Samsung SDI: profit up 106-fold, but the whole gain rests on a low-base recovery
Contents
In its Q3 2026 report Samsung SDI posted revenue of KRW 4,685.6bn (+53.5% YoY), EBITDA of KRW 1,371.2bn and net profit of KRW 600.4bn, against KRW 5.7bn a year earlier. Growth accelerated: revenue was up 18.5% in Q2 and 53.5% in Q3. The EBITDA margin rose to 29.3% from 5.1%, but the comparison is against a quarter when EBITDA was negative, so the multiple says nothing about durability. Our conclusion: rather attractive – the portal model puts the upside to fair value at +31%, even though on EV/EBITDA the stock trades 39% above its own three-year average. Our view: the share is rather attractive at the current price; upside on the portal model +31%.
Our view: is there upside
Our view in numbers
| Metric | Our read |
|---|---|
| Verdict | Rather attractive |
| Upside on the portal model | +31% |
| How the model works | fundamental value-creation model: EBITDA growth times the target multiple, against market cap |
| EV/EBITDA now and 3-year average | 15.6 vs 11.2 – a 39% premium to its history |
| Net debt / EBITDA | 3.27 – high leverage |
| Strongest case for | The portal model puts the upside to fair value at +31%, and it rests on EBITDA growth already visible at KRW 1,371.2bn. |
| Strongest case against | EV/EBITDA of 15.6 against its own three-year average of 11.2 is a 39% premium to history, and the 29.3% margin was achieved on a weak base. |
| What would change the view | The verdict would move one step down if the EBITDA margin in the next quarter falls materially below 29.3%, and one step up if net profit holds near KRW 600bn for two quarters in a row. |
Key takeaways
— Revenue rose 53.5% YoY, and growth accelerated from 18.5% a quarter earlier
— EBITDA turned positive at KRW 1,371.2bn, but it was negative a year ago, so the multiple is distorted by the base
— The 29.3% EBITDA margin against 5.1% a year earlier reflects a recovery, not a new level of profitability
— Net profit of KRW 600.4bn rose 106-fold, also a low-base effect
— Leverage of 3.27x EBITDA is a level that limits room to manoeuvre
— EV/EBITDA of 15.6 against its own three-year average of 11.2 is a 39% premium to history
— The portal model puts the upside to fair value at +31%
Key figures, KRW bn
| Metric | Q3 2025 | Q3 2026 | Change |
|---|---|---|---|
| Revenue | 3 052 | 4 686 | +53.5% |
| EBITDA | 155 | 1 371 | +782.1% |
| Operating profit | -591 | 1 371 | to profit |
| Net profit | 5.66 | 600 | +10499.8% |
| Operating cash flow | -241 | — | — |
| Capex | 554 | — | — |
| EBITDA margin | 5.1% | 29.3% | +24.2 pp |
| Net margin | 0.2% | 12.8% | +12.6 pp |
Revenue rose 53.5% YoY, and growth accelerated from 18.5% a quarter earlier
In Q3 2026 Samsung SDI's revenue was KRW 4,685.6bn, up 53.5% from a year earlier. Growth was 18.5% in Q2 and 12.6% in Q1, so it has accelerated for three quarters running. For comparison, in Q3 2025 revenue was falling 22.5% YoY.
The acceleration rests on last year's low base: the year-ago quarter was one of the weakest in the series. There is sequential growth too – from KRW 3,768.8bn in Q2 to KRW 4,685.6bn in Q3 – but the annual figure looks more impressive precisely because of last year's slump.
The facts do not disclose what produced the increase. We record the fact of acceleration and its scale, but do not attribute it to a specific segment or a one-off contract.
Revenue and EBITDA by quarter
| Period | Measure | bn KRW |
|---|---|---|
| Q4 2024 | Revenue | 3754.458206 |
| Q1 2025 | Revenue | 3176.817731 |
| Q2 2025 | Revenue | 3179.406066 |
| Q3 2025 | Revenue | 3051.8206349999996 |
| Q4 2025 | Revenue | 3858.686247 |
| Q1 2026 | Revenue | 3576.353183 |
| Q2 2026 | Revenue | 3768.8081230000003 |
| Q3 2026 | Revenue | 4685.585759 |
| Q4 2024 | EBITDA | 228.776236 |
| Q1 2025 | EBITDA | 72.73740399999998 |
| Q2 2025 | EBITDA | 111.14568300000002 |
| Q3 2025 | EBITDA | −76.58853100000002 |
| Q4 2025 | EBITDA | 273.334538 |
| Q1 2026 | EBITDA | 428.038776 |
| Q2 2026 | EBITDA | 802.8405 |
| Q3 2026 | EBITDA | 1371.157976 |
EBITDA margin
| Period | Measure | % |
|---|---|---|
| Q4 2024 | EBITDA margin | 6.1 |
| Q1 2025 | EBITDA margin | 2.3 |
| Q2 2025 | EBITDA margin | 3.5 |
| Q3 2025 | EBITDA margin | −2.5 |
| Q4 2025 | EBITDA margin | 7.1 |
| Q1 2026 | EBITDA margin | 12.0 |
| Q2 2026 | EBITDA margin | 21.3 |
| Q3 2026 | EBITDA margin | 29.3 |
EBITDA turned positive at KRW 1,371.2bn, but it was negative a year ago, so the multiple is distorted by the base
Q3 2026 EBITDA was KRW 1,371.2bn, and the same figure is shown as operating profit. A year earlier EBITDA was negative – minus KRW 76.6bn – so the reported 782.1% growth reflects a move out of loss, not a multiplication of profit.
Operating profit in Q3 is also positive at KRW 1,371.2bn, whereas a year earlier there was a loss of KRW 591.3bn. This is the first quarter in the available series when the operating result is solidly positive: in Q2 2026 it was KRW 203.8bn.
The 8.8-fold (782.1%) EBITDA growth can only be read together with the sign of the base. When the base is negative, a growth percentage loses meaning, and the only correct conclusion is that the company has returned to profit at the operating level.
Net profit by quarter
| Period | Measure | bn KRW |
|---|---|---|
| Q4 2024 | Net profit | −204.579437 |
| Q1 2025 | Net profit | −226.16874900000002 |
| Q2 2025 | Net profit | −161.28750200000002 |
| Q3 2025 | Net profit | 5.664301 |
| Q4 2025 | Net profit | −207.849434 |
| Q1 2026 | Net profit | −28.140411 |
| Q2 2026 | Net profit | 342.191505 |
| Q3 2026 | Net profit | 600.405963 |
The 29.3% EBITDA margin against 5.1% a year earlier reflects a recovery, not a new level of profitability
The Q3 2026 EBITDA margin was 29.3% against 5.1% a year earlier. The 24.2 percentage point rise follows from revenue growing faster than costs, but the facts do not disclose the exact cause of the cost movement.
The net margin rose to 12.8% from 0.2% a year earlier. Net profit of KRW 600.4bn on revenue of KRW 4,685.6bn gives exactly that margin, and it is three times lower than the EBITDA margin – the gap is depreciation, interest and taxes.
Whether the 29.3% margin holds is the main question for the next report. One quarter at that level of profitability does not yet make it the new norm, especially when last year's base was abnormally weak.
Net debt
at each reporting date, current net debt / EBITDA (LTM) = 3.27
| Period | Measure | bn KRW |
|---|---|---|
| Q4 2024 | Net debt | 9692.824825000002 |
| Q1 2025 | Net debt | 10262.674551999999 |
| Q2 2025 | Net debt | 9263.860445999999 |
| Q3 2025 | Net debt | 9292.847929 |
| Q4 2025 | Net debt | 9079.915269 |
| Q1 2026 | Net debt | 9916.014469 |
| Q2 2026 | Net debt | 10605.443901 |
Net profit of KRW 600.4bn rose 106-fold, also a low-base effect
Q3 2026 net profit was KRW 600.4bn against KRW 5.7bn a year earlier – growth of 10,499.8%. The 106-fold multiple looks impressive, but last year's base was almost zero, so the growth percentage says little here.
The sequential picture is more modest: net profit was KRW 342.2bn in Q2 2026 and a loss of KRW 28.1bn in Q1. So profit is rising quarter on quarter, but Q3 is the first quarter when it exceeded KRW 600bn.
The facts do not name any one-off items that would explain the jump. We do not attribute the profit to asset sales or revaluation – there is no such indication in the available data.
Valuation vs its own history
now vs the 3-year average
| Period | Measure | × |
|---|---|---|
| EV/EBITDA | Now | 15.580935783264396 |
| EV/EBITDA | 3-year average | 11.186796843434056 |
Leverage of 3.27x EBITDA is a level that limits room to manoeuvre
Net debt at the latest annual balance-sheet date was KRW 9,079.9bn, and the ratio of net debt to trailing-twelve-month EBITDA was 3.27. That is a level at which the company can still service its debt, but there is little headroom for major investment or generous payouts.
Trailing-twelve-month EBITDA is KRW 2,777.5bn, revenue KRW 15,889.4bn and net profit KRW 706.6bn. These figures describe the annual period, not a single quarter, and they are the basis for the 3.27 ratio.
Operating cash flow and capital expenditure for Q3 2026 are not disclosed in the facts, so we cannot judge how the company is funding debt and investment in this quarter. The direction of change in leverage relative to earlier periods is also undefined – the facts contain no earlier value of this ratio.
EV/EBITDA of 15.6 against its own three-year average of 11.2 is a 39% premium to history
On EV/EBITDA the stock trades at 15.6 against its own three-year average of 11.2, a 39% premium to its history. That means the market has already priced in the profit recovery, and the margin for error is thin.
The trailing-twelve-month P/E is 48.4, which corresponds to profit of KRW 706.6bn and a market cap of KRW 34,195.5bn. A high P/E with a modest ROE of 9.7% suggests that profit has not yet caught up with the valuation.
The premium to its own history is the main argument against buying. If profit does not hold at the Q3 level, the multiple will quickly look even higher, and with it the risk of re-rating.
The portal model puts the upside to fair value at +31%
Our own valuation model, based on EBITDA growth and a target multiple, puts the upside to fair value at +31%. This is not a market consensus or a target price – it is the portal model's output, and it rests on current profit and the current valuation.
The model and the multiple point in different directions: EV/EBITDA 39% above its own history suggests overvaluation, while the model suggests upside. For this stock we treat the model's upside as decisive, because it captures the EBITDA growth already visible in the report, whereas the historical average multiple reflects periods when profit was negative.
If profit holds at the Q3 level, the premium to history will be justified by growth in the denominator. If not, the model will be wrong, and the multiple will prevail.
Valuation on the latest reported figures
| Metric | Value |
|---|---|
| Market cap | 34 196 bn KRW |
| P/E (LTM) | 48.4 |
| EV/EBITDA (LTM) | 15.6 |
| P/B | 1.59 |
| Net debt / EBITDA (LTM) | 3.27 |
| ROE | 9.7% |
| EV/EBITDA, 3-year average | 11.2 |
Bottom line
The strong side of the report is the return to profit at every level: EBITDA of KRW 1,371.2bn, operating profit of KRW 1,371.2bn and net profit of KRW 600.4bn, with revenue growth accelerating to 53.5%. The weak side is the base: a year ago EBITDA was negative, so the growth multiples exaggerate the scale of the improvement. The real question for a holder is whether the 29.3% EBITDA margin and KRW 600bn profit hold, because that is what determines whether the 39% EV/EBITDA premium to history is justified. Our conclusion: rather attractive – the portal model puts the upside to fair value at +31%. Our view: the share is rather attractive at the current price; upside on the portal model +31%.