Indosat Ooredoo Hutchison

1. 01Build the network once, then sell access to it repeatedly

   Spectrum, base stations and backhaul are bought up front and then carry traffic for years. Capital spending ran IDR 9.4tn, 10.4tn, 11.8tn, 12.5tn and 12.4tn across FY2021 to FY2025, which is 21.9 percent of revenue in FY2025.

   EconomicsSpending a fifth of revenue every year on assets that then depreciate is why the income statement understates this business. Depreciation is a real past cost but not a current cash cost, and it is the single largest reason profit and cash diverge here.

2. 02Sell prepaid data to a saturated market

   Indonesian mobile penetration is already above one connection per person, so subscriber growth is largely finished. Revenue grew from IDR 31.4tn to IDR 56.5tn between FY2021 and FY2025, but the step change came from combining two networks rather than from finding new customers.

   EconomicsWith volume growth gone, the levers are price per gigabyte and cost per site. That is why operating margin rose from 12.6 percent to 19.1 percent while revenue growth slowed to 1.1 percent in FY2025: the gains came from the cost side.

3. 03Take the cash in before you deliver the service

   Prepaid customers pay first and consume afterwards, so the working capital position is structurally favourable. The cash conversion cycle was minus 4.1, then 6.1, 14.2, 8.9 and 6.7 days.

   EconomicsA near-zero cycle means growth does not consume working capital. The capital intensity in this business is entirely in the network, not in the customer.

4. 04Service the debt that built the network

   Debt of IDR 54.9tn against equity of IDR 39.5tn funds the infrastructure, costing IDR 4.2tn of interest in FY2025. Interest cover improved from 1.32 times in FY2021 to 2.57 times.

   EconomicsThis is the step most readers skip, and it is where the return actually goes. Operating profit of IDR 10.8tn becomes IDR 5.5tn of net profit largely because interest and tax take the difference, so the equity return depends on refinancing cost as much as on tariffs.

Cost structureDepreciation-led, which is the defining feature. Gross margin is high at 55.0 percent because network operating costs are modest relative to airtime revenue, yet operating margin is only 19.1 percent, and most of that 36 point gap is depreciation of assets already paid for. Below the operating line, interest of IDR 4.2tn takes a further large slice of the IDR 10.8tn operating profit. So the cost structure has two dominant items that are both consequences of past capital decisions rather than current trading, which is why cost discipline shows up slowly and capital discipline shows up for years.

Cash cycleStructurally favourable and close to zero: minus 4.1 days in FY2021, then 6.1, 14.2, 8.9 and 6.7. Prepaid customers pay before they consume, so the business is funded by its own subscribers rather than by a supply chain, which is the opposite of the credit-driven cycles at industrial names on this site. The practical consequence is that ISAT never needs working capital to grow. Every rupiah of financing it raises goes into the network, which is also why the balance sheet risk is concentrated in capital spending decisions rather than in trading conditions.

Unit economicsNet income is the wrong number for this business, and the gap is enormous rather than technical. Operating cash flow covered net income 1.63, 3.32, 3.92, 3.89 and 3.91 times across FY2021 to FY2025. In FY2025 that is IDR 21.6tn of operating cash against IDR 5.5tn of reported profit, because depreciation on a network already built is subtracted from profit but never leaves the bank. Anyone valuing ISAT on earnings is discounting roughly a quarter of the cash it actually produces. The discipline runs the other way too: of that IDR 21.6tn, capital spending consumed IDR 12.4tn, so free cash flow was about IDR 9.2tn. Judge a network operator on operating cash flow minus the capital spending needed to keep the network competitive, never on the profit line alone, and never on cash flow alone either.

Read FY2022 onward as the company and treat FY2021 as a different entity, because the merger that created the current business closed at the start of 2022 and it reset every ratio on the page. The evidence is unmistakable: revenue jumped from IDR 31.4tn to IDR 46.8tn, total equity from IDR 10.3tn to IDR 31.4tn and total assets from IDR 63.4tn to IDR 113.7tn in a single year. That is why FY2021 shows a return on equity of 73.5 percent, which is an artefact of a small pre-merger equity base rather than an operating achievement, and why FY2021 net margin of 21.5 percent EXCEEDED its operating margin of 12.6 percent, a combination only possible with a large non-operating gain. Anyone drawing a five-year trend line through those two points is measuring a corporate transaction, not performance. From FY2022 the record is consistent and clean: return on equity 16.7, 14.7, 14.7 and 15.2 percent, operating margin climbing 12.5 to 19.1 percent, and cash covering profit near four times every year with no gains propping up the result. The one genuine caution in the current numbers is that reported profit is small relative to the cash and the asset base, so it is sensitive to interest and depreciation assumptions rather than to trading.

Cash conversionExceptionally strong and the most important figure on the page: operating cash flow covered net income 1.63, 3.32, 3.92, 3.89 and 3.91 times. Free cash flow after capital spending was about IDR 9.2tn in FY2025, from IDR 21.6tn of operating cash less IDR 12.4tn of capital expenditure. The four-times conversion is not a quality signal in the way it would be at a consumer company, it is a structural feature of depreciating a network that was already paid for. Judge the business on cash after capital spending, since that is the amount actually available to service debt and reward equity.

A builder with no choice about it, which is the honest description of a network operator. Capital spending has run between IDR 9.4tn and IDR 12.5tn every year, never below 21 percent of revenue, because a mobile network that is not continuously upgraded loses customers to one that is. The encouraging part is that the spending is now funded from operations rather than from the balance sheet: debt has been broadly flat at around IDR 55tn since the merger while equity grew from IDR 31.4tn to IDR 39.5tn, so leverage fell from 178.5 percent of equity to 139.0 percent without shrinking investment.

DeploymentNearly everything goes into the network, and the discipline is visible in the ratio rather than the absolute. Capital spending was 29.9 percent of revenue in FY2021 and 21.9 percent in FY2025, so investment grew more slowly than revenue, which is what operating leverage looks like in this industry. Working capital absorbed nothing because the cycle is near zero. The remainder went to strengthening the balance sheet: cash sits at IDR 5.1tn and debt at IDR 54.9tn, roughly unchanged in absolute terms since FY2022 while operating cash flow rose from IDR 15.7tn to IDR 21.6tn, so the same debt is now covered by far more cash. Interest cover improved from 1.32 times in FY2021 to 2.57 times, which is the single clearest measure of progress here.

Returns trendReturns are stable but modest once the merger year is excluded: return on equity 16.7, 14.7, 14.7 and 15.2 percent for FY2022 to FY2025, with return on invested capital lower at 7.5, 11.1, 9.6 and 10.1 percent because the asset base is large and debt-funded. That spread between the two is leverage, exactly as at INDF on this site, and it means the equity return depends materially on the cost of refinancing IDR 54.9tn of debt. The valuation is where the capital intensity shows up as an opportunity. At IDR 1,900.16 the market capitalisation of about IDR 61.3tn sits against a forward model value of IDR 3,596.79 per share, a gap of PLUS 89.3 percent. Note a limitation of the relative check rather than hiding it: the EV/EBITDA anchor of 5.44 times is IDENTICAL to the peer median, because Indonesian telecoms has very few listed comparables and ISAT is effectively setting the median it is measured against. So the relative test is uninformative here and the case rests on the cash flow model. What that model is really asking is whether IDR 12.4tn of annual capital spending can be held while operating cash flow keeps rising, since that is the only route by which a capital-hungry network turns into shareholder value. Both figures are model outputs, not targets.