing cycle.
I bought the dream of money while you sleep and chased it for years. Here's what passive income actually costs — and what I do now instead.
I tried every budgeting app and spreadsheet and quit them all within a month. Here's why budgets fail and the boring system that finally stu…
I spent years saving the hardest task for when I 'felt ready.' Doing it first instead quietly fixed my focus, my dread, and my output.
Comments
Sign in to join the conversation
No comments yet. Be the first to share your thoughts!