I kept a notebook on the windowsill above the sink for six weeks last year, and every time I started making something instead of buying it, I wrote down the time. Start and stop. Not an app, not a spreadsheet, just a pencil on a string because the pencil was already there. By the end of the six weeks the notebook had disappointed me thoroughly, which is the most useful thing a notebook has ever done for me.
Why I bothered
I had been telling myself a story for a couple of years. The story was that all this making saved us money, and that the making was therefore a kind of work, and that the work was worth doing. Nobody was auditing the story. Certainly not me.
What started the notebook was a Saturday in February when I spent most of a day on a project, looked at the result on the counter, and could not honestly say whether I was ahead or behind. Not roughly. Not even directionally. I had no idea. That bothered me more than a bad answer would have.
The three numbers I could not get around
To know whether an hour of making is worth it, you need three things, and I had been carefully avoiding two of them.
The first is what the bought version actually costs you. Not the fancy version you would buy if you were being indulgent. The one you would genuinely put in the cart on a normal Tuesday. I had been comparing my homemade goods to the premium shelf, which flatters the homemade version enormously and is basically cheating.
The second is what the homemade version costs in materials, counting the ingredients you bought in bulk and did not finish, counting the ones that went off, counting the energy to run an oven or a stove for a long time.
The third is the real time, and this is the number people fudge the hardest. Not the hands on minutes. All of it.
Where the hour actually disappears
The making itself is rarely the bulk of it. Here is what the notebook caught that I had never been counting.
Setup and cleanup, which for anything involving a large pot or a mixer or flour in the air is not trivial. Sourcing, meaning the extra stop, the special order, the trip because I was short one thing. Storage, meaning washing the jars, finding the lids, labeling, and rearranging a shelf to make it fit. The learning curve, which is real time spent producing bad results before you produce good ones, and which nobody amortizes into the first year even though they should. And the failure rate, meaning the batches that get thrown out, which cost you the full time plus the full materials plus the trip back to buy the thing anyway.
Add those up and a project that felt like forty minutes was reliably closer to double that. Every time. Not once did the notebook say I had overestimated.
The list, sorted honestly
When I divided what I saved by the hours I really spent, the results split into three groups, and the middle group is the one that changed my behavior.
A small number of things paid decently. Bread, because I make it constantly and the skill has stopped costing me anything. Stock from bones and vegetable ends, which is close to free time because it happens while I am doing something else and the input would otherwise be trash. Mending, which is the sleeper winner of this entire notebook and which I now think is the single most underrated skill on the list.
A larger number came out somewhere between poor and insulting. Candles were the worst offender by a distance, once I counted the wax, the wicks, the containers, the melting, the cleanup, and the ones that tunneled and got thrown out. Small batch yogurt was not worth the pot washing. Anything decorative was not worth anything at all. Several other projects landed in the same bracket, and the honest summary is that for most of the items on my list, buying beat making, and it was not close.
Then there was a third group that broke the frame entirely, which I will get to.
An hour is not an hour
Here is the correction I had to make to my own method, because the arithmetic above is too clean.
My Saturday morning hours are expensive. They are the hours where I could be fixing the fence or driving somewhere with the kids or actually resting, and spending one on a project that saves a small amount is a bad trade dressed up as thrift.
My eight in the evening hours are cheap. My hands are free, I am half listening to something, and I am not going to accomplish anything else with that time regardless. Mending lives there. Anything that can be done sitting down with my hands busy and my brain elsewhere is nearly free, and should be judged on a completely different scale.
So the question is not only how many hours a project takes. It is which hours it demands. A project that insists on Saturday mornings has to earn a great deal more than a project content with Tuesday evenings. Once I started sorting by that instead of by total time, half my decisions made themselves.
What I do with the answer now
I did not quit everything that lost money. That is the part I would have gotten wrong if I had only trusted the notebook.
I kept a couple of losers on purpose, because I like doing them and enjoyment is a legitimate reason that does not need to disguise itself as economics. What I stopped doing is calling them thrift. They are hobbies. They come out of the hobby budget in my head, not the household one, and that single relabeling took a surprising amount of low grade guilt out of my week.
What I dropped was the middle group: the things I was doing neither because they paid nor because I loved them, but because I had absorbed an idea that a certain kind of person makes their own. That was most of the list. Letting it go freed up more hours than any single project ever saved me.