Where the margin actually sits
Nine energy profit pools, six of them created by AI forecasting
Traditional energy thinking counts three profit pools: generation spread, retail margin, capacity payments. AI forecasting creates six more: congestion rent capture, ancillary dispatch, demand response timing, battery arbitrage, behind-meter optimization, and data-driven residual instruments.
60% of energy profit pools did not exist before real-time forecasting. The old hedging playbook captures four of nine.
Reading the pool
Each rectangle represents one of nine profit pools in the US power sector. Width shows revenue share — how much of the $500B market flows through that activity. Height shows operating margin — the AI-addressable fraction within 3-5 years. Area is the prize: total addressable profit. The largest rectangle is not always the richest. Congestion rents are 4% of revenue but 35% margin. Generation dispatch is 22% of revenue but only 8% margin. The shape of the pool tells you where intelligence creates value versus where scale alone wins.
Read width for volume. Read height for margin. Read area for the actual prize.
Where energy margin concentrates
Revenue share and operating margin across the nine profit pools that make up the $500B US power sector.
Where margin concentrates
Three of nine pools — congestion rents (35%), demand response (30%), and mining energy (25%) — carry margins above 25%. These are the fastest-responding activities where AI timing creates direct alpha. The remaining six pools sit below 20% margin but carry larger revenue bases. Generation dispatch alone is $110B in volume at 8% margin — $8.8B addressable. The power market rewards two distinct strategies: high-margin/low-volume timing plays, and low-margin/high-volume efficiency compression.
$42B in total addressable margin. $18.5B sits in the three highest-margin pools.
Projected margins after AI optimization (3-5 year horizon)
Same nine pools, projected forward. Operating margins shift as AI forecasting, automation, and dispatch intelligence mature. Total addressable margin grows from $42B to $61B as efficiency gains compound.
Three wedges, one evidence layer
Attribution drives trading margins through ERCOT-grade forecasting at 3.21% MAPE. Billing automation through voice AI compresses operational cost from invoice generation to customer support. Residual curves convert battery, solar, and rig telemetry into financial instruments that reflect actual asset condition, not book depreciation.
ERCOT-grade forecasting at 3.21% MAPE drives trading margins through precise load prediction and price arbitrage timing.
Voice AI compresses operational cost from invoice generation through dispute resolution to churn prediction.
Battery, solar, and rig telemetry converted into financial instruments reflecting actual asset condition.
Together, the three wedges address $42B in margin across nine pools. The old hedging playbook reaches four of nine.
Explore each profit pool
Generation dispatch and fuel optimization→
Day-ahead and real-time market bidding, heat rate optimization, fuel procurement timing. AI forecasting compresses fuel
Congestion rents and CRR bidding→
Transmission constraint exploitation, congestion revenue rights, FTR portfolios. AI pattern recognition captures rents t
Ancillary services and frequency regulation→
Frequency regulation, spinning reserves, responsive reserve service. AI-dispatched BESS and DR assets capture premium cl
Retail billing and customer operations→
Customer acquisition, billing operations, churn management, call center operations. voice AI voice AI compresses operational
Behind-the-meter optimization→
Solar self-consumption, demand charge avoidance, battery scheduling for C&I and residential. AI sizing and scheduling ma
Demand response and load curtailment→
Grid-level and industrial demand response payments, ERCOT 4CP management, emergency load shedding. AI timing captures pa
Battery arbitrage and revenue stacking→
Charge/discharge spread capture across day-ahead and real-time markets. AI dispatch with degradation-aware scheduling be
Data center power and cooling efficiency→
PUE optimization, cooling energy reduction, power density management. AI load profiling and predictive thermal managemen
Mining energy optimization and curtailment→
Hash rate management, curtailment timing, power procurement optimization for crypto mining. AI reforecasting every 5 min
AI shift timeline
Watch the margins move in real time
The profit pool is not static. See how AI forecasting, billing automation, and residual instruments reshape each activity over the next 5 years.
View the timelineSee where your margin sits
The profit pool map shows which activities AI compresses and which it creates. The question is which pools your operations currently capture.
Map your profit pools
Explore the cluster
More on power and utilities
Cluster overview→
Nine profit pools, three structural transitions, and the AI activities reshaping US energy.
AI thesis→
The investment thesis for AI in power and utilities — where capital should flow and why.
AI shift timeline→
How each profit pool activity transforms over the next five years as AI adoption accelerates.
power and utilities AI profit pool: common questions
How are the nine profit pools defined?
Each pool represents a distinct margin source in the energy value chain: generation dispatch, congestion rents, ancillary services, retail operations, behind-the-meter savings, demand response payments, battery arbitrage, data center efficiency, and mining energy optimization. Revenue shares are fractions of the $500B US power sector.
What does AI-addressable margin mean?
The fraction of each pool's revenue that AI forecasting, dispatch, or automation can either compress (reduce cost) or create (unlock new revenue). A 0.08 operating margin means 8% of that pool's revenue is capturable through AI optimization within 3-5 years.
Why do 60% of pools not exist without AI?
Congestion rent capture, battery arbitrage, automated demand response timing, and real-time behind-the-meter optimization require forecasting accuracy and dispatch speed that manual operations cannot achieve. These pools exist only when the operating system is fast enough to capture them.
Which pools have the highest near-term impact?
Generation dispatch optimization and demand response timing show the fastest ROI because they build on existing metering infrastructure. Battery arbitrage and congestion rents have higher margins but require phase 2 forecasting accuracy to capture.