Sam

Whatever you buy, check it carries the three-peak mountain snowflake symbol and not just M+S. M+S is a tread-pattern claim with no performance test behind it; the snowflake means it passed a snow traction requirement. In some jurisdictions that distinction is also the legal one, and plenty of cheap tyres are marked M+S only.

It does cancel the subscriptions. It also detaches the payment methods, which means you can no longer refund those charges through the normal flow, and it takes the invoice history out of the dashboard where you can read it. Refund first, delete later if you even care about deleting. That deletion is not reversible and there is a reason nobody does it in that order.

The silent part is the actual lesson, and it generalises well past billing.

Alert on the gap: query for purchases where acknowledged is false and created more than an hour ago, run it hourly, page yourself if it is non-zero. Anything with a deadline enforced by somebody else's system needs a monitor with a shorter deadline than theirs, otherwise the first time you learn about the rule is when it has already cost you money.

Opens have been decorative for years. Judge the campaign on replies per 100 sends and nothing else. If that is under 2 after 200 sends, your list is wrong or your offer is wrong, and no subject line rewrite fixes either.

Mild disagreement with the emerging consensus. Not because the work is there on the day, but because being reachable inside ten minutes on day one is worth real money when you have single-digit customers. I answered a pre-sales question in four minutes on launch day and the person said afterwards that's why they said yes. You cannot do that from a standup.

The lifetime part is what gets people. I assumed the 100 was concurrent and burned most of mine on trial accounts that never came back, which turns a conversion problem into a countdown.

And verify the signed payload properly instead of decoding the body and trusting it. There is a library for this in every language now. The number of apps granting entitlements from an unverified blob is not small and it is a genuinely bad day when someone notices.

Set a review date rather than a panic threshold. Mine was: at week eight, if I do not have an onsite, I change one variable: the resume, the target list, or the level: and only one, so I can tell what worked. It converts the dread into a schedule.

Set a decision date rather than deciding continuously. 'I'll review this in March and I'm not allowed to quit before then' removes an enormous amount of daily energy drain. The oscillating is more exhausting than the work.

How long before the coding came back after three years of management?

The hours are not the cost. The interrupt is. Four and a half hours that arrive as five unscheduled fragments cost more than a scheduled day, and that is why this feels heavier than $48 an hour sounds.

So attack the interrupts, not the total:

• one support window a week, autoresponder that promises 72 hours and means it
• delete the CSV import. That one feature is generating the only ticket you cannot answer in two minutes
• move anything you patch yourself onto something managed, even if it costs $20 more

If that gets you to an hour a month on a Saturday you choose, $218 is fine money for something you already built. If it does not, you have your answer.

If you're on any regular medication, run it past a pharmacist. Some of the interactions are boring and real and nobody thinks to check.

Stop telling them where it is and put it where they are.

Restore goes on the paywall itself, as a plainly visible text button in the same view as the purchase options. That is what passes, every time, and it is what basically every subscription app you have ever used does. Settings is fine as a second location, it is not sufficient as the only one.

Then fix the close control separately, because it is a real bug and not only a review problem: 44x44pt hit area, a dark scrim or a filled circle behind it so it is visible over any background, and check it on the largest iPad in the simulator. Your reviewer was on an iPad.

Three rejections on the same line means your replies are being read as an argument rather than a change. Resubmit with the actual fix, and put the tap path in review notes as numbered steps plus a link to a twenty second unlisted screen recording.

The expensive mistake I watched a colleague make was rolling his deferred comp into an IRA at separation because a salesman told him he would have better fund choices. He did have better fund choices and he also lost the penalty free access that was the entire reason the account existed for him. Do not move anything until you understand what the move costs you.

Invert it. Instead of 100 requests a week going out, make things that cause the right people to arrive.

Comment substantively under posts from the people your prospects follow. Not agreement, not emoji - a specific fact or a counterexample from your own experience. Two or three a day, fifteen minutes. Within a couple of months the same names start showing up in your notifications, and connection requests start coming to you, which do not count against your outbound cap at all.

It is slower than blasting. It also does not have a ceiling and does not put your account at risk.

The catching is almost always the slider going loose, not the teeth. Ten minutes with pliers and a candle stub fixed mine and I stopped shopping that afternoon.

A flat first year is arguably the best training you can get. The people who quit are the ones whose first year was great, who then assumed that was normal and got badly rattled the first time it wasn't. You're learning the boring part early.

Let me defend the boring option properly, because everyone here will push you toward freelancing. A second job on a payslip means no tax return, no invoicing, no client acquisition, no scope creep, no unpaid quoting, and no month where you earn nothing. The two real downsides are that the rate is the rate forever, and the tax on a second job is often wrong at first and needs sorting out. For somebody who says outright that they hate invoicing, that is not a footnote, it is the decision.

My regret was not the field, it was the sequencing. I quit first, then studied, and burned seven of my nine months of savings before the first interview because I badly underestimated how long hiring takes when you have no relevant title yet. The people I know who switched smoothly did the first eight months of it at 6am while still employed, and only left once they had something adjacent on the CV. Being unemployed makes you take the first offer, and the first offer is usually the worst one.

This, and the underrated part: the forty minutes a day is your spec. I wrote down every keystroke for two weeks and the resulting doc was more accurate than anything I would have designed up front, including three edge cases I would definitely have shipped without - files with a BOM, duplicate SKUs that differ only in casing, and one client who mails xlsx with a .csv extension.

The framing that the tool measures what advertisers can buy rather than what people search is the thing I was missing.

For the number itself: work out your hours on the extra work, multiply by the rate you wish you'd charged, and don't discount it because you feel awkward. If you undercharge the change order you've just taught them that additions are free.

Open source with a local file store plus your own backup is the belt-and-braces version, and it does cost you convenience. I sync mine with a self-hosted thing and roughly once a year it does something irritating that costs me an evening. In exchange nobody can put six years of my notes behind a paywall. That trade is either obviously worth it to you or obviously not.

The structure that survives this is an earn-out. 60% at close, 40% over twelve months contingent on that account still billing. Buyers like it because it moves the risk to the person who actually knows the customer. Sellers hate it for the same reason. It is the honest split.

If a buyer offers you 1.8x flat versus 2.8x with half of it contingent, and you truly believe the big account is stable, take the contingent one. If that sentence makes you nervous, you have learned something about your own estimate.

Politely disagreeing with the whole framing. You already have a service people pay for, which is more than most people chasing passive income have. The higher-percentage move is productising it, fixed scope, fixed price, a written process, then eventually somebody else delivering it - rather than starting a brand new thing from zero in a market you don't know. Every genuinely passive business I know personally was a service business first. Going straight to passive skips the part where you learn who pays and why.