Case: the September 2027 price change

After a price change, the first month is the one everyone looks at, and the one most likely to mislead. Buyers don't only react to a new price; they react to its announcement, and that moves purchases in time.

Klasna announced +50 UAH per lesson on every package on 1 June 2027, effective 1 September; packages bought before then kept the old price. English went to 700 / 670 / 630 for the 8, 16 and 32-lesson packs. In September 2027 English sold 4,992 lessons against 8,240 in September 2026: −39.42%. Leo read it as "we broke the business".

That number is real, but it answers the wrong question. It compares one month that lost its buyers to the summer before it with a normal month a year earlier. To judge the change, you need to know where those buyers went, and that means reading the months around the change, not the month after it.

When people know a price will rise on a date, many buy before it. That's pre-buying (also called stocking up or forward buying), and with packages it has a typical shape: buyers choose the biggest package to lock in as many lessons as possible at the old price.

Klasna's summer shows it clearly. In June–August 2027 families bought 195 English 32-packs, against 90 in the same months of 2026: +116.67%. Summer lessons overall rose from 9,440 to 13,952. Those families didn't disappear in September; they had already bought.

Pre-buying isn't a loophole. Klasna told families three months ahead that purchases before 1 September keep the old price, which is what makes the rise fair. It simply moves sales from September into the summer, so the summer looks too good and September too bad. Neither month alone tells you what the price did.

If pre-buying moves sales from one month to another, the fix is to read a window long enough to hold both: the months of buying ahead and the dip that follows. Inside the window the shift cancels out, and what's left is closer to the real effect of the price.

For Klasna's change the window is June to October: from the announcement to the first month after the dip. English sold 26,128 lessons in June–October 2027 against 24,640 a year earlier: +6.04%. The chart shows the shape: summer bars well above last year, a deep September, October back above 2026.

Choose the window by the mechanism, not by the result you hope for. It starts when buyers learn about the change and ends after the pre-bought demand has worked through. If pre-bought lessons will still replace purchases later, the window isn't closed yet, so plan to read it again.

A price rise should change money more than volume: each sale is worth more, even if there are a few fewer. So read both over the same window.

For English in June–October: lessons +6.04% (26,128 vs 24,640), sales in UAH +9.74% (16,540,160 vs 15,072,000). Money grew faster than lessons because many lessons sold at the new price. It grew less than the +50 alone would suggest because much of the summer was sold at the old price, a lot of it in 32-packs with the lowest price per lesson.

To get the money figure, sum sales_uah for June to October in each year and compare: 16,540,160 / 15,072,000 − 1. Don't average the monthly percentages; months of different size would get the same weight.

The gap between the two growth rates is a first sign of what the price did. The next lesson in the module, the yearly review, splits it properly into price, volume and mix.

"Up on last year" sounds like success, but last year is only the right yardstick if nothing else was changing. If the business was already growing, the honest comparison is last year plus that growth: what you'd have sold anyway.

In May 2027, before the announcement, English was already selling 7.95% more lessons than a year earlier. At that trend, June–October would have sold about 26,598.56 lessons (last year's 24,640 grown at May's rate). Klasna sold 26,128: 1.77% below the trend. So the price did cost some volume, about 1.8%, while sales in UAH grew 9.74%.

Both simple readings were wrong. Leo's first one, −39% from September alone, was far too gloomy. His second, "lessons are up 6%, we lost nothing", ignores growth that was already there. The trend line gives the answer in between, and that's the one to act on.

Inside a window, some months are distorted and some are clean. The cleanest is the first month when buyers face the new price as a normal choice: the announcement is old news, and pre-bought demand has mostly worked through.

For Klasna that's October 2027. English sold 7,184 lessons against 6,960 in October 2026, +3.22%, while sales grew 11.75% and the average price per lesson sold rose from 611.61 to 662.14. In plain words: families were buying at the new price, about as many as before, and each lesson brought in more.

A clean month is still one month. Some families who stocked up in summer are still using those lessons and will buy again later than usual, so November and December may stay a little soft. That's why Klasna reads the change once more in January, when the pre-bought lessons are used up.

Pre-buying moves cash forward and the work later. A lesson bought in August at the old price is taught in autumn, when the costs are already the new ones. So for a while the business delivers old-price lessons at new costs.

At Klasna, a 32-pack lesson bought at 580 in August is taught this autumn at the new payout of 410, with video and AI at 31.47: 580 × 0.975 − 410 − 31.47 = 124.03 UAH. The same lesson sold at the new 630 leaves 172.78. Until the pre-bought lessons are used up, part of the autumn earns less than the new price list suggests.

The rise still reaches the money, just later than the price list says. Plan for it: the cash arrives in summer, the lessons come in autumn, and tutors need the hours then. Don't spend the summer's cash as if those lessons were already delivered.

Pre-buying tempts people to cut the notice: two weeks instead of three months, so fewer families stock up. That trades fairness for a prettier September. The notice is what made the rise fair; pre-buying is the price of that fairness, and you can plan for it.

The other temptation is to react to the loudest month. Rolling the change back in September would have taught families that a bad month makes Klasna blink, and the next change would be harder. The protection is to decide before the change which window and which yardstick will judge it: here June to December against the trend, never September alone, plus what result would make you reverse.

How it fits together: don't judge by the month after; expect pre-buying; read a window in lessons and money; compare with the trend; use the cleanest month; plan for old-price lessons at new costs. The last module asks how to test a price before changing it, and how to review prices every year.