Using AI as an independent baseline to validate production uplift
Last updated: July 2026
When a new chemical, proppant, frac design, drilling method, or spacing pattern is introduced, the practical question is simple: did it produce a measurable effect?
Most teams still answer that question with internal results or with decline-curve forecasts built after the fact. Both can be influenced by expectations or by how the curves are drawn. In practice, service companies often end up building custom Excel models, while operator teams typically do the work in Spotfire. These are internal constructs that reflect the assumptions, filters, and curve-fitting choices of the people running the work, so the result is often less objective than the data itself would support.
A cleaner approach exists
AlphaX Sky first establishes an independent AI-based expectation for well performance using historical production and completion characteristics across the basin. That expectation becomes the baseline — a data-driven reference point generated from patterns already present in the historical record.
Wells that received the new treatment are then grouped by treatment type. Their actual monthly production is compared directly against the baseline. The difference is the measured effect.
How the baseline workflow works
- Build the baseline from basin-scale historical production and completion patterns.
- Group treated wells by chemical, proppant, frac design, or spacing pattern.
- Compare actual monthly output against the independent forecast baseline.
This sequence matters. Because the expectation is set from basin-scale historical relationships rather than from the trial results themselves, the comparison sits outside the team running the test. It does not rely on internal type curves or forecasts that were shaped after production arrived. The measurement rests on observed monthly data versus a pre-existing, independent reference.
Why this is more defensible
Operators can evaluate their own field trials with less internal influence. Service and chemical companies can use the same approach to present results that rest on an external baseline instead of solely on their own projections. In both cases the goal is the same: make the observed effect clearer and easier to defend.
The method does not change the underlying models. It simply supplies a consistent reference against which actual performance can be measured.
To see how the baseline approach works on a specific set of wells, run a deal or contact us.