Promotions without training buyers to wait

The list price is the published starting point: 700 a lesson for a maths 8-pack, 5,600 for the pack. Every step after it takes something away: a public sale, a promo code, a refund, the payment fee. What's left per lesson is the realized price, the money Klasna actually gets.

The gap matters because each step has a different owner. Nazar runs the sales, support hands out codes, finance sees refunds and fees. Nobody sees the sum unless someone adds it up, so a 5% rise on the list can vanish into a few more codes without anyone noticing.

Watch the realized price, not only the list.

Follow a single order to see the realized price. A maths 8-pack, list 5,600 (700 a lesson), bought with CHAT10:

• CHAT10 takes 10%: 5,600 − 560 = 5,040 paid;
• the payment fee is 2.5% of what was paid: 126;
• Klasna gets 4,914, or 4,914 ÷ 8 = 614.25 a lesson.

That's 12.25% below the 700 on the list, from one code and one fee. Do the same sum for any order: list, minus each reduction the buyer sees, minus what never reaches you, divided by units.

Drawn as falling steps for a whole year, the walk from list to realized price is a price waterfall:

1. list price of the package;
2. minus public sale discounts;
3. minus promo-code discounts;
4. = what families paid;
5. minus refunds and payment fees;
6. = realized revenue, divided by lessons for the realized price.

The reductions the buyer sees come first, then what never reaches Klasna. Its value is that every step gets a size and an owner, so a slide in one step shows up instead of hiding in the total.

A sale open to everyone gives its discount to every buyer, including those who would have paid full price. Last November, NOV15 (15% off everything for three days) brought 190 orders and cost 322,980 of discount. Only 38 orders came from first-time families: 38 ÷ 190 = 20%.

So four-fifths of the discount went to families Klasna already had, and some simply moved a December purchase into November. Judge a sale by the new buyers it brings, not by its order count.

One sale is a decision. The same sale on the same dates every year teaches buyers to wait. The signs show up around the sale, not in it:

• purchases dip in the weeks before the usual dates;
• the share of packages sold at full price keeps shrinking;
• support gets “when is the next sale?” questions;
• most sale buyers have bought before.

The sale's own numbers look great because they include purchases pulled in from the weeks around it. New families arriving at full price between sales are the healthy sign.

A price cut and a bonus can look equally generous to a family and cost very different amounts. Ten percent off an English 16-pack takes 992 from the price, and after the 2.5% fee that's 967.20 of contribution. A free 17th lesson costs Klasna only what it pays out: 390 to the tutor and 30 for video and AI, 420, while the family sees a whole extra lesson worth 620.

A bonus has a second advantage: it doesn't lower the price families remember. After a 10% cut, the old price feels like a rise; after a bonus, the price never moved.

Promotions aren't banned; they need a purpose. A good one is:

• aimed at the people you want, such as a first package for accounts with no previous purchase;
• tied to a give, like paying before a date;
• off the fixed calendar, so nobody learns to wait for it;
• dated, with the price returning when it ends. As of October 2026, sale announcements in Ukraine should state start and end dates and the reduction against the previous price, according to the state consumer service.

A bonus lesson with a first package, with dates, fits all four.

How it fits together: follow the realized price down the waterfall, judge sales by the new buyers they bring and design promotions that don't teach waiting. Next: the module's case, a year of codes and a discount policy.