capability 01 · outcomes engineer

Opportunity Discovery

The most common way to waste an AI budget is to build something impressive that nobody needed. Opportunity discovery is the skill that prevents it.

Removes: understand customers · align to customer needs

By · V-Squared AI

The most common way to waste an AI budget is to build something impressive that nobody needed. Opportunity discovery is the skill that prevents it: finding gaps where customers can't get what they need, sizing the value trapped there, and defining it before anyone builds.

Point capability at trapped value

It's how you point capability at trapped value instead of pointing technology at whatever's convenient. The core of it is rigorous customer understanding: how customers decide what they need, and what they'll actually buy to meet those needs.

Get the model right and everything aligns

Get that model right, and everything downstream aligns to it. Get it wrong, and no amount of engineering saves the roadmap. This is the capability that removes the understand-the-customer and align-to-customer-needs bottlenecks, and it's usually where an outcomes engineer starts.

Point capability at trapped value, not technology at whatever's convenient.
learn it · highroiai.com

AI Opportunity Discovery

Build this capability through the AI Opportunity Discovery course on High ROI AI. Often employer-reimbursable, and available self-paced or instructor-led.

Explore the course
the market is already hiring for this

Forward Deployed Engineer

This capability removes (G4,F1)/(G4,F3), the customer modeling and alignment cells an FDE embeds to remove. When AI commoditizes the technical work, that removal is the job, not the tooling.

See the roles
the evidence · earnings calls, july 2026

The vendors are selling opportunity discovery as a service now.

This capability says the most common way to waste an AI budget is building something impressive nobody needed. Four earnings calls in July 2026 described the same fix, and in every case it starts before anything gets built.

peer-reviewedsurvey researchcompany filingsmarket dataearnings call
Microsoftearnings callFY26 Q4 · Jul 29, 2026

The engagement starts with co-design, not deployment

Frontier Company embeds 6,000 experts with customers to co-design and continuously improve AI systems, after a year of piloting across 330 projects with 164 customers. That is roughly two projects per customer rather than one large program.

Source: Microsoft FY26 Q4 earnings call, July 29, 2026.
what it confirms

Two projects per customer is a discovery pattern, not a delivery pattern. You find the trapped value, size it, prove it, and go again. The vendor built its largest new organization around doing that inside the customer's business.

Metaearnings callQ2 2026 · Jul 29, 2026

A well chosen opportunity, with the conversion number attached

Movida, a Brazilian rental car company with roughly 400 locations, put a business agent on WhatsApp to run selection, pricing, and payment. Daily bookings rose 44% year over year in a single month, with 85% of conversations resolved without a human. More than a million businesses now use Meta Business Agents weekly.

Source: Meta Q2 2026 earnings call, July 29, 2026.
what it confirms

The booking flow was the right target because that is where customers could not get what they needed. Same technology pointed at an internal reporting workflow produces a demo. The choice of where to point it is the capability.

Alphabetearnings callQ2 2026 · Jul 2026

The moat is the customer's problem, not the model

Asked what the moat is when every competitor has comparable models, Pichai said the model is just an ingredient in the solution, and that what customers need is their own data and trajectories kept confidential with nothing flowing back.

Source: Alphabet Q2 2026 earnings call, July 2026.
what it confirms

When the model is an ingredient, the differentiator is knowing which problem to point it at and what the customer will actually pay to have solved. That is opportunity discovery described by a vendor who no longer expects to win on capability.

Sources: Alphabet Q2 2026, Microsoft FY26 Q4, Meta Q2 2026, and Amazon Q2 2026 earnings calls and releases, July 2026. Figures are as stated by company executives. Amazon reported after market close on July 30, so Amazon figures come from the release and initial call remarks rather than a full transcript.

the wider evidence · what actually kills AI budgets

The failure is upstream of the build, and it has been measured repeatedly.

If AI projects failed on engineering, better engineers would fix them. Four independent datasets point somewhere else: the problem was chosen badly, defined vaguely, or never connected to something a customer wanted.

peer-reviewedsurvey researchcompany filingsmarket dataearnings call
S&P Global Market Intelligencesurvey research2025 · 1,000+ enterprises

Abandonment more than doubled in a single year

The 2025 Voice of the Enterprise survey of more than 1,000 organizations across North America and Europe found 42% of companies abandoned most of their AI initiatives, up from 17% the previous year. The average organization scrapped 46% of its AI proofs of concept before they reached production.

Source: S&P Global Market Intelligence, Voice of the Enterprise: AI & Machine Learning 2025, as reported by CIO Dive and Fortune.
what it confirms

These are funded programs with staffed teams and executive sponsors. They didn't fail because the technology stopped working, and the doubling happened in the year the models got dramatically better. Something upstream of the model is selecting the wrong work.

RAND Corporationsurvey research2,400+ AI initiatives

Problem definition ranks above every technical cause

RAND's analysis of enterprise AI initiatives found more than 80% failing to deliver intended business value, roughly twice the failure rate of comparable IT projects without AI. The root causes identified are led by misunderstood problem definition and a technology-first mentality, ahead of data and infrastructure issues.

Source: RAND Corporation analysis of enterprise AI project outcomes.
what it confirms

Misunderstood problem definition is not a euphemism for weak engineering. It means the team built the thing they were asked for and the thing they were asked for was not the trapped value. That is the specific failure this capability removes.

Gartnersurvey research2024 to 2026 forecasts

The named causes are commercial, not technical

Gartner forecast that at least 30% of generative AI projects would be abandoned after proof of concept by the end of 2025, naming poor data quality, inadequate risk controls, escalating costs, and unclear business value. The firm has separately forecast that 60% of AI projects unsupported by AI-ready data will be abandoned.

Source: Gartner research on generative AI project outcomes, 2024 to 2026.
what it confirms

Unclear business value sits in the same list as data quality and cost. It is the only one on that list you can eliminate before a single engineer is assigned, and it is the one that makes the other three worth solving.

Metaearnings callQ2 2026 · Jul 29, 2026

What the successful version looks like from the outside

Meta reported the Movida agent lifting daily bookings 44% year over year in one month, and said it expects to evolve business agent products toward its advertising model, where businesses only pay when Meta delivers results for them.

Source: Meta Q2 2026 earnings call, July 29, 2026.
what it confirms

Pay-on-results only works if you can identify, in advance, where the value is and what it is worth. A vendor willing to underwrite the outcome is a vendor that has done opportunity discovery properly. That is the standard this capability sets.

The RAND and Gartner figures are widely cited and directionally consistent with each other and with the S&P Global survey. Failure rate definitions differ across the three, so the agreement on causes matters more than the exact percentages.