Your Customers Expect Fixed Prices. The Market Doesn’t Care.
Predict+ Protects the Margin In Between.
Razor-Thin Margins in a Market That Won’t Stay Still
Energy retailers live in the gap between the fixed prices they promise customers and the volatile wholesale prices they pay for supply. Every percentage point of load forecast error translates directly to margin exposure. When real-time prices spike to $4,000/MWh during a single afternoon — as they did in ERCOT in April 2025 — the retailers with accurate forecasts are pre-positioned. The rest are scrambling.
The challenge is compounding. Renewable intermittency is reshaping net demand curves. Behind-the-meter DER adoption is making customer load profiles less predictable. And the correlation between a customer’s peak usage and wholesale price peaks means that under-forecasting load during high-price hours is the most expensive mistake a retailer can make. Legacy regression models trained on yesterday’s load shapes can’t keep up with today’s market.
ERCOT real-time price spikes (April 2025)
More accurate than legacy spreadsheet methods
Forecast accuracy
Predict+ Solutions for Your Retail Operations
Predict+ delivers 97.5% forecast accuracy across diverse customer portfolios — and has maintained that performance through the market conditions that destroy retail margins. COVID-19 demand shocks. Extreme weather price spikes. Sudden shifts in customer behavior. Our adaptive AI models automatically recalibrate to regime changes without manual intervention, so your portfolio forecasts stay accurate when the market is most volatile.
That accuracy is built on a meter-level foundation — AI models trained on real consumption data from 140K+ endpoints, augmented with weather ensembles, economic indicators, and behavioral signals. The result: load forecasts that understand your customers’ actual patterns, not just weather-adjusted historical averages.
For retailers serving customers with on-site solar, Predict+ can also incorporate device-level generation data from Tigo MLPE, improving the net load accuracy for those customer segments. But the core value — adaptive, crisis-tested demand forecasting — delivers margin protection regardless of your customers’ DER mix.
.svg){kind=avatar}
How Predict+ Powers Retail Energy Operations
From product pricing to portfolio risk management — every capability is built to protect the margin between what you promise customers and what the market charges you.
Protect Margins with Accurate Load Forecasting
Portfolio-level load forecasts that reduce unhedged market exposure.
Price Products with Confidence
Forecasting backbone for profitable product pricing across rate structures.
Optimize Hedging & Procurement Strategy
Shape-aware hedging aligned to hourly customer load profiles.
Manage Customer Portfolio Risk
Segment portfolio risk by forecast confidence and load volatility.
Win and Retain Customers with Green Products
Generation forecasting for green tariff pricing and PPA verification.
Reduce Imbalance & Settlement Costs
Adaptive hourly forecasts that minimize real-time settlement exposure.
Enterprise-grade security
Your forecasting data is protected by enterprise-grade security infrastructure. Predict+ is built on a zero-trust architecture with end-to-end encryption, role-based access controls, and continuous monitoring — so your energy data stays yours.
Certified information security management
Audit in progress — on track for 2026
How we work together
Rapid validation and implementation so you are up and running as quickly and seamlessly as possible.
Discovery
Align on use cases, forecast horizons, stakeholders, and what success means for your organization.
Data
Connect sources — smart meters, SCADA, weather APIs, market feeds — validate quality, and establish repeatable pipelines.
Backtest
Run parallel validation against your historical data to demonstrate accuracy and fit before expanding scope.
Go-Live
Operational rollout with customized models tuned to your portfolio, with continuous monitoring from day one.
Optimization
Ongoing accuracy tuning for seasonal patterns, market changes, and evolving load profiles. Regular reviews to expand use cases.