The traits that predict sticking with it

Nothing about this work requires a skill you would put on a CV. What it requires is a temperament, and the people who leave overwhelmingly leave for temperament reasons rather than because they turned out to be bad at it.

So the useful self-assessment is not "could I do this" - almost anyone could - but "would I still be doing it in four months".

The five questions that actually predict it

Can you do a small dull task on a schedule nobody is enforcing? This is the big one. There is no manager, no shift, and no consequence for skipping a week except a slow decline in placement that you will notice a month later. People who need external structure find this much harder than they expect, and it has nothing to do with how motivated they feel in week one.

Can you tolerate four quiet weeks without concluding anything? The first month is usually thin, and the honest reading of a thin month is that it contains too few data points to mean anything. People who need early feedback to keep going are the largest single group that leaves. What a realistic first month looks like is worth reading before you start rather than during, because reading it while discouraged is much less effective.

Are you genuinely comfortable with the subject matter? Not "can you handle it" - comfortable enough to be matter-of-fact about it for hours, in writing, repeatedly. Discomfort does not fade with exposure as reliably as people assume, and it shows in the work. The conventions around what buyers submit and how it gets handled are set out from the other side in the account of photo submissions and their handling, and reading that is a fair test of whether the material is a non-event for you.

Can you write something specific about a thing you have looked at for four minutes? The whole product is specificity. A person who can only produce general praise is producing what a free tool already produces, and will be priced accordingly.

Do you have two or three usable hours in the evenings or at weekends? That is when demand clusters. Someone available only on weekday mornings can make this work, but is starting harder than they realise.

Three yes answers out of five is workable. Two is a warning, and the two that matter most are the first and the third.

What does not predict it

Speed at writing. Prior freelance experience. Confidence. Any equipment you own.

None of these show up in who is still working four months later, and two of them - confidence and speed - are mildly negatively associated in practice, because both encourage taking too much on in the first fortnight.

Nor does interest in the subject, which is a different thing from comfort with it. The work is repetitive, and enthusiasm is not the resource that gets you through repetition.

The reasons people actually leave

Worth knowing in advance, because each has a countermeasure and none of them is a character flaw.

The first month was quiet and they read it as a verdict. Countermeasure: decide in advance how long you will give it, write the date down, and do not revisit before then.

They priced at the bottom, got the hardest buyers, and burned out on them. This is the most preventable one. Why the cheapest tier attracts the hardest buyers is the mechanism, and it catches almost everyone who starts by undercutting.

Scope crept until the hourly rate was insulting. Countermeasure is a written scope before you need one, and the phrasing in the scope sentence that prevents most disputes takes about a minute to adopt.

One bad interaction and no plan for it. Difficult requests and the occasional unpleasant buyer are part of the trade at low frequency. Having decided in advance what you decline, and how, converts these from crises into two-line messages. It is common across paid platform work generally: in Pew Research Center's 2021 survey of US gig work, 37% of people who had earned money this way said they had been treated rudely at least sometimes, and 13% often.

Their life changed and the work did not fit it any more. Perfectly normal, and the only one on this list that is a genuine ending rather than a preventable exit.

The two things to sort out before you decide

Both are decisions rather than tasks, and both are harder to change later.

How separate this stays from the rest of your life. Working name, separate email, what you tell people. Separation is easy to set up at the start and awkward to retrofit, and the general treatment of keeping identity and material separated is the right frame for it. The earner-specific version is anonymity as an earner.

Whether you want it to grow. A practice you intend to keep small should be priced and scheduled completely differently from one you intend to build on, and defaulting to neither is how people end up with the worst of both.

If the answers point yes

The thing that separates the people who stay from the people who do not is almost entirely front-loaded: they survive the evidence-free stretch at the start, and after that the work becomes ordinary.

Buyers are also less intimidating than the anticipation suggests. The buyer-side account of how someone chooses who to commission reads as risk-management rather than judgement of talent, which is a useful thing to have in mind the first time an enquiry lands.

If the answers point yes, the next question is what you actually need in place before listing, and it is a shorter list than most people assume - what you need before your first listing covers it. The platform-specific requirements sit alongside that on Rate Cock's judges page, and those are worth reading last rather than first, because the general decision should not be made inside one marketplace's onboarding flow.