Asset Performance & Optimization for Energy is a new decision solution from data², powered by the reView™ platform. It brings drilling, geology, production and business records into one connected view of every well. Teams can benchmark drilling and production, and see which results the data can actually support. It launches at ADIPEC 2026.

What did data² launch at ADIPEC 2026?

On November 2, 2026, data² launched Asset Performance & Optimization for Energy at ADIPEC 2026 in Abu Dhabi. It is a decision solution built on reView, the data² decision intelligence platform.

Every well carries a capital decision. Where should we drill next? Which rig and frac vendor should we hire? Which producing wells should we fix first? Each of these starts with one question: given a place in the earth, what do we know about it, and how much should we trust each fact?

Most data platforms can answer the first half. Very few can answer the second. This solution was built to answer both.

Why is well data so hard to trust today?

For most operators, the record is split. Daily drilling reports, rig time logs, cost estimates, service tickets, well logs and production history sit in separate systems. Someone joins them by hand for every study.

Trust does not travel with the data, either. A paper-era survey that was re-keyed by hand can land in the same column as a checked wireline reading. Once the values are combined, nobody can tell which one drove the number.

Adding AI on top makes it worse. An AI model answers just as fluently either way. Without a record of where each value came from, it cannot tell you what it does not know, or what you should go and measure.

"In energy, a decision about a well is only as good as your ability to defend it," said Jon Brewton, founder and CEO of data². "Operators already have the data. What they lack is a way to prove where each number came from and how far it can be trusted, to a partner, a board or a regulator. Asset Performance & Optimization for Energy puts that proof behind every answer, whether an engineer or an AI is asking."

What can asset teams do with it?

The solution gives teams four ways in, with one connected graph underneath.

• Benchmark drilling. Compare operators, rigs or pads on rate of penetration, drilling efficiency, wellbore curvature, cost per meter and non-productive time, inside any area drawn on the map. Minimum evidence rules decide what can be ranked, so every ranking rests on data that holds up.

• Benchmark production. Track each well against its type curve, compare expected recovery against completion design, rank frac vendors, and get a ranked list of the wells that need attention.

• See the earth. Explore a 3-D view of every well. Click any point along a wellbore to trace it to its formation, daily drilling report, rig, crew and invoice.

• Ask reView. Send any question, with its full context, to the reView Reasoner. Users see the prompt before it goes, and every answer comes back with the evidence behind it.

How does it decide which data to trust?

Every value carries a trust level. Six tiers run from measured and verified down to inferred. Each tier sets what a value may be used for, from reserves booking down to background context.

An unchecked value can never drive a ranking. When the evidence is too thin, the app will not rank it. Instead, it tells the team what to go and measure.

"Most AI tools will give you a number for any question you ask," said Jeff Dalgliesh, co-founder and Chief Research Officer of data². "We built this to refuse a ranking when the evidence is not there, and to tell the engineer exactly what to go and measure. That refusal is what makes every other answer worth trusting."

How detailed is the view of each well?

The solution models the rock along each wellbore in 30-centimeter cells. Each cell holds about 90 attributes across drilling mechanics, direction, mud, vibration, geology, lost time and cost.

Every cell links back to its well, the activity that drilled it, the rock formation and the budget behind it. That is evidence-level traceability: a single meter of hole can be traced to the rig that drilled it and the vendors who billed for it. This kind of cross-system reasoning used to take a manual study. Now it is a walk through the graph.

Does it work with the data standards operators already use?

Yes. Business records follow PPDM 3.9, the industry data model for wells. New operator data arrives in WITSML and PRODML, the standard formats for drilling and production data, with its chain of custody attached. Data comes in through a gateway that checks it and issues a receipt, and staff review it before anything is loaded.

On the read side, the user app is read-only. Writes and reads are kept apart by design.

How do we know the AI answers are right?

data² does not take the AI's answers on trust. It tests them. Engineers write and confirm questions with known answers for six roles, from executive to well engineer. The same questions run three ways: reView with the graph and its domain pack, a strong AI model on the raw graph, and the same AI on the data as database tables. One independent judge scores every answer, and each score breaks into checks you can open. The gap between the scores shows what the graph and the domain pack add.

Who is it built for?

The solution serves everyone who makes or reviews well decisions: executives setting capital plans, asset managers, and drilling, completions and production engineers. Because every number carries its trust level and its source, a figure that reaches a board deck or a regulator can be traced back to where it came from. That explainability is the core of every reView Decision Record™: the evidence, the governing rule, any conflicts, a confidence level, and a link back to the source.

Where can I see it at ADIPEC 2026?

ADIPEC 2026 runs November 2 to 5 at ADNEC Centre Abu Dhabi. data² is part of the Al Masaood exhibit in Hall 3, Stands 3430 and 3530. Visitors can see live demos of drilling benchmarks, production benchmarks, the 3-D well view and Ask reView. The solution is available now.

Frequently Asked Questions

What is Asset Performance & Optimization for Energy? It is a decision solution from data², powered by reView. It joins drilling, geology, production and business records into one view of every well, and shows how far each result can be trusted.

How is it different from other well analytics tools? Every value carries one of six trust tiers. The app refuses to rank results when the evidence is too thin and tells the team what to go and measure.

What data standards does it support? PPDM 3.9 for business records, and WITSML and PRODML for incoming drilling and production data, with chain of custody attached.

Can AI answers be traced to their source? Yes. In Ask reView, users see the prompt before it goes, and every answer comes back with the evidence behind it.

Who uses it? Executives, asset managers, and drilling, completions and production engineers who make or review well decisions.

Where can I see a demo? At ADIPEC 2026, November 2 to 5, Al Masaood exhibit, Hall 3, Stands 3430 and 3530, ADNEC Abu Dhabi. Or request a demo at data2.ai.

See It Live

Visit data² at the Al Masaood exhibit, Hall 3, Stands 3430 and 3530, during ADIPEC 2026. Can't make it to Abu Dhabi? Book a demo and see decisions you can defend.