ActionWell ActionWell
Assess my portfolio
Assess my portfolio

Engineering

Engineering judgment that carries into the next well

Inspect the evidence behind an analysis. Correct the assumptions that matter. Keep that judgment available where the engineering conditions match.

A reason to trust the next decision

Engineering requires more than a plausible answer

The source record, operating context and method must support the conclusion. ActionWell makes that reasoning inspectable so your engineers can challenge it and decide what comes next.

Evidence

The right history belongs to the right well

Check identities, units and operating periods. Preserve conflicting records and identify which calculation they affect. A missing test or work record becomes a specific information request.

Cutaway of an oil well from the surface wellhead and gathering line down through the wellbore into layered rock and the producing formation.

Engineering reasoning

Assess the explanations that fit the evidence

A production change can reflect operating work, equipment behavior, reservoir response or reporting. Compare the alternatives and identify the observation that would distinguish them.

Methods and economics

Use the method the decision needs

Choose calculations and forecasts for the operating conditions they describe. Keep inputs and assumptions inspectable. Use the operator’s response estimates, ownership, costs and constraints when evaluating an action.

Engineering knowledge

Keep the correction and the reason behind it

The recorded workflow shows an engineer ending one window at a documented production change and keeping the reason with the correction. The same reasoning reaches similar wells, each at its own change, while a reviewed result stays protected.

Recorded product workflow · Demo engineer Watch the 90-second workflow

Decision outcomes

What happened should inform the next case

The target workflow connects the expected response with executed work, actual costs and subsequent production. Preserve the original expectation so the team can examine the difference.

Target workflow · Schematic
Outcome measured against a frozen baseline Measured production declines up to a marker labeled work carried out. After the marker, a dashed baseline frozen before the work continues the decline inside a band of normal variation, and measured production runs above it. A bracket marks the difference measured on the same basis. Work carried out Measured production Baseline frozen before the work Normal variation Measured history

The bracket is the difference, measured on the same basis. No values are shown.

Bring your engineering question

Use an assessment to examine one asset with your team and establish what would change the decision.

Assess my portfolio
Keep the reason, improve the next review · 1:30Recorded product workflow · Demo engineer
Transcript: on-screen findings and narration
  1. Production columns: reported oil, Jul 2025–Jun 2026; color shows the change vs the prior year (within ±2% neutral). Recorded example · well C-1 · oil: automatic analysis, not reviewed. A production history can contain several distinct intervals. Each needs to be understood in its own context.
  2. Well C-1 · reported oil per calendar day, every month (bbl per calendar day; full history 2014–2026). Automatic intervals: 2015, 2016–17, 2018–25. Selected window: Jan 2018–Aug 2025. Data · public monthly reports. ActionWell identifies candidate intervals automatically. The engineer can inspect the evidence and the analysis within each one.
  3. What is wrong with this window: one curve spans 88 months; production behavior changes inside the window. Evidence · change screen: from May 2019, a sustained rise to a new level, ×1.82 against Sep 2018–Apr 2019. Here, one window includes observations that follow a different production trend.
  4. Separate the later production interval. The engineer’s correction (demo profile, fictional): “End this window at the production change. The later observations follow a different trend and should be analyzed separately.” The observations after the change are ringed; the window’s right edge moves to the change (corrected window). The engineer explains the correction: end this window at the change and analyze the later interval separately.
  5. Recalculated: window Jan 2018–Apr 2019 · 16 observations · a descriptive curve, not a forecast. The other intervals stay visible. The boundary moves. The selected analysis updates, while the full history and the other intervals remain visible.
  6. Saved with the correction · read back from the record: the reason (the engineer’s words, kept verbatim); the evidence (a sustained change of level at May 2019, from the window’s own change screen); what changed (window revision 1: the end moves from Aug 2025 to Apr 2019). Demo profile, fictional · local, unauthenticated. ActionWell keeps the correction together with its reason and supporting evidence. The next review can trace exactly what changed and why.
  7. Another well, C-11 · the same reasoning, its own evidence: a sustained rise from Dec 2023 (×2.22); its window now ends Nov 2023, before its own change (provisional, awaiting review). Protected · reviewed well C-2: its reviewed result is kept; an impact notice asks a person to compare. On another relevant well, that reasoning guides the review. The boundary comes from that well’s evidence. Reviewed results stay protected.
  8. Where the reasoning applies (this recorded example): 12 wells checked · updated: 7 wells, each at its own change · protected: 1 reviewed result kept, with a notice · not applicable: 4 wells without such a change. Saved as a provisional rule · engineering endorsement: none. This is how engineering judgment becomes useful across a portfolio: the team keeps the reasoning and applies it where it fits.
  9. ActionWell. Keep the reason. Improve the next review. Assess my portfolio. actionwell.com ActionWell. Keep the reason. Improve the next review.