This is investment research, not personal financial advice.
SEEK Limited (ASX:SEK) fell A$2.30, or 14.3%, to A$13.77 on 12 August after a result that looked healthy until management described the next year. FY26 net revenue rose 10% to A$1.20 billion, EBITDA rose 15% to A$529.9 million and free cash flow increased 21% to A$246.0 million. Yet paid-job-ad volume fell 6% across APAC, and the midpoint of FY27 guidance implies only about 3% growth in adjusted profit. An independent same-day account also recorded the 14% fall and the weak outlook as the trigger (SEEK 2026a; Rask 2026).
The market's response was broadly proportionate. SEEK proved that it can extract more revenue from each listing, but the result did not prove that hirers will keep absorbing double-digit yield increases while volumes contract. The post-result price still values the core business at roughly ten times a reasonable FY27 EBITDA base after giving partial credit to the investment portfolio. That is not a distressed rating. It asks pricing, product investment and portfolio liquidity to work at the same time.
The result split growth into price and volume
The headline growth came from yield. Core job ads and enhancements produced A$1.08 billion of FY26 net revenue, up 10%, while expanded sourcing and other revenue rose 8% to A$122.7 million. Paid-ad yield increased 18% across APAC, the sixth consecutive year of double-digit growth. Paid-ad volume declined 6% (SEEK 2026a).
Those two numbers describe the investment question better than adjusted profit does. A marketplace that can raise yield by 18% while volume falls 6% has pricing power. A marketplace that needs another year of roughly 10% yield growth to offset lower volume has not escaped the cycle.
ANZ was the stronger half. Revenue rose 12%, paid-ad yield rose 14%, and volume fell 2%. Management attributed the yield gain to upgraded ad tiers, the Advanced product launched in the second half of FY25, value-based pricing and higher adoption of enhanced listings. New Zealand volume rose 8% after several weak years, but Australian hiring softened in the second half.
Asia showed the limit more clearly. Revenue rose 3% as yield increased 20%, but paid volume dropped 12%. Some of that decline is deliberate because SEEK is moving several Asian markets to a freemium model. A free Lite listing can increase inventory and candidate utility without appearing in paid volume. Some is economic weakness, particularly in Hong Kong. The accounting result cannot separate those effects cleanly, so the FY27 volume comparison matters more than one year's consolidated decline.
The macro evidence is consistent with a soft, not collapsed, hiring market. Australian job vacancies were 329,500 in May 2026, down 2.1% from February and 30.3% below the May 2022 peak (ABS 2026). That helps explain why hirers posted fewer ads. It does not by itself explain an 18% yield increase. The additional yield came from product mix and pricing decisions, which is why customer response now matters.
FY27 guidance sets the next test. Management expects revenue of A$1.21 billion to A$1.28 billion, EBITDA of A$530 million to A$580 million and adjusted profit of A$185 million to A$215 million. At the midpoint, that is about 5% revenue growth, 5% EBITDA growth and 0.5% adjusted-profit growth against continuing-operations FY26 adjusted profit of A$199.1 million. Against total adjusted profit of A$206.7 million, the midpoint is roughly 3% lower. The different bases are a reminder that portfolio and discontinued items can distort a neat growth comparison. Either way, the implied acceleration is modest (SEEK 2026a; SEEK 2026b).
A marketplace with a toll, not an annuity
SEEK matches candidates with hirers across Australia, New Zealand and six Asian markets. Hirers pay for listings, upgraded placement, sourcing tools and adjacent services. Candidates supply profiles, applications and behavioural data. More candidate activity can improve matching; better matching can lift a hirer's return on spend; that return supports higher listing yield. This is the reinforcement loop behind management's value-based pricing.
The unified APAC platform, completed in FY24, matters because one product stack can deploy search, recommendations, ad tiers and AI tools across markets. It also concentrates execution risk. SEEK now reports APAC as one operating segment because costs and resources are managed on a consolidated basis. Geographic revenue remains visible, but the cost base is no longer allocated in a way that allows a clean ANZ-versus-Asia return comparison (SEEK 2026a).
The moat evidence is tangible but incomplete. SEEK reports number-one placement share in all core markets, strong brand awareness in ANZ and the highest recent placement share in Asia. Its May 2025 investor day linked proprietary data and AI capability to placement growth and yield (SEEK 2025b). Six straight years of double-digit yield growth supports the claim that hirers receive enough value to accept higher spend.
Counter-evidence sits beside it. ANZ volume fell despite placement leadership. Asia's paid volume fell much faster as freemium and weak hiring overlapped. Competitors can aggregate listings, use general search traffic or build AI-led matching without copying SEEK's historical interface. Recruit Holdings' Indeed provides a global peer with a larger data pool and a pay-for-performance model, which keeps pressure on local marketplaces to prove measurable hiring outcomes rather than charge for visibility alone (Recruit 2026).
AI cuts both ways. SEEK can use it to rank candidates, improve high-fit applications and reduce hirer screening time. The company is adding product, technology and AI staff, as well as compute infrastructure. But its FY26 risk report also says AI may change discovery and user behaviour, introduce biased or inaccurate outcomes and increase dependence on external model providers. A product tool becomes moat only when it improves placement share or yield faster than it raises cost and substitution risk.
That leaves the moat classified as stable rather than widening. Placement share and yield are holding. Volume, capital intensity and a changing discovery layer stop the stronger conclusion.
Five years show a resilient core and a noisy perimeter
The history below uses reported continuing-operations figures in Australian dollars. NPAT is statutory profit attributable to SEEK owners from continuing operations, not management's adjusted profit. OCF is net cash from operating activities. The ROIC column is an author-computed operating-capital proxy described below, and therefore does not appear as a source-reported figure in the source ledger.
| Year | Revenue (A$m) | EBITDA (A$m) | Statutory NPAT (A$m) | OCF (A$m) | Capex (A$m) | Net debt (A$m) | Net debt / EBITDA | Computed operating ROIC proxy |
|---|---|---|---|---|---|---|---|---|
| FY22 | 1,116.5 | 509.1 | 240.8 | 406.3 | 132.9 | 1,053.1 | 2.07x | n/m |
| FY23 | 1,225.3 | 546.1 | 202.7 | 325.2 | 189.4 | 1,063.8 | 1.95x | 21.0% |
| FY24 | 1,084.1 | 468.9 | (59.9) | 312.8 | 169.0 | 1,032.8 | 2.20x | 13.1% |
| FY25 | 1,090.4 | 459.2 | 238.3 | 332.8 | 130.0 | 947.1 | 2.06x | 11.1% |
| FY26 | 1,198.9 | 529.9 | (306.5) | 397.7 | 151.7 | 962.4 | 1.82x | 14.2% |
Sources: SEEK annual reports for FY22 through FY26 (SEEK 2022; SEEK 2023; SEEK 2024; SEEK 2025a; SEEK 2026a). Parentheses denote losses. Capex differs slightly from the total-expenditure table in years where cash paid and capitalised additions differ.
Revenue ended FY26 only 7% above FY22, while EBITDA rose 4%. The path was anything but smooth. Revenue and EBITDA peaked in FY23 as labour markets remained tight, fell through FY24 and FY25, then recovered in FY26. The FY26 EBITDA margin reached 44.2%, up from 42.1% in FY25 and close to FY22's 45.6%.
Statutory NPAT is a poor guide to the core marketplace. FY24 included a Zhaopin impairment. FY26 included A$304.4 million of continuing-operations significant items, mainly a A$283.6 million Zhaopin impairment, and a A$201.2 million after-tax loss from SEEK's share of the Growth Fund. Continuing-operations statutory NPAT was therefore negative A$306.5 million even though adjusted profit was A$199.1 million (SEEK 2026a). The charges are non-cash, but they are not meaningless. They record prior capital whose expected recoverable value fell.
The operating return proxy removes the carrying values of the Growth Fund, Zhaopin and other equity-accounted investments from year-end equity, then adds net debt. Adjusted continuing-operations profit is divided by that residual capital base. It produces 21.0% in FY23, 13.1% in FY24, 11.1% in FY25 and 14.2% in FY26. This is not textbook ROIC because it uses year-end capital and adjusted profit after interest rather than NOPAT on average invested capital. The structure of SEEK's accounts makes the conventional denominator misleading: up to 48% of assets have sat in equity-accounted investments outside the core marketplace. The proxy is useful for direction, not precision.
The direction says the core still earns an attractive return, but incremental investment has yet to restore FY23 economics. Capital expenditure rose from A$130.0 million in FY25 to A$151.7 million in FY26. Depreciation and amortisation reached A$162.0 million as five years of product development flowed through the income statement. The unified platform is complete, yet investment in product, technology, AI and compute continues.
Free cash flow needs one more subtraction
SEEK defines free cash flow as net cash from operating activities less capital expenditure. On that basis, FY26 free cash flow was A$246.0 million:
| FY26 owner-cash bridge | A$m |
|---|---|
| Net cash from operating activities | 397.7 |
| Less capital expenditure | (151.7) |
| Company-defined free cash flow | 246.0 |
| Less Growth Fund capital contributions | (34.0) |
| Less Growth Fund management fees | (20.0) |
| Less fees for other managed assets | (5.0) |
| Add distributions from equity-accounted investments | 7.8 |
| Cash after portfolio funding and fees | 194.8 |
| Less share-based payments, treated as an economic cost | (29.3) |
| Author-computed owner cash | 165.5 |
On an attributable-per-share basis, the first three lines are A$0.83, A$0.42 and A$0.46 respectively.
The first three lines are reported. The remainder is an author calculation using the FY26 cash-flow review and share-based-payment expense (SEEK 2026a). It is deliberately conservative. Share-based payments do not consume cash in the period, but ignoring them treats dilution as free. Fund contributions may later return as proceeds, but they were not available to owners in FY26.
The post-fall market capitalisation was A$4,929.25 million. Company-defined free cash flow represents 5.0% of that equity value. The stricter owner-cash measure represents 3.4%. SEEK paid A$175.0 million of dividends during FY26, slightly more than this stricter cash measure. Another A$16.4 million of net borrowing entered the financing bridge.
Cash conversion was strong on management's definition. Operating cash before interest, transaction costs and tax was A$551.6 million, or 104% of EBITDA, helped by supplier-payment timing. Yet net debt still rose A$15.3 million to A$962.4 million. That combination explains why the owner-cash bridge matters. The marketplace generated more cash, but dividends and the investment perimeter absorbed it.
Balance-sheet survival is not the immediate issue. SEEK had A$136.6 million of cash and A$554.9 million of undrawn facility capacity at June 2026. Net debt was 1.82 times FY26 EBITDA on a simple author calculation, while management's covenant framework uses a defined borrower-group measure. The constraint is allocation: debt, dividends, product investment and fund contributions are all drawing on one core cash engine.
The Growth Fund is an asset and a discount
SEEK carried its 83.8% interest in the Growth Fund at A$1.50 billion at June 2026, down A$276.6 million over the year. The fund's total portfolio value, including distributions, fell 13% to A$1.98 billion. Four assets, Employment Hero, Go1, HiBob and Online Education Services, represented 74% of portfolio value. Since inception, the fund had distributed about A$174 million to SEEK, equal to 12% of invested capital (SEEK 2026a).
This creates a valuation trap. Adding the full carrying value to the core business assumes that private marks convert into cash without tax, fees, minority leakage or timing risk. Ignoring the fund entirely assumes assets with external funding evidence and an announced divestment program are worthless. Neither treatment fits the evidence.
Management expects the fund's next phase to include near-term divestments of investments currently valued above A$1 billion. Employment Hero is already part of that process, and an update is expected before the end of calendar 2026. The word "valued" is doing work. It is not proceeds. FY26 included A$18.3 million of fund management fees, A$34.0 million of capital contributions and only A$7.4 million of distributions to SEEK.
Zhaopin offers the cautionary comparison. SEEK recorded A$356.0 million of total FY26 impairments across its equity-accounted stake and a net receivable. The remaining equity-accounted investment was A$172.3 million, with a A$7.7 million net receivable. A non-core asset can be economically valuable and still produce repeated write-downs before liquidity arrives.
Capital allocation therefore has two scorecards. The core marketplace is funding product investment and dividends. The portfolio has produced selective gains and optionality, but its fees, contributions and valuation swings obscure the return on that core. A completed transaction with disclosed net proceeds would narrow the discount. Another mark without cash would not.
What A$13.77 already assumes
The post-result equity value is A$4.93 billion using 357.97 million shares. Adding A$962.4 million of net debt gives an enterprise value of A$5.89 billion, or 11.1 times FY26 EBITDA. The stock also trades at 24.8 times FY26 adjusted continuing-operations profit. Those simple multiples include the fund and Zhaopin inside the equity value, so they overstate the multiple on the core if the investments realise meaningful proceeds.
A sum-of-parts bridge is more suitable than a single earnings multiple. The scenario values below use:
(core EBITDA × enterprise multiple + realised non-core value − net debt) ÷ 357.97m shares
The method gives no credit for portfolio marks until a scenario assigns a discounted amount. It also leaves net debt at FY26 levels, which avoids treating uncompleted asset proceeds as balance-sheet repair.
| Case | Core EBITDA | EV / EBITDA | Realised non-core value | Midpoint value per share |
|---|---|---|---|---|
| Severe downside | A$400m | 6x | A$0 | about A$4.0 |
| Bear | A$480m | 8x | A$200m | about A$8.6 |
| Base | A$550m | 10x | A$500m | about A$14.1 |
| Bull | A$650m | 12x | A$900m | about A$21.6 |
The frontmatter ranges widen those point estimates to reflect debt, tax and execution uncertainty. They were built from business drivers before comparison with the market price.
The A$13.77 close sits near the base midpoint. Reverse the calculation and assume A$500 million of net non-core value: the market is assigning about A$5.39 billion of enterprise value to the core. That is 10.2 times FY26 EBITDA. If no non-core value arrives, the implied core multiple is 11.1 times. If A$900 million arrives, it falls to 9.4 times.
Sensitivity is concentrated in two variables:
| Core EBITDA / multiple | 8x | 10x | 12x |
|---|---|---|---|
| A$480m | A$9.4 | A$12.1 | A$14.8 |
| A$550m | A$10.9 | A$14.1 | A$17.1 |
| A$650m | A$13.2 | A$16.9 | A$20.5 |
This table assumes A$500 million of non-core value and A$962.4 million of net debt. Every two turns of EBITDA multiple changes equity value by roughly A$3.1 per share at A$550 million of EBITDA. Every A$100 million of portfolio proceeds changes value by about A$0.28 per share before tax or other leakage.
The market has not priced a collapse. It has removed the premium that allowed investors to look through falling listings and trust yield, AI investment and portfolio value to bridge the gap.
The anti-thesis is stronger than a cyclical rebound
The positive reading starts with FY26 execution. SEEK grew net revenue 10% against lower paid volume, expanded EBITDA margin by about two percentage points, converted EBITDA into cash and reduced simple net-debt-to-EBITDA from the FY24 peak. The Advanced tier and value-based pricing are producing measurable revenue. A recovery in vacancies would add volume to a higher-yield base rather than replace it.
There is also hidden balance-sheet optionality. The Growth Fund's A$1.50 billion carrying value and Zhaopin's remaining value are not needed to produce core EBITDA. Net proceeds from a large divestment could reduce debt, fund capital returns or reveal that the market's portfolio discount was too severe. A$1 billion of gross asset value is material beside A$4.93 billion of equity value.
The opposing reading goes beyond the economic cycle. Hirers may be near the point where another price increase causes downgrades, lower listing frequency or migration to outcome-based alternatives. AI may lower the cost of matching enough to weaken the scarcity value of a traditional listing marketplace. SEEK's own spending shows that defending the position is not cheap: FY26 capex rose 17%, product investment has lifted amortisation, and compute expense is growing.
The fund can also distract from core economics. Private portfolio marks fell in FY26, Zhaopin suffered another impairment, and management fees continue regardless of distributions. A proposed transaction might convert value into cash, but a delayed or discounted outcome would validate the market's scepticism.
These readings meet at the same fact: yield cannot be assessed apart from volume. A cyclical volume recovery would make FY26's pricing achievement look valuable. Continued contraction after another 10% yield increase would suggest price is supporting revenue by consuming future listing demand.
Three disclosures will decide the verdict
The first is the FY27 half-year result, expected around February 2027. It should reveal whether paid-ad volume stabilised, whether 10% yield growth held and whether hirers absorbed another product-mix increase. A volume decline worse than 5% alongside double-digit yield would keep the burden on price.
The second is the promised calendar-2026 Growth Fund update. The useful facts are binding transaction value, net cash to SEEK, timing, tax and any retained exposure. A portfolio valuation without those terms will not resolve the discount.
The third is cash. FY27 owner cash should be rebuilt from operating cash, capex, fund contributions, management fees, distributions and share-based payments. A figure below A$170 million, or net debt above A$1.0 billion at the half year without offsetting proceeds, would show that the A$246 million headline free-cash-flow number is not translating into balance-sheet room.
Placement share is the operational cross-check. Stable or rising share would support the argument that SEEK is improving outcomes despite fewer listings. Falling share after heavier AI and product spending would challenge both the moat and the return on incremental capital.
The 14.3% fall correctly recognised that FY26 growth was purchased from yield while the volume base shrank. At A$13.77, the valuation still assumes a durable core marketplace, a normal rather than distressed multiple and some portfolio recovery. The next six months will show whether SEEK's pricing engine is financing a cyclical bridge or masking a demand response.
Source notes and confidence
Confidence is high for the FY22-FY26 reported history, FY26 market move and stated FY27 guidance because the underlying annual reports, results release, ASX issuer page and independent event account were fetched in this run. Author-computed ROIC, owner cash and valuation scenarios have lower confidence than the reported inputs: SEEK does not publish a clean standalone invested-capital balance for the core marketplace, and the APAC segment no longer allocates costs between ANZ and Asia.
Missing information is concentrated in three places. SEEK has not disclosed customer churn by ad tier after the latest price changes, a transaction price or net-proceeds bridge for the Growth Fund's proposed divestments, or a geographic cost allocation that would permit separate ANZ and Asia returns. The Recruit peer filing was fetched, but its embedded font prevented reliable automated text extraction, so it supports only the qualitative identification of Indeed as a global peer and carries no figures. Those gaps, and reader-liveness checks that returned 403 for several older SEEK PDF links after the same documents had been downloaded and read in this run, are why verification is marked partial and the scenario ranges are wide.
References
- ABS 2026, Job Vacancies, Australia, May 2026, Australian Bureau of Statistics.
- ASX 2026, ASX company page for SEEK Limited (SEK).
- Google Finance 2026, Market snapshot for SEEK Limited (ASX:SEK).
- Rask 2026, Seek share price sinks 14% on FY26 result, weak outlook, 12 August 2026.
- Recruit 2026, Recruit Holdings FY2025 Annual Report.
- SEEK 2022, SEEK Limited FY2022 Annual Report.
- SEEK 2023, SEEK Limited FY2023 Annual Report.
- SEEK 2024, SEEK Limited FY2024 Annual Report.
- SEEK 2025a, SEEK Limited FY2025 Annual Report.
- SEEK 2025b, SEEK Limited 2025 investor day presentation.
- SEEK 2026a, SEEK Limited FY2026 Annual Report, 12 August 2026.
- SEEK 2026b, SEEK Limited FY2026 full-year results presentation, 12 August 2026.
- SEEK 2026c, SEEK Limited FY2026 half-year report, 17 February 2026.