The Asset-Light Arms Race: Mining Services Without the Mining Risk
Four ASX mining services companies earn 10-45% ROIC without commodity exposure. Our analysis finds only Lycopodium (LYL) trades below fair value at A$14.98 vs A$16.26. GNG and PRN are fairly priced, while IMD carries a 27% premium.
Australia's mining services sector contains a group of companies that earn exceptional returns on capital without deploying a single dollar into commodity exposure. They build, engineer, drill, and digitise the mining industry while leaving the resource price risk entirely with their clients. Four of them reported results through the February 2026 reporting season: GR Engineering Services (ASX: GNG), Lycopodium (ASX: LYL), Perenti (ASX: PRN), and Imdex (ASX: IMD). They illustrate four distinct layers of the asset-light spectrum -- and the market has priced three of the four with more optimism than our analysis supports.
The common thread is returns on invested capital (ROIC, a measure of how much profit a business generates relative to the capital it employs) that range from roughly 10% to over 40%. What separates these businesses from commodity miners is that their capital base is paper-thin -- intellectual property, engineering expertise, and long-term client relationships -- while the resource companies carry the exploration risk, development cost, and commodity price exposure. The question worth examining is not whether this is a good model (it is), but whether all four are priced to reflect it.
Our analysis suggests the answer is no. LYL trades at a modest discount to fair value, PRN and GNG are priced at or above our estimates, and IMD carries a 27% premium that requires its digital transformation thesis to succeed faster and more completely than the evidence so far supports.
Scorecard
| Company | Rating | Price | Fair Value | Gap | Credibility | ROIC | Key Metric |
|---|---|---|---|---|---|---|---|
| GNG | Fair | $4.65 | $3.68 | -21% | 7.6/10 | ~45% | 12% EBITDA margin, zero debt, $86.5m cash |
| LYL | Fair | $14.98 | $16.26 | +8.5% | 7.5/10 | ~32% | $415m backlog, 10% NPAT margin target |
| PRN | Fair | $2.43 | $2.22 | -9% | 7.9/10 | ~10.5% | 0.6x leverage, $5.8bn work-in-hand |
| IMD | Overvalued | $4.00 | $2.93 | -27% | 7.8/10 | ~13.5% (ex-goodwill) | 32% EBITDA margin, four simultaneous acquisitions |
Prices as at reporting season February 2026. Fair values are point estimates from our DCF pipeline. Gap = fair value vs price.
GR Engineering Services (GNG): The Highest-ROIC EPC Contractor in Australia
GR Engineering is the purest expression of the asset-light engineering model. The company builds mineral processing plants under fixed-price EPC (engineering, procurement and construction) contracts and operates gas and power infrastructure under long-term O&M retainers. It owns no mines, no processing plant, and no heavy equipment fleet. Its capital base at December 2025 was equity of $70m and lease liabilities of $7.4m, against $57.2m of EBITDA on $479m of revenue, producing a ROIC of roughly 45%. That figure reflects genuine competitive advantage built on repeat-client relationships and a study-to-EPC conversion pipeline with 31 active studies at the half-year. FY26 revenue guidance of $500-520m was reaffirmed with H2 expected to step up as construction phases ramp.
Our probability-weighted fair value is $3.68 per share. At $4.65, GNG is approximately 21% above our estimate, but the gap is almost entirely a rate debate: the market is pricing a through-the-cycle WACC of roughly 9.3%, while our locked 11.0% WACC reflects the current 10-year bond yield at 4.86% and the RBA's February 2026 hike to 3.85%. At 9.5% WACC, our model produces ~$4.48, within 4% of the market price. The counter-argument for holding is real: fully-franked dividends of 24 cents annualised (5.2% yield, 7.4% grossed up), a fortress balance sheet with $86.5m cash and zero debt, and contracted visibility of 2-3 years. For domestic investors who can use the franking credits, GNG is not expensive. For those anchoring to current rates, it is.
Lycopodium (LYL): Gold Processing Expertise with a Modest Valuation Edge
Lycopodium is the only company in this group where our analysis finds a positive gap to market. LYL is an EPCM (engineering, procurement and construction management, where the client funds procurement and LYL provides management expertise) specialist focused on gold processing plants globally. The key distinction from GNG is that EPCM places less balance sheet risk on the contractor, producing higher NPAT margins (12.4% in FY25, though management targets 10% through-cycle) at lower revenue scale ($340m FY25). ROIC is approximately 32% on book equity, supported by $70m net cash and minimal capex. The $1.3bn project pipeline with $415m of committed contracts provides 12-18 months of visibility, backed by gold prices at record levels above $2,900/oz driving project approvals at rates not seen since the 2019-2020 supercycle. FY26 guidance was revised to revenue of $370-410m and NPAT of $37-41m, a downgrade management attributed to project timing slippages rather than demand.
Our dynamic weighted fair value of $16.26 implies approximately 8.5% upside at $14.98, modest but the only unambiguous positive gap in this group. The primary risk is gold price mean reversion: a sustained correction below $2,200/oz would defer 3-5 pipeline projects and reduce FY27 revenue by an estimated 15-20%. The July 2025 acquisition of 60% of SAXUM (Latin American engineering, Americas segment revenue $26.4m at 27% margin in H1 FY26) adds geographic diversification but brings integration risk in volatile jurisdictions. Our bear scenario (25% probability, $10.08 per share) captures gold correction and SAXUM integration risk together.
Perenti (PRN): Scale and Discipline at a Fair Price
Perenti occupies a different part of the spectrum. It deploys a $1.2bn fleet of underground mining equipment and employs around 10,000 people across Australia, Africa, and increasingly North America. The model is not asset-light, but it has systematically converted scale into a business with 10.5% ROIC, $5.8bn of contracted work-in-hand, 0.6x net debt to EBITDA, and four consecutive years of guidance delivery, which is uncommon in a sector where cost overruns are routine. FY25 revenue reached a record $3.49bn with EBIT(A) (earnings before interest, tax and amortisation, adjusted for non-cash acquisition costs) of $333.5m at a 9.6% margin. In 1H26, AUD appreciation compressed USD-denominated revenue translation and guided FY26 EBIT(A) to $335-350m, though underlying Contract Mining margins held at 11.1%.
Our DCF fair value is $2.22 against a market price of $2.43, a -9% gap that reflects a conservatively locked 20% probability on a severe scenario rather than a strong sell signal. PRN is efficiently priced for mid-cycle conditions. The moat here is different from GNG and LYL: underground mining is difficult, dangerous, and relationship-intensive, as evidenced by Perenti's 85%+ contract renewal rate and ability to win long-life gold contracts (Goldrush, Obuasi) against larger global competitors. The investment debate is margin sustainability: if current 18-19% EBITDA margins are structural, fair value moves toward $2.75-$2.95; if they compress toward 17% as our model assumes, current pricing is approximately fair. The CEO transition from Mark Norwell to Pat Howard (effective February 2026) is the near-term uncertainty.
Imdex (IMD): The Technology Premium Debate
Imdex is the most structurally distinct business in the group. It provides an integrated suite of downhole sensors, drilling optimisation tools, and cloud-based geological data analytics to resource companies across 100+ countries. Revenue splits roughly 68% into rentals, services, and software (recurring) and 32% into sale of goods. The central analytical question is the EBITDA margin: IMD maintained 29% through the FY25 downcycle, then expanded to 32% in 1H FY26 as exploration budgets recovered. Our analysis assigned 52/48 probability to this being cyclical recovery versus structural transformation, a near-even split. Complicating the picture is a $160m+ M&A program across four simultaneous acquisitions (ESA, Datarock, ALT/MSI, Krux) to consolidate the digital orebody intelligence market before larger platforms can. The strategic logic is coherent; four simultaneous integrations under a new CFO is without precedent in the company's history.
At $4.00, IMD trades 27% above our dynamically-weighted fair value of $2.93. The gap is driven by our WACC of 10.7% (versus market-implied ~8.5%) and terminal EV/EBITDA of 10.5x (versus trading at ~14x), reflecting the unresolved margin question. Reported ROIC on goodwill-inclusive capital is approximately 7.5%, below the WACC and technically value-destroying; strip out $329m of goodwill and it rises to 13.5%, which is the core tension in any acquisition-heavy technology business. Our analysis suggests the business is attractive below $2.50 and expensive above $3.50. At $4.00, the market is pricing the successful version of a thesis that has not yet been confirmed.
Synthesis: What the ROIC Spread Tells You
The 35-percentage-point spread in ROIC across these four companies -- from GNG's 45% to PRN's 10.5% -- is not primarily a story about management quality. It reflects three structural differences: contract type, capital intensity, and operating model.
GNG and LYL sit at the top of the ROIC range because they earn fees for expertise with almost no capital deployed. GNG's equity base is $70m against $479m of revenue. LYL's is $134m against $340m. The fixed-asset base in both cases is measured in millions, not hundreds of millions. A project wins and ROIC spikes because the incremental capital required to deliver it is near-zero. The risk to the ROIC model is binary: a fixed-price EPC overrun (GNG's core risk) or a project pipeline that dries up when commodity prices fall (LYL's gold exposure). Neither is a sustained erosion of the business model -- they are episodic risks, which is why scenario analysis matters more than steady-state DCF for these two.
Perenti's 10.5% ROIC reflects the capital cost of deploying a $1.2bn fleet. The model earns a spread of roughly 100 basis points over its weighted average cost of capital -- thin, but positive and defensible given the $5.8bn work-in-hand, rise-and-fall cost provisions in contracts, and scale procurement advantages. The appropriate valuation multiple for a business earning a thin economic spread is lower than for one earning 45% ROIC -- which is exactly what the market reflects with PRN trading at 4.2x EV/EBITDA versus GNG at roughly 11x. The question is whether the 10.5% ROIC is the structural ceiling or whether North American expansion into premium-priced underground contracts can push it toward 12-13%.
Imdex occupies a different category. Its reported ROIC on goodwill-inclusive invested capital is approximately 7.5% -- below its WACC, meaning that as currently measured, the business is destroying value. Strip out the $329m of goodwill (accumulated from a decade of acquisitions) and ROIC rises to approximately 13.5%, above the 10.7% WACC. The resolution matters: if the goodwill is productive (the acquisitions built a platform that will generate above-cost returns for years), the true economic ROIC is 13%+. If goodwill impairment comes, the equity base shrinks and historical returns look better retrospectively, but the capital is gone. This is the core analytical tension in any technology-acquisition-heavy business.
Contract structure amplifies the differences in risk profile. GNG takes fixed-price EPC risk (cost overrun falls on GNG, cost saving goes to GNG). LYL takes EPCM risk (schedule delays affect fee conversion, but procurement overruns stay with the client). PRN earns cost-plus with rise-and-fall provisions (fuel and labour inflation is passed through, limiting margin volatility). IMD earns tool rental and software fees with near-zero marginal cost of delivery (revenue scales, cost base is largely fixed in the 1-3 year horizon). These structures create different EBITDA margin volatility profiles -- which is why a blanket EV/EBITDA comparison across all four misleads. GNG's 12% EBITDA margin is harder to earn and harder to lose than PRN's 19%. IMD's 32% has more upside leverage but also more downside exposure to exploration cycle turns.
Bottom Line
Mining services without mining risk is a genuinely attractive business model -- the ROIC data across GNG and LYL makes this plain. The complication is that the best-quality version of the model (GNG, 45% ROIC, zero debt, fully franked yield) is priced at a 21% premium to our current-rate fair value, because the market is pricing in rate normalisation that the RBA has not yet signalled. LYL offers the only unambiguous positive gap, with 8.5% upside supported by a gold pipeline at record prices and a management team with a 7.5/10 credibility score. PRN is fairly priced -- a sound business at a fair price, not a bargain. IMD carries the most structural uncertainty: the digital transformation story is compelling, the evidence is one half-year old, and four simultaneous acquisitions are a significant execution test. Our analysis suggests the business is good at any price below $2.50 and expensive above $3.50. At $4.00, the market is pricing the successful version of a thesis that has not yet been confirmed.
Analysis generated by the Alpha Insights AI research pipeline. All fair values are point estimates subject to model uncertainty. This is not financial advice. Do your own research before making investment decisions.