Test the Demand Before You Sign Anything

A free session tells you nothing

A free session fills a room and settles nothing. The parents who turn up are answering a different question from the one you need answered, which is whether this class is worth money to them in a month that already has school fees in it.

Marketing Science research on zero pricing found that dropping a price to zero shifts choice far more than the size of the cut would suggest, because people treat free as adding value rather than only removing cost. A free class is not a cheap class. It is a different product, and the people who come to it are a different group from the people who would have paid.

A survey is weaker still. Saying yes to an interested question costs a parent nothing and commits them to nothing, and you will collect agreement from people who were never going to appear on a Saturday.

The one signal that survives contact with a real Saturday is money leaving an account before the session starts. Everything below is built on that.

What does a pilot actually cost to run?

Less than one month of the lease you were about to sign. Rooms for hire are an ordinary commodity in Indonesian cities, and XWORK, a national booking marketplace, advertises meeting rooms from Rp 50.000 per hour.

Cheaper options sit underneath that. A community hall, a cafe with a back room, a friend's studio on its closed day, a mosque or church hall, and your own front room all host a first cohort perfectly well, and none of them asks you to commit to an address.

Borrow the equipment rather than buying it. If the class needs mats, keyboards, robotics kits or art supplies, borrow, hire or buy the smallest quantity that lets a session run, because the point of the whole exercise is to avoid owning anything you cannot resell.

The one thing worth spending on is a room you are not embarrassed by. A pilot in a bad space produces a false negative, and a false negative at this stage costs you the business you were about to build.

What do you charge for it?

Charge what one session of the real class would cost. If you intend to sell a term at a given price, divide it and charge that, because a discounted pilot measures appetite for a discount and you are not planning to sell one of those.

Take the money before the session rather than at the door. Payment on arrival sits close enough to free to carry the same problem, and a parent who transfers on Tuesday for a Saturday class has made a decision you can count.

Say the refund out loud and keep it honest. Offer the money back if the session does not run or the child hates it, because the price has to work as a test rather than as a barrier, and you are going to ask these same parents for the truth afterwards. Taking a paid place through a booking page for trials and sessions rather than a chat thread also means the payment and the seat arrive attached to each other.

How many sessions is enough?

Three or four, run on the day and at the hour you actually intend to teach. One session tells you whether a novelty draws a crowd, and it is the second and the third that tell you whether the same families come back without being chased.

Run them a week apart rather than as a block. A holiday intensive is a different product from a weekly class, and if the weekly slot is what you plan to sell, the pilot has to be weekly or it is testing something you are not going to offer.

Cap the room and say the cap out loud. A pilot that can absorb everybody teaches you nothing about scarcity, and a pilot that fills is the only version of this that ends with somebody waiting for the next one.

The three numbers worth counting

Count enquiries against posts rather than against weeks. One post in one parents' WhatsApp group either produces conversations or it does not, and a monthly total hides which post did the work and which audience was actually listening.

Count paid seats separately from attendance. A friend's child sitting in as a favour is not a customer, and the figure you carry forward is the number of unrelated families who transferred money without being asked twice.

Count the parents who ask whether it runs weekly. That question is the real signal in a pilot, because it is a parent doing your planning for you, and it comes from someone who has already decided that the answer matters to them.

Two of the three stay straight on their own if the pilot is sold as a ticketed session instead of a chat thread. Happy Kamper sells a session by timeslot and price band, holds a capacity cap with a live count, and takes the booking directly, so a paid place and an empty one are recorded as they happen rather than reconstructed afterwards from a phone.

The third number stays yours to keep. Happy Kamper does not carry an enquiry pipeline with stages, owners or follow-up dates, so the parents who asked and have not yet paid live in whatever notebook or spreadsheet you already use, and somebody has to look at it on a named evening.

A failed pilot is usually about the format or the hour

Most pilots that fail are not telling you that nobody wants the class. They are telling you about the format or the time of day, and those are the two cheapest things you own to change.

Read the shape of the failure. Nobody enquiring at all is a reach problem rather than a demand problem, because the class was never seen by anyone in a position to want it. Enquiries that never turned into payment point at the price, the day or the description, and paid seats that did not come back point at the session itself.

Change one thing and run it again. A Saturday morning pilot that empties in the second week and a Wednesday afternoon pilot that fills are the same class taught by the same person, and the gap between them is the entire finding.

The failure worth respecting is silence at every stage. Posts that produce no conversations, in more than one place, across more than one attempt, is the version of this that should stop you spending, and it is a far cheaper way to learn that than a year of rent on an empty room.

When does the paperwork start to matter?

Later than most people fear, and that is the argument for piloting first. Permendikbud 81/2013 governs the establishment of a nonformal education unit in Indonesia, the category that covers a course provider and a bimbel, and the file it asks for describes a settled business rather than an experiment.

One requirement sets the order of everything else. Among the administrative documents is evidence that you own or hold the right to use the teaching premises for three years, which puts the premises before the licence, and the pilot has to come before both unless you are willing to commit to an address in order to test an idea.

The rest is less daunting than its reputation. The founder can be an individual rather than a company, the application goes to the education office of your own kabupaten or kota rather than to a province or a ministry, and that office has thirty working days from receipt to approve or refuse it.

What the regulation does not settle is whether a handful of paid sessions in a borrowed room amounts to establishing a unit at all. That is a question of fact about your particular arrangement, and the office that would issue the izin is the one to put it to, in writing, before a pilot becomes a term.

If the pilot works and you do turn it into a term, the solutions page for activity providers covers what the running version of this looks like. This is general information about the rules named here and not advice on any specific arrangement, so put your own circumstances to a professional before you act on it.