Was it worth it?

Seven lessons of mechanics. This one is the question they were for: is this spending — money and time — actually earning its place?

Most people never ask, and the two ways they answer when they do are both wrong.

The two standard errors

“I use it every day, so it must be worth it.”

Frequency measures habit. Plenty of things are used daily and produce nothing — that is what a habit is. The question is not whether you reach for it but whether the outcome was better than not reaching for it.

“It saved me hours.”

Almost always estimated, almost always generously, and almost always excluding L4’s verification and correction time. The saving is remembered; the checking is not.

Both errors share a shape: they measure your experience of the tool rather than the difference it made.

Ask the counterfactual

The only question that measures anything:

What would have happened without it?

Not “did it help” — everything helps a little. What would the document have looked like? How long would it have taken? Would you have done it at all?

That last one is worth attention, because it is the honest answer surprisingly often. Some things you now do were not previously worth the effort. That is real value, and it does not show up as time saved because the baseline was zero.

Equally, some tasks would have been fine without it, done slightly differently, in about the same time. Also real, and worth admitting.

Three questions, once a quarter

Enough to be useful, short enough to actually happen.

1. What did I stop doing manually, and would I go back? The strongest signal available. If having it removed would genuinely hurt, it is earning its place. If you would shrug, it is not.

2. What can I now do that I could not before? New capability, not faster old capability. E9’s access argument is the extreme version of this — and for most readers it will be something smaller and still real.

3. Where did it cost me? Time spent correcting, something shipped wrong, understanding you did not build because you handed the thinking over (E8·L1). Count these. A review with nothing in this column is not a review.

Do not let the subscription decide

The sunk-cost trap arrives specifically here.

Having paid for a year, people use it to justify the payment — reaching for it on tasks where it does not help, because not using it feels wasteful.

The money is gone either way. The only question is whether the next month is worth paying for, and that question is not affected by what you have already spent. E8·L8’s review-date discipline exists for exactly this: decide on a schedule, not on a feeling that arrives after a large charge.

When the honest answer is “not really”

It is a legitimate outcome, and it will be true for some readers of some tools.

Cancel it. Drop to the free tier. Keep the two tasks where it genuinely helps and stop forcing the rest.

Nothing in this curriculum has argued you should use these tools more. It has argued you should use them well, which includes using them less where less is correct — and being able to say so is the point of having measured anything.

Try it now (6 minutes)

Answer the three questions for one tool you pay for.

Be specific: name the task you stopped doing manually, name the thing you can now do, and name a cost. If you cannot fill the third column, you are not looking hard enough.

Check your understanding

1. The counterfactual question is:

Did this help? What would have happened without it? How often do I use it?

2. Why is 'I use it every day' weak evidence of value?

Daily use suggests over-reliance Frequency measures habit, not whether outcomes improved Usage statistics are unreliable

3. Having already paid for a year should affect your decision to keep using it:

Yes — you should get value from what you paid for No — the money is gone either way; only next month is the question Only if the tool has improved

Check answers

Recap

Ask what would have happened without it, not whether it helped. Review quarterly on three questions: what you stopped doing and would not go back to, what you can now do that you could not, and where it cost you — and a review with an empty third column has not been done properly. Ignore what you have already paid. And if the honest answer is that it is not earning its place, stopping is a legitimate result. That completes E11.