The platform
Battery energy storage, fully utilized.
One platform from the first utility bill to the signed credit paper. Tariffs, dispatch, revenue stacking, and project finance for distributed storage in North America.
Intake
Paperwork becomes structured input.
Utility bills, tariff filings, and meter files arrive as paperwork. They leave as structured inputs, every value traceable to its source page. AI does this intake work, and only this work. Then its job is done.
Nothing downstream reads the original document again. The engines read the structured record, and the record keeps a pointer back to the page it came from, so a reviewer can check any figure against the paperwork it was taken from.
Model
Tariffs read the way the utility bills them.
Tariff engines model demand charges, time-of-use windows, ratchets, riders, and export compensation as the filing writes them, not as a blended average. Change the tariff version and the bill impact changes with it.
Dispatch optimization then values the battery the way it will actually run: against the site load, the tariff it is billed on, and the markets it can reach. Degradation and augmentation sit in the math, not in a footnote.
Underwrite
Built for the credit process.
Pro formas, downside cases, and debt sizing are produced for the people who will question them. Sensitivities run on the same engines that produced the base case, so the downside is not a second spreadsheet carrying its own private assumptions.
Every figure traces to a source document, a tariff version, and a methodology version. That is what lets a case survive the committee's questions instead of restarting after them.
Operate
Actuals against the pro forma.
Once the asset is running, measured data goes back through the same engines that underwrote it. You see where the project stands against the case that financed it, stated in the same terms.
The boundary
Where AI stops.
Deterministic by design
No result comes from a language model. Every number comes from deterministic engines: same inputs, same outputs, every run. Versioned, explainable, and re-runnable, the way model-risk review expects.
AI-accelerated
Utility bills, tariff filings, and meter files arrive as paperwork. AI turns them into structured inputs, every value traceable to its source page. Then its job is done.
When the OCC, the Federal Reserve, and the FDIC revised their model risk management guidance in April 2026, generative AI stayed outside the framework banks use to validate the models they rely on. Deterministic engines are built for that framework, and the methodology behind them is published for anyone who has to check it.
Who it serves
The same engines, read three ways, depending on the decision in front of you.
For developers
Screen sites against real tariffs and size systems against real load. When diligence starts, walk in with a pro forma that holds up to the questions it is going to get.
For operators
Compare dispatch performance to plan and see where the two diverge. Revenue is verified against the tariff the site is billed on, not against a guess at what the month should have looked like.
For lenders
Downside stress, DSCR sensitivity, and a methodology your credit team can read before the first meeting. Nothing in the case rests on a number you have to take on faith.
Access
How access works.
Access is by invitation. Engagements are scoped to the decision at hand, whether that is a site screen, a financing case, or a performance review after the asset is running.
Outputs are delivered with a methodology version and an input manifest. What you take to a credit committee can be re-run and checked, by you or by whoever reviews your work.
Utilyzed fits if you
- are siting distributed storage and need to know what the tariff does to the case before capital is committed
- are running assets and need measured performance checked against the plan that financed them
- are sizing debt against a storage revenue stack and need the assumptions written down and re-runnable
Frequently asked
Questions
What is Utilyzed?
Utilyzed is an independent battery energy storage analytics platform for developers, operators, and lenders in North America. It covers behind-the-meter and distributed storage underwriting: tariff modeling, dispatch optimization, revenue stacking, and project finance pro formas.
How is Utilyzed different from Modo Energy?
Modo Energy focuses on wholesale, front-of-meter battery markets, with benchmark indices and forecasts. Utilyzed is built for behind-the-meter and distributed storage underwriting in North America: tariffs, dispatch optimization, revenue stacking, and pro formas. Utilyzed publishes its methodology publicly, holds no market position and administers no benchmark, and every output number comes from deterministic engines rather than generative AI.
Does Utilyzed use AI?
AI is used for document intake only: turning utility bills, tariff filings, and meter files into structured inputs traceable to their source pages. Every output number comes from deterministic engines. This matters under the April 2026 revised U.S. interagency guidance on model risk management, which keeps generative AI outside the framework banks use to validate the models they rely on.
Can lenders use Utilyzed outputs in a credit process?
Outputs are delivered with a methodology version and an input manifest, so results can be re-run and checked. The full methodology is public at utilyzed.com/methodology.
What does Utilyzed cover?
Retail tariffs, wholesale markets, feed-in programs, and incentives, modeled across North America.
How do I get access?
Access is by invitation. Request access at utilyzed.com/contact with a note about your project and the decision it needs to support. There is no self-serve signup; a person reads every request.
Access is by invitation
Tell us what you need to underwrite.
Send the project and the decision it has to support. A person reads every request.