Built for ESCOs & energy services contractors
Pursue every opportunity. With the bench you have.
You are not short of buildings. You are short of the engineering weeks it takes to turn each one into a savings number a customer will sign against. Energy Pilot compresses that work from eight weeks to three days — running your calculations, not ours.
Proposals per engineer, per quarter. The analysis cycle drops from eight weeks to three days.
The economics
Three of every four engineering weeks produce no revenue.
Performance contracts need defensible savings before anyone signs. That means a site visit, an equipment inventory, utility data collection, a model, and an engineer's week — spent before you know whether the project proceeds.
At a 25–35% win rate, most of that investment returns nothing. And the opportunities you can't justify staffing never enter the pipeline at all, which is the part that doesn't show up in any report.
The instinct is to be more selective. But selectivity is just rationing a constrained resource — it caps your revenue at whatever your engineering bench can process this quarter.
The alternative is to make the analysis itself cheap enough that pursuing an opportunity is no longer a staffing decision.
What changes
Same bench. More pipeline.
Illustrative figures based on industry benchmarks and pilot data.
Built around your practice
Your ECMs. Your calculations. Your reports.
Your engineers have spent years building savings calculations they will defend in front of a customer, and a report format that customer already recognizes. Replacing them is not an upgrade. We load them in instead.
The asset layer
The site walk produces data, not a notebook.
Most of the three-to-five days between a walkthrough and a usable equipment list is transcription. Photograph the nameplate instead: make, model and serial are read on the spot, then matched against our equipment database to fill in the specifications your engineer would otherwise chase down from a manufacturer's site.
Depth on demand
Screening rigor and contract rigor, same building record.
An early number and a number that survives a performance contract are the same record at different depths. Add data, the range narrows — and no engineer rebuilds anything when the deal gets serious.
Open the conversation
Bills in, benchmark and bill analysis reports out. Something credible in the customer's hands before you spend a site visit on them.
Meter · platform fee
Capture the building
Capture walk, equipment inventory, telemetry connection. One visit produces the structured record everything downstream runs on.
Meter · per building assessed
Reach a defensible number
Your equipment ECMs and auto-calibrated energy models, with an 80% confidence range on every measure in the scenario.
Meter · per building engineered
Hold the guarantee
The calibrated baseline keeps running against incoming meter data. IPMVP-aligned reporting, with drift on guaranteed savings surfaced early.
Meter · per building / year
One building, from bills to signed-ready.
- A folder of utility bills becomes twelve months of parsed meter dataImport wizard · PDF, Excel, CSV, photos of bills
- Benchmark and bill analysis reports go back to the customerDeliverable · the artifact that earns the site visit
- A capture walk photographs nameplates; specs fill themselves inEquipment · make, model, serial, capacity, age
- BAS and telemetry connect; scheduling and control faults surfaceTelemetry · Load Analysis
- Your ECMs run against real equipment, ranked with confidence rangesLab · custom measures alongside the standard library
- A scenario becomes your proposal, in your report formatScenario → Case → Deliverable
Eight weeks of analysis. Three days of work. Nothing rebuilt twice.
“Once the fire smolders to coals, hard to reignite.”
Under the hood
Fast enough to screen. Rigorous enough to sign.
Three calculation methods behind one interface. The platform picks the right fidelity for the stage of the deal and the data that exists — and reports how confident it is either way.
Quantile regression trained on 10M+ calibrated simulations, returning 10th–90th percentile savings ranges from as little as an address, area and annual energy.
Engineering calculations against captured nameplate specifications and the specifications of the replacement unit — including your own methods.
Auto-constructed building energy models, calibrated against utility data at the click of a button. Export the model file for offline analysis.
Transfer learning holds up where change-point regression collapses. A partial bill history produces a full baseline matching the building's real seasonal pattern.
HVAC equipment and controls, lighting, envelope, renewables, storage and demand flexibility — plus every measure you bring with you.
Median error against calibrated simulations, inside ASHRAE Guideline 14 tolerances for monthly calibration.
The offer
Send one building from your backlog.
Pick something real — ideally one with incomplete utility data that has been sitting in the queue. Send the address, the square footage and whatever bills you have. You get back a benchmark, a ranked measure list with confidence ranges and a sample proposal, inside one business day. Let your engineers judge the baseline themselves.
Or reach us directly: sales@energy-pilot.ai