It started, as these things often do, with a credit card statement I didn't want to read. Eleven recurring charges. One annual renewal. The total, after I'd added it up twice in disbelief, came to $4,827 a year. Not on hardware. Not on advertising. On software. Tools I had bought at some point because somebody, somewhere, had convinced me my business needed them.
I want to be clear about what this essay is not. It is not a takedown of software-as-a-service. SaaS made my career possible. The first time I sent an automated email sequence, the first time I tracked a click-through rate, the first time I generated a PDF lead magnet without opening InDesign — those moments were small private revolutions. SaaS gave me leverage I did not deserve.
What this essay is, instead, is an accounting. A year ago I sat with that credit card statement and asked a simple question: which of these tools am I actually using? The answer surprised me. So did the experiment that followed.
The audit
I made a spreadsheet. Three columns: tool, monthly cost, last meaningful use. I defined "meaningful use" strictly — not just opening the dashboard, but actually shipping work that depended on the tool. By the end of the audit I had divided the twelve subscriptions into three groups.
The first group was the genuinely active tools. I used them weekly, they made money, cancelling them would cost more than they cost. Three tools. Not eleven. Three.
The second group was the dormant ones. I had logged in within the past quarter, but couldn't point to a single shipped project that depended on them. Five tools. $173 a month combined.
The third group was the embarrassments. Subscriptions I had forgotten existed. One I'd been paying for since 2022 without logging in once. I won't tell you which one because the marketing team is reading this and they will laugh.
The dormant tools weren't the problem. The dormant tools were the symptom.
— From the essay
The cancellation week
I gave myself one week to cancel everything in groups two and three. Nine subscriptions. Some required calls. Some required forms. One required emailing a customer success manager named Brian who tried to convince me to stay on a discounted retention plan. Brian was good at his job. I cancelled anyway.
By the end of the week I had freed up $382 a month. Almost five thousand a year. I expected to feel guilty. I felt the opposite. I felt lighter. My password manager had nine fewer entries. My email inbox had nine fewer "thanks for your payment" emails. My weekly review meeting with myself had nine fewer dashboards to glance at.
The strangest part: nothing broke. Not a single workflow stopped functioning. Not a single client noticed.
This was, in retrospect, the most important data point in the whole experiment. If I cancelled nine tools and nothing changed, then those tools were not doing what I thought they were doing. They were not enabling work. They were performing a kind of theatrical preparedness — a stage set of capability that I was paying rent on but rarely walking onto.
What stayed
Three tools survived the audit. I'll be vague about specifics because that's not the point of this essay, but their roles were:
- One that handled email — list, broadcast, and automation, all in one platform
- One that handled lead magnet creation and opt-in pages, end-to-end
- One that handled site audits and SEO checks for the consulting work
That's it. Three tools, total monthly cost of $163. The remaining nine I'd been paying for were either redundant (multiple tools for the same job), aspirational (I'd buy a tool because I planned to do something I never did), or sold to me by a marketing funnel I respected too much to refuse.
Lead Magnet Engine is one of the three tools I kept
It replaced two subscriptions ($44/mo combined) and saved me about six hours per lead magnet shipped. It's $67 one-time, which made the math obvious.
See Lead Magnet Engine →The lesson I didn't expect
I started the audit thinking it was about money. It wasn't. The money mattered — five thousand a year is not nothing — but the real lesson was about cognitive overhead.
Each tool I cancelled was a tool I no longer had to think about. No more fitting it into a workflow that didn't quite need it. No more trying to extract value from a feature set I'd outgrown. No more loginging in once a quarter and feeling vaguely guilty for not using it more. The act of removal was itself the value.
The dormant tools weren't the problem. The dormant tools were the symptom. The problem was that I had been adding to my stack without removing from it. Software, like clutter, accumulates by default. The skill I had to learn was not which tool to add — that question had a thousand answers a week. The skill was which tool to subtract, and that question had to be asked deliberately, on a schedule, against my own resistance.
The skill is not which tool to add. The skill is which tool to subtract.
— The turning point
Twelve months later
It's been a year. I've added two tools since the audit and removed one. The current count is four. I've stopped justifying purchases by their potential and started justifying them by my last 30 days of demonstrated use. If I can't point to a project that depended on a tool in the past month, I cancel.
This rule has held through one launch, three new client projects, and a brief experiment with an AI agent that I returned within a week. The rule does not generate excitement. The rule does not feel like growth. But the credit card statement, when I open it now, no longer triggers a flinch.
If you have not done this audit, I would suggest you do. Not because the money matters most — though it might — but because the act of subtraction will tell you something about your work that addition never could. Most of what you think you are doing is being done by three things. The other nine are theatre.
Most of what you think you are doing is being done by three things. The other nine are theatre.
— The take-away