i-Scream Media (KOSDAQ: 461300) is a Korean edtech micro cap trading at four times earnings, with 91% of its market value in cash and securities. Its software runs 93% of Korea's primary classrooms. A full deep dive from the DART filings.
The main risk: value diversion by the controlling owner.
The risk is qualitatively the same as that faced by other family businesses, but unusually pronounced - a large amount of liquid assets, an active related-party channel and historically weak minority-shareholder protections in Korea. In the case of a typical family business with value in its operations and sound governance, this point would be almost irrelevant; here, it is the central issue of the investment case.
You’re paying next to nothing for the operating business and if the cash ever flows back to the shareholders, the upside potential is considerable - but the whole case hinges precisely on that ‘if’.
First of all, thank you for reading it and digging into it! I appreciate it really.
What you say its fair I don't disagree that it's the central issue. But it's less static than it looks. Park Young-Ok, a well-known local retail investor, has started pushing the company publicly to return at least 50% of free cash flow and management has already committed to a 40% payout, and the dividend more than doubled last year. And Korea's governance reforms are slowly making this kind of hoarding harder to justify.
So nothing is guaranteed but the pressure is real and building and the odds are better than it looks.
You're right that they already had plenty of cash. The 2024 IPO was timed to the AI textbook program the government was pushing for 2025, which was the big growth story at the time. They raised about ₩76bn at ₩32,000 a share for "R&D and new business."
The idea made sense: if those AI textbooks had become mandatory nationwide, it would've been a big, recurring subscription market — and they'd have had a real reason to spend that money scaling up. But the early development was cheap enough to cover from normal cash flow, so most of the raise just sat in the bank for the time being.
Then in 2025 the government made the AI textbooks optional instead of required. The big spend never became necessary, the growth story faded, and the stock is now worth about half the IPO price — so the cash ended up in their "investment portfolio" instead.
Thanks a lot for a wonderful writeup Jakub! Any information on financials before 2021 and why suddenly the operating margins improved so drasitically post FY22? What about their financials during historical existence before 2021? Any insights into that?
Thanks! Until 2021 the company was mainly a school e-commerce business. Commerce made up about 76% of revenue at a gross margin near 31%, so operating margin stayed between 6 and 9%. In 2022 it entered the certified textbook market, which it sells through its free iScream S platform without a sales force, while commerce revenue roughly halved. That mix shift lifted gross margin from 31% to 65% and operating margin from 6% to 29%.
I also got very interested as I stumbled upon it recently along with finding your writeup on it. I'm very new to Korean small cap space. But here is something that is concerning me somewhat as I was looking through it. The debtor days seem to have gone from mid 20s to about mid 50s given the changes on the publishing side and it is becoming dominant pillar for the business. The cash conversion has been helped by offsets from a payable increase and rising in other liabilities on consol level. Stripped off, the conversion does somewhat weaken. Also, their treasury has shifted to higher risk taking via the investment portfolio that is being directed to listed Korean equities and some direct corporate bonds and hedge funds. Especially during the 2024-25 Korean market boom. This is indicative of added risk-taking appetite by the management. Even if they got lucky on Hynix and Samsung and Micron, the risk appetite has gone up it seems which I might discount for in overall portfolio concentration. But a really interesting company and situation indeed.
Thanks a lot for the feedback, I really appreciate it!
The debtor days don’t worry me. Many of the new textbooks are for the first semester of the new school year, which starts in March, and they ship in November and December. So much of the year’s sales is still unpaid on 31 December. The state pays through the Korean textbook association, so the credit risk is close to zero.
On the portfolio you are absolutely right, and it is my main concern too. They were still adding Samsung, Hynix and Hyundai in June. The Park family also controls the company which is another concern. After the sister company iScream Edu listed in 2019 the Park family sold about 22% of it while its value fell from around 200bn won to under 40bn. Management says dividends and buybacks come first and the portfolio will shrink, but I take that with a grain of salt.
I see. Thanks for the insights into the debtor cycle! I also noticed the gifting of shares from the Chairman to his son. This transition also backs continuity of dividend payments if they were to transfer the remaining shares in future given the tax bill that might arise to the holdco.
But overall, I do largely agree with your assessments and analysis. I also look at it as a business that has good competitive position and cash generation but maybe relatively bad capital allocation that is evident from their past decision making. That to me seems more positive - as bad capital allocation hasn't really hurt cash generation unlike for many other businesses. At current price, still looks a lot asymmetric compared to many shitter things that are out there. Lastly, I thank you again for your wonderful writeup!
Thanks for the writeup. Do you have any insight into why they did not provide guidance for FY26? Historically, they only provided guidance for revenue & operating profit once (July 3 '25) and seemingly have elected to not provide any for this year.
Providing guidance would assist the rerating story as it would likely trade at <4x fwd P/E assuming >5% EPS growth (versus 44% EPS growth & 34% operating profit growth in FY25).
Thanks a lot for reading and your feedback! Good spot that they’ve only done it once, and you’re right that guidance would help the rerating. I’ve no inside information but here’s how I read it: they got burned. The July forecast backfired four weeks later when the AI textbook law passed, and they ended up about 6% short. In Korea, missing a forecast can also earn you penalty points, so there’s little incentive to try again.
On the 5%, I’d watch the costs rather than the sales. No new grades this year, so sales growth should slow. But in Q1 they halved their R&D spend and kept overheads flat, so profits may hold up better than sales do. Q2 lands in the next two weeks. Happy to compare notes with you then!
Yes, I’d expect solid growth in Q2. Some of the new grade 5-6 books will be delivered for the second semester for the first time, so there’s still a bit of that lift left in Q2. It’s Q4 where I’d expect growth to flatten out, once the expansion is fully in the base.
Great find. I love such weird companies outside of traditional markets. Thank you for this post.
I see a couple of warning flags here.
1. With textbooks for grades 3-6 now done, growth is partly depleted.
2. Looks like the family is running the company as a personal investment portfolio. They probably got hit by the sell-off this month, which would show up as investment losses in upcoming quarters even if the underlying operating business is fine.
3. With 1.2M margin‑called retail accounts retail risk appetite is getting crushed. This must depress small/micro‑cap valuations.
Thanks so much for taking a look! I really appreciate the feedback. Regarding the warning flags:
They are still expanding into new subjects and grades (like middle school). Since they can still grow their market share and the price per unit increases every year, I see plenty of room for growth.
I completely agree that capital allocation is the biggest issue here. However, roughly 90% of that portfolio is actually sitting in assets like corporate bonds, US treasuries, and index ETFs. Only about 8B won is tied up in venture funds with related parties. Also, with the KOSPI still higher than its March Q1 level, I actually expect an investment gain in Q2, not a loss.
I honestly see the retail crush as an opportunity. The core operating business is excellent, so if the price drops without the fundamentals deteriorating, I will gladly keep buying.
Hi, thanks for reading my post! You’ve spotted something real. A lot of Korean companies, still don’t publish in English, and that barrier is part of why the Korea discount exists.
I just download every statement and disclosure, translate it in full, and read it start to finish. Nothing skipped.
The main risk: value diversion by the controlling owner.
The risk is qualitatively the same as that faced by other family businesses, but unusually pronounced - a large amount of liquid assets, an active related-party channel and historically weak minority-shareholder protections in Korea. In the case of a typical family business with value in its operations and sound governance, this point would be almost irrelevant; here, it is the central issue of the investment case.
You’re paying next to nothing for the operating business and if the cash ever flows back to the shareholders, the upside potential is considerable - but the whole case hinges precisely on that ‘if’.
First of all, thank you for reading it and digging into it! I appreciate it really.
What you say its fair I don't disagree that it's the central issue. But it's less static than it looks. Park Young-Ok, a well-known local retail investor, has started pushing the company publicly to return at least 50% of free cash flow and management has already committed to a 40% payout, and the dividend more than doubled last year. And Korea's governance reforms are slowly making this kind of hoarding harder to justify.
So nothing is guaranteed but the pressure is real and building and the odds are better than it looks.
Jakub, thanks for highlighting i-Scream Media CO., LTD. (461300 KS).
Looks interesting.
Why do they need to IPO?
The company had so much cash even before IPO.
Thanks
Thanks so much for reading and digging into it!
You're right that they already had plenty of cash. The 2024 IPO was timed to the AI textbook program the government was pushing for 2025, which was the big growth story at the time. They raised about ₩76bn at ₩32,000 a share for "R&D and new business."
The idea made sense: if those AI textbooks had become mandatory nationwide, it would've been a big, recurring subscription market — and they'd have had a real reason to spend that money scaling up. But the early development was cheap enough to cover from normal cash flow, so most of the raise just sat in the bank for the time being.
Then in 2025 the government made the AI textbooks optional instead of required. The big spend never became necessary, the growth story faded, and the stock is now worth about half the IPO price — so the cash ended up in their "investment portfolio" instead.
Thanks again, great question!
Thanks for sharing!
Thanks a lot for a wonderful writeup Jakub! Any information on financials before 2021 and why suddenly the operating margins improved so drasitically post FY22? What about their financials during historical existence before 2021? Any insights into that?
Thanks! Until 2021 the company was mainly a school e-commerce business. Commerce made up about 76% of revenue at a gross margin near 31%, so operating margin stayed between 6 and 9%. In 2022 it entered the certified textbook market, which it sells through its free iScream S platform without a sales force, while commerce revenue roughly halved. That mix shift lifted gross margin from 31% to 65% and operating margin from 6% to 29%.
I also got very interested as I stumbled upon it recently along with finding your writeup on it. I'm very new to Korean small cap space. But here is something that is concerning me somewhat as I was looking through it. The debtor days seem to have gone from mid 20s to about mid 50s given the changes on the publishing side and it is becoming dominant pillar for the business. The cash conversion has been helped by offsets from a payable increase and rising in other liabilities on consol level. Stripped off, the conversion does somewhat weaken. Also, their treasury has shifted to higher risk taking via the investment portfolio that is being directed to listed Korean equities and some direct corporate bonds and hedge funds. Especially during the 2024-25 Korean market boom. This is indicative of added risk-taking appetite by the management. Even if they got lucky on Hynix and Samsung and Micron, the risk appetite has gone up it seems which I might discount for in overall portfolio concentration. But a really interesting company and situation indeed.
Thanks a lot for the feedback, I really appreciate it!
The debtor days don’t worry me. Many of the new textbooks are for the first semester of the new school year, which starts in March, and they ship in November and December. So much of the year’s sales is still unpaid on 31 December. The state pays through the Korean textbook association, so the credit risk is close to zero.
On the portfolio you are absolutely right, and it is my main concern too. They were still adding Samsung, Hynix and Hyundai in June. The Park family also controls the company which is another concern. After the sister company iScream Edu listed in 2019 the Park family sold about 22% of it while its value fell from around 200bn won to under 40bn. Management says dividends and buybacks come first and the portfolio will shrink, but I take that with a grain of salt.
I see. Thanks for the insights into the debtor cycle! I also noticed the gifting of shares from the Chairman to his son. This transition also backs continuity of dividend payments if they were to transfer the remaining shares in future given the tax bill that might arise to the holdco.
But overall, I do largely agree with your assessments and analysis. I also look at it as a business that has good competitive position and cash generation but maybe relatively bad capital allocation that is evident from their past decision making. That to me seems more positive - as bad capital allocation hasn't really hurt cash generation unlike for many other businesses. At current price, still looks a lot asymmetric compared to many shitter things that are out there. Lastly, I thank you again for your wonderful writeup!
Thanks for the writeup. Do you have any insight into why they did not provide guidance for FY26? Historically, they only provided guidance for revenue & operating profit once (July 3 '25) and seemingly have elected to not provide any for this year.
Providing guidance would assist the rerating story as it would likely trade at <4x fwd P/E assuming >5% EPS growth (versus 44% EPS growth & 34% operating profit growth in FY25).
Thanks a lot for reading and your feedback! Good spot that they’ve only done it once, and you’re right that guidance would help the rerating. I’ve no inside information but here’s how I read it: they got burned. The July forecast backfired four weeks later when the AI textbook law passed, and they ended up about 6% short. In Korea, missing a forecast can also earn you penalty points, so there’s little incentive to try again.
On the 5%, I’d watch the costs rather than the sales. No new grades this year, so sales growth should slow. But in Q1 they halved their R&D spend and kept overheads flat, so profits may hold up better than sales do. Q2 lands in the next two weeks. Happy to compare notes with you then!
Thanks for the explanation! Do you think Q2 will be a decent print? Given it’s one of their ‘strong’ quarters
Yes, I’d expect solid growth in Q2. Some of the new grade 5-6 books will be delivered for the second semester for the first time, so there’s still a bit of that lift left in Q2. It’s Q4 where I’d expect growth to flatten out, once the expansion is fully in the base.
Great find. I love such weird companies outside of traditional markets. Thank you for this post.
I see a couple of warning flags here.
1. With textbooks for grades 3-6 now done, growth is partly depleted.
2. Looks like the family is running the company as a personal investment portfolio. They probably got hit by the sell-off this month, which would show up as investment losses in upcoming quarters even if the underlying operating business is fine.
3. With 1.2M margin‑called retail accounts retail risk appetite is getting crushed. This must depress small/micro‑cap valuations.
Thanks so much for taking a look! I really appreciate the feedback. Regarding the warning flags:
They are still expanding into new subjects and grades (like middle school). Since they can still grow their market share and the price per unit increases every year, I see plenty of room for growth.
I completely agree that capital allocation is the biggest issue here. However, roughly 90% of that portfolio is actually sitting in assets like corporate bonds, US treasuries, and index ETFs. Only about 8B won is tied up in venture funds with related parties. Also, with the KOSPI still higher than its March Q1 level, I actually expect an investment gain in Q2, not a loss.
I honestly see the retail crush as an opportunity. The core operating business is excellent, so if the price drops without the fundamentals deteriorating, I will gladly keep buying.
Hi, thanks for reading my post! You’ve spotted something real. A lot of Korean companies, still don’t publish in English, and that barrier is part of why the Korea discount exists.
I just download every statement and disclosure, translate it in full, and read it start to finish. Nothing skipped.
For Korean, I'd try Papago first, and DeepL is also solid for the bulk. I lean on Claude for the trickier notes since it actually gets the accounting.
Give it a go — and if you come across anything I missed or didn't think of, please let me know. I'd be happy to add it to the analysis.