Actionable insights — Hastings on D3 Energy's drilling program

1. The wellhead-payout matrix — price × content × volume, minus wellhead OPEX, divided into well cost

1. Build a single-well revenue line from the components you can actually observe: price per mcf for each saleable component (here helium and methane), the content of the stream (impurities %, helium %, methane %), and the volume in MCFD. Multiply out to annual revenue.
2. Subtract a wellhead OPEX estimate only — deliberately excluding gathering, plant processing and other facility costs, "as those are not associated with the wellhead effort." That isolates the drill-another-one decision from the midstream build.
3. Haircut the resulting operating cash flow by the program's success rate (Hastings uses 83%), so the number reflects a drilling campaign, not a single lucky hole.
4. Divide the well cost by the adjusted monthly cash flow to get payout in months — the one figure that makes cheap-well plays comparable to conventional ones.
5. Build it with a scenario block at the bottom: your "most likely" row, and an empty row underneath for the reader to drop in their most-likely variables. A model someone else can re-run is a model that gets stress-tested.

• Whether a quoted payout is wellhead-only or full-cycle — a one-month wellhead payout says nothing about the facility capital still to be spent.
• The assumed success rate and realized price per mcf — the two inputs that move payout most.
• Actual flow rates (MCFD) from the production tests, to replace the assumed volume with a measured one.

2. When wells are cheap, dry holes are an information budget — not a loss

1. Compare the cost of one well against the value of the product and against the company's treasury. If a failure is immaterial at the corporate level, stop scoring wells as hit-or-miss.
2. Re-frame the failed well as a purchase of information: it calibrates the seismic, constrains the fault model, and adds to reserve definition — "the much greater good."
3. Judge the operator on whether they are using that latitude — drilling to learn — rather than on a headline success rate.
4. Sanity-check your scale intuition against a real high-cost analogue so you don't apply offshore risk discipline to a $200k onshore well.

• A well cost that is a rounding error against the company's cash and against per-well revenue — the precondition for this logic.
• An operator drilling additional holes into uncertainty rather than only into de-risked locations.
• The counter-case: if well costs creep up materially, the whole "dry holes don't matter" argument stops holding.

3. Seismic-after-wells is a valid sequence when you have well control — check the precondition, not the convention

1. When an operator drills before shooting seismic, ask the precondition question: is there an existing producing well nearby — ideally between the new locations? That is what makes the environment genuinely low-risk.
2. Understand what the post-well survey buys: shoot the seismic after, then match the seismic signature to each well's known producing zone(s) — you get a calibrated template instead of a guess.
3. Use that calibrated template to step out further, where there are no well corollaries, at lower risk. The riskier the next location, the more the earlier calibration is worth.
4. Distinguish a cost-saving shortcut from a sequencing choice. If the survey is being shot anyway, and shortly, it wasn't skipped — it was ordered second on purpose.

• An existing producer inside or adjacent to the new drilling pattern (the precondition).
• A funded, scheduled survey following close behind the wells — versus seismic that keeps being deferred.
• The next locations: are they step-outs beyond well control (where the calibration pays), or infill where it doesn't matter?

4. Read the seismic shoot map as a forward map of the development

1. When a survey is announced, chase down where it will be shot — the footprint, not just the fact of it. Survey money goes where drilling money is intended to follow.
2. Overlay that footprint on the permit map and on the block the company most needs to prove up (here, the area under production-right application).
3. Rank the next wells above the current ones in importance when they will have both well corollary and seismic behind them — that combination is where the risk reduction actually lands.

• Publication of the 2D seismic layout at Nooitgedacht — specifically whether the lines run south toward the PR016 application area.
• Two further wells this year, positioned toward the PRA.
• Any survey footprint that points somewhere unexpected — that is a change in plan worth asking about.

5. Underwrite the reclassification event — Contingent → Proven/Probable/Possible via the independent certifier

1. Identify who the independent certifier is (here Sproule) and what inputs they need before they can re-book volumes — new well results plus the new seismic, together.
2. Treat the reclassification, not the drill-bit result, as the value event: contingent resources are volumes believed present but not commercially demonstrated; proven/probable/possible reserves are what underpins a production right, financing and a valuation.
3. Sequence your expectations accordingly: wells → tests → seismic → certifier's interpretation → re-booked reserves → the permitting/valuation consequence. Don't price the re-rating before the certifier has the data.

• A Sproule (or equivalent) update following the NGT245 E/D tests and the 2D survey.
• The size of any Contingent→2P/3P migration, and whether it extends the reserve base north of the existing PRA.
• Follow-on Production Right applications, which the company has said the test results will inform.

6. Read operator intent from well placement — geometry is a disclosure

1. When an experienced operator does something geometrically odd — e.g. two wells spaced closer together than reservoir drainage would justify — assume intent and ask what the layout would make sense for.
2. Cross-reference against anything the company has said quietly or in passing (a "second target," a new zone). A layout that "perfectly ties in" with an offhand disclosure is corroboration.
3. In fractured geology, consider that the second well may be testing a different depth (a deeper or shallower zone) rather than a different area — or that the operator has mapped a structural high at that spot.
4. Weight the inference by the operator's track record: from a good, experienced operator, an unexpected choice is a signal; from a poor one, it's noise.

• Disclosure of a second (deeper or shallower) producing zone at Nooitgedacht.
• Well spacing and orientation in the next pair of wells — whether the tight-twin pattern repeats or the operator reverts to wide step-outs.
• Any language about a structural high / "geologic peak" at the NGT245 location.

7. Use pressure across wells to map fault-system connectivity — and watch helium %, pressure, flow

1. In fracture-hosted gas, treat the fault system as the reservoir: the question is not how porous the rock is but whether the cracks are connected.
2. Test it with pressure: measure it well by well, then compare across the growing well set to see whether the systems are connected or separate — both laterally and by depth.
3. Accept that this needs a population of wells, not one — which is why cheap wells and connectivity mapping reinforce each other, and why the next pair of wells is drilled on the results of the current pair.
4. Track a short, fixed signal list per well so results stay comparable: helium content (is it >4%?), pressure, flow — and let the independent certifier make the formal assessment.

• Helium concentration above 4% in the NGT245 E and D results (the permit has shown up to 8%).
• Reported reservoir pressure per well and how it compares across the pair — the connectivity read.
• Flow rates from the production tests once intermediate casing is set and NGT245 E is re-entered and tested.
• Whether well cost stays near the ~A$200k budget as the program extends — the assumption every one of these methods leans on.