Every route to a private student runs through the same narrow point: somebody who has never met you decides to pay you for an hour. That is the job the marketplace was doing in exchange for its commission, and it is the only part of going independent that is genuinely hard.
Everything else transfers. You can already teach, you already have a lesson shape, and you can already tell within twenty minutes what a new student needs. None of that is what the platform was selling you.
Eight adjustments the screen forces
- shorter stages than in a classroom. Attention online runs out faster, so a stage that ran fifteen minutes in a room should run eight.
- breakout rooms replace pair work, and they need a clear task and a stated time or they collapse into silence.
- share a document, not a slide. A document everybody can type into turns a broadcast into an activity.
- annotate a shared screen, which is the online replacement for the board and is underused.
- the chat box is a speaking alternative, and quiet students will often write what they would not say.
- cameras optional, participation not. Insisting on cameras costs more in goodwill than it returns; insisting on a response does not.
- prepare the link before the lesson and paste it into the chat, because a minute of technical fumbling costs the room's attention.
- end five minutes early. Online lessons run over more easily and the last five minutes are worth least.
The first private students are not your platform students
This has to be said plainly because it is the assumption everybody starts with. Marketplaces prohibit taking students off the platform, the prohibition is in the terms you agreed to, and enforcement is real. Building your independent list out of people who arrived through a platform is not a plan, it is a risk with your current income attached to it.
Which means the list has to be built from people who never came through one, and that is a different activity from teaching. It is the activity that needs unbooked hours, and being fully booked is the state that removes them, which is the trap the platform arrangement produces on its own without anybody intending it.
A narrow thing, rather than English lessons
Nobody searches for an English teacher. They search for help with a specific problem, on a specific date, usually with a number attached: an interview next month, a band score they are short of, presenting to a board in a language they read comfortably and cannot speak in.
A teacher who is visibly the person for one of those is chosen without a comparison. A teacher offering English lessons is compared with several thousand others, which is exactly the position the marketplace put you in, reproduced independently and without the marketplace's traffic.
The narrowing feels like giving up most of the market and it is what makes the remaining part reachable at all.
Where it stops: waiting for word of mouth
The stall is expecting referrals to start on their own. They do eventually and they need two things first: a cohort of satisfied students large enough that somebody in it meets somebody who needs you, and a reason for them to mention you that is more specific than that you were good.
A teacher who is good at English lessons is hard to recommend, because the recommender has to explain what the value was. A teacher who got somebody through a specific exam or a specific interview is recommended in one sentence, and the sentence writes itself.
So the first year of independence is not a marketing problem, it is a positioning one, and the teachers who conclude that they are bad at self-promotion have usually just never given anybody an easy sentence to say.
Something that works while you are asleep
A video answering one narrow question is the cheapest thing that keeps working when you are not there. Free to publish, permanently searchable, and it does the exact job the platform's listing page was doing: it lets a stranger form a view of you before deciding to pay.
It replaced an advert in a local paper or on a noticeboard, which stopped working on the day it came down and reached only the people who walked past it.
The limitation is the timescale and it is why most teachers abandon this. Nothing happens for months, the first video reaches almost nobody, and the returns arrive long after the effort. That makes it something to start while the calendar is still full, and it makes starting it when the calendar empties close to useless.
What it looks like when it goes wrong
The recognisable failure is a full calendar and a flat rate for the third year running, with the free hours that would have changed it never appearing, because an unbooked hour feels like lost income rather than like the work.
The second is a teacher who does everything right except the narrowing, and offers general English lessons on their own site at a rate the marketplace would not have charged. Nothing is wrong with the teaching and nobody can tell why they would choose this person.
Which certificate opens which door before any of this begins is which certificate is asked for, and by whom, and the reason preparation rather than price is what usually caps an hourly rate is a shape that works with one student or with twelve. Both sit under turning your own English into something you are paid for.
The open question is which narrow thing to be the person for, and the honest answer is the one you can already do that most teachers cannot, which is usually the hardest corner of the market rather than the most pleasant one: the first lessons, with no language in common.