Twelve errors, in the order they usually happen

The common ones are underpricing, working free, skipping the scope sentence and keeping no records, and none are signs of poor judgement. They are what a reasonable person does when they have no evidence, no measured numbers and no idea which of the hundred things in front of them matters.

They happen in roughly this order, and knowing one is coming is most of what it takes to skip it.

Before the first job

1. Preparing instead of listing. Most preparation is procrastination wearing a to-do list, and the fourth rewrite of a profile teaches you nothing about which version worked. Build the four things a buyer actually scans and publish - what you need before your first listing is the short version of what those are.

2. Offering everything. An experienced generalist reads as versatile; a new generalist reads as someone who has not decided. Lead with one nameable format and put the rest below it - choosing your first format.

3. Pricing before measuring. You cannot price an hour you have never timed, and every first estimate of minutes per job is wrong in the same direction, usually by about double. Do one complete job on practice material with the clock running before you write a number - the practice run.

4. Setting the price at the bottom of the band. Undercutting communicates inexperience, which is the one thing you are trying not to communicate, and it selects for the buyers who ask for most. Price inside the band rather than under it; why two earners charge very different rates explains why the gap is structural rather than qualitative.

In the first month

5. Working free to build a reputation. It produces a discounted form of evidence, transfers value to people who were never going to pay, and sets your floor at zero, which is a different category from a low price. Free samples are a tax on new earners is the full argument, and it is worth having in mind before somebody asks rather than while they are asking.

6. Starting work without a scope sentence. Two sentences before you begin - what you will deliver, by when - is your entire case in the one job out of fifty that goes wrong. It takes twenty seconds and it is the highest-value habit in the trade.

7. Saying yes to every addition. "While you are at it, could you also" is the most expensive sentence in this work, and the correct answer is not refusal but a price. Buyers accept being charged for extra work far more readily than new earners expect, because a person who quotes reads as a professional rather than as difficult.

8. Being unreachable in the busy hours. Early bookings go disproportionately to whoever answers first, and demand clusters in evenings and weekends everywhere. Being the first competent reply during a narrow window beats being available all week - responding fast without living online.

As the first jobs land

9. Taking a job that is clearly wrong for you. A first job that goes badly costs a review, and the review outlives the fee by a year. Decline in one sentence, quickly, without an explanation, because an explanation invites negotiation - declining work politely and quickly.

10. Not asking for the review, or asking three times. One plain sentence at the moment of delivery converts; a follow-up costs more goodwill than the review is worth. The content matters more than the star: a detailed three-star review sells better than a vague five-star one, which is the point of what makes a review persuasive.

11. Keeping no records because there is nothing to record yet. The reason to have kept them always arrives months after the moment you could have, and reconstructing a year of mixed transactions is the worst afternoon this work will ever produce. One row per job takes under a minute - records you will wish you kept - and reserving tax as the money arrives rather than at a deadline removes the other common surprise. In the US, for example, the IRS's self-employment tax is 15.3% - 12.4% Social Security plus 2.9% Medicare - charged on 92.35% of net self-employment earnings, before income tax. Every jurisdiction differs on the specifics, so check yours rather than taking a figure from a forum.

12. Concluding something from month one. Judging this trade by its first month is like judging a shop by the morning it opened, and almost everybody does it anyway. What a realistic first month looks like sets the expectation, and the number that matters - your minutes per job - is at its least representative point in exactly this window.

Why they cluster like this

Every one of these is a decision made under an absence of information, and the absence is the actual condition of the first month rather than a personal failing. That is also why they are cheap to avoid: the remedy is nearly always a decision made once, in advance, when nothing is at stake, rather than a judgement made well under pressure.

Two of them look like mistakes and are actually market conditions. Pricing at the bottom of the band feels rational because the bottom of the band is what generic assessment is now worth, and generic assessment is worth very little because it is free and instant - the technical account of what automated scoring reliably does is the map of which part of your offer is exposed to that. Working free feels rational for the same reason, and it is not, because the thing a buyer is paying a person for was never the generic part.

The precision-of-measurement question that new earners often think they need to answer first is documented on the site that owns it, and it is rarely the constraint. What buyers are actually deciding between is not quality but risk, which is the single most useful thing in the commissioning side's own account and the correction that makes most of the twelve above obvious in retrospect.

The thirteenth mistake, for completeness, is reading a list like this and acting on all of it at once. Change one thing, leave it a fortnight, see what moved. Anything faster is noise, and the platform-specific parts of the setup - listing rules, verification, payout methods - are documented by the platform itself rather than worth guessing at.