Forecasting and Decision Science
Predicting what’s likely.
Preparing for what’s possible.
User Metrics builds probabilistic forecasts and scenario models that help organizations plan for demand, risk and change.
Independent forecasts and evidence for the decisions that shape energy policy.
Load growth, electrification and new technology are driving billions in grid spending. We help utilities, agencies and policy makers see what's likely, test what they're being told, and decide with confidence.
Scenario modeling · Independent forecast review · Program evaluation · Target tracking
Industries we serve
Energy and Utilities
Load, renewable generation and electrification forecasts for planning and operations.
Government and Policy
Scenario modeling, program evaluation and tracking progress toward climate targets.
Retail and Consumer
Demand, sales and inventory forecasts at the store and product level.
Supply Chain and Operations
Demand planning and capacity forecasts that hold up when conditions change.
Built for decisions, not dashboards.
Single numbers are a liability.
A point forecast hides the risk you most need to see. We forecast the full range of outcomes and the probability of each, so you plan for what could happen, not just what probably will.
Messy data isn't an excuse.
Short histories, structural breaks and shifting policy break most models. Bayesian methods are built for exactly these conditions. They combine the data you have with what is already known, and they tell you how confident to be.
If you can't explain it, you can't use it.
Black-box models don't survive a board meeting, a regulator or a rate case. Every model we deliver shows its assumptions, drivers and track record in plain language.
Who we are.
Senior expertise from start to finish.
User Metrics is led by Ryan Clukey, a data scientist and product leader with more than 25 years of experience turning data into decisions.
Ryan led machine learning product work at Kinaxis, where he worked on a demand forecasting platform used by global supply chains, and at Dell Technologies, where he led machine learning and large language model initiatives. Earlier in his career, he ran quantitative research for Intel, Microsoft, T-Mobile and Nokia, studying how people actually use information to make decisions.
That combination is rare in forecasting: rigorous statistical modeling, product experience shipping tools people rely on, and a background in experimental psychology. The result is forecasts that are accurate, and also understood and used.
Every engagement is led directly by Ryan.
Based in San Diego. Working with clients from around the world.
Let’s go!
Tell us what you need and we’ll figure out how to get it done.