How to Measure Progress When You Are Still Broke

The bank balance is the last number to move, which makes it the worst one to grade yourself on this week. It reports decisions you made a while back, filtered through other people's payment timing. Judge the week by it and you can decide you failed on a week where you did everything available to you.

So you need a second set of numbers. Not motivational ones. Countable ones, written somewhere you can check, about things you did yourself.

Count the actions you took, not the outcomes you received

An action is a thing you finished. You sent the message. You posted the listing. You walked into the shop and asked for the manager. Nobody else has to agree for it to have happened, and nobody can take it back off the page by not replying.

An outcome is what someone did in response. A sale, a callback, a booking, a deposit. Outcomes are the point eventually. They also sit on someone else's schedule, and no amount of wanting one tonight produces one tonight.

So the top line of your count is actions, and you keep it embarrassingly literal. "Asked six shops about window cleaning" is a count. "Worked on the hustle" is not, because on Sunday you cannot tell whether it happened.

This only holds if the actions are the ones the work actually needs, which is upstream of counting — building a goal out of actions you control comes first and is a different job.

Conversations, attempts and replies are three separate counts

Lump them into one total and you hide the exact place you are stuck.

An attempt is you reaching out: a message, a call, a knock, an application. A reply is any human response at all, including no. A conversation is a back-and-forth about the actual work — what they need, when, whether you can do it.

Keep the three in three columns and the shape tells you something a single number never will. Attempts with no replies points at your opening line, or at the list of people you are working from. Replies that never become conversations points at the offer being unclear, or at you fumbling the moment someone asks what it would cost. Conversations that never become work points at the end of the call: who was supposed to do what next, and whether you said it out loud.

You cannot fix a total. You can fix a column.

Write down what you learned in a form you can reread

"I learned a lot this week" is not retrievable. One line per attempt is.

Keep a single note, newest entry at the top, three things in one sentence: what you said, what came back, what you would change. "Asked at the front desk, got told to email, never got given the address — next time ask for the email before I leave." That is a note you can act on in April.

Read it as a stack rather than a diary. Across a run of entries the same obstacle turns up wearing different clothes, and the change you keep writing down is the change you have not made yet.

The weekly review that takes ten minutes

Pick a day and keep it. Put a ten-minute timer on your phone so the review stays a review and does not turn into an evening of replanning.

Tally the three columns. Read the week's learning lines. Then answer one question in writing: what is the one change I am making next week? One. Not a new plan, not a new business, one change to the thing you already started.

Then close it. A review that ends with you redesigning your whole approach is avoidance wearing a clipboard.

Money in and money out belongs somewhere else entirely — keeping a simple budget with a notebook and your phone covers that ledger, and folding it into this one turns ten honest minutes into an hour of dread.

Metrics that flatter you and should be dropped

Some counts climb while your situation stands still. Hours "worked" with the laptop open. Videos watched. Tools installed. Business names considered. Tabs read. Followers.

The test is blunt. Could you post a huge number in this metric for a whole week and still have nobody new know that you exist and what you do? Then it is a comfort metric. It measures your relationship with the idea rather than your contact with the world.

Drop it from the count. Not because screen work is worthless — setup has to happen once — but because a metric you can max out alone will quietly become the only one you feed.

Follower counts deserve their own warning. They move for reasons that have nothing to do with whether anyone will pay you, and they are extremely satisfying to refresh.

When all the numbers move and nothing improves

Counts can climb for a long stretch with no money at the end of them. That is information, not a verdict on you. Something upstream is wrong and your columns are pointing at it.

Work through it in order. Wrong people: you are reaching a group who do not buy this, however well you ask. Wrong ask: you are describing yourself instead of naming the job and saying what happens next. Wrong work: it has no buyers where you are, at the hours you have, without the transport you do not have.

Sometimes the answer is not in the count at all. If the real blocker is a dead phone, no childcare, or no right to work where you live, no measurement system touches that, and the honest move is to spend the week on the blocker instead of on attempts.

If the counts are healthy and payments have started arriving in ones and twos, you have a different question in front of you — turning a side hustle into repeat income starts after the first yes, not before it.

And if you are deep in the flat part where these columns are the only evidence you have, that stretch has mechanics of its own: keeping going when nothing is working yet.

Start the count tonight, on paper if the phone is dead. Three columns, one learning line, and the day of the week you will read it back.