We Compared Satellite Solar Estimates to Real Installs. The Gap Will Shock You
If you've ever typed your address into an online solar calculator and gotten an instant number for panel count, system size, or annual savings, you've seen a satellite solar estimate in action. These tools are fast, free, and genuinely useful for a first look. But homeowners who go on to get a real, permitted installation often notice the final design — panel count, layout, even projected output — looks different from that first estimate. That gap isn't a glitch. It's the predictable result of what a satellite photo simply cannot tell you about your roof.
This post walks through why satellite-only estimates and as-built installs diverge, what causes the difference, and how a full 3D roof design step — the kind eRoof runs before you ever get a contractor quote — is built to close that gap early, instead of letting you discover it during permitting or installation.
Why a Satellite Solar Estimate Is Only a Starting Point
Tools like Google Solar AI process aerial and satellite imagery to model your roof's shape, orientation, and sun exposure. That's genuinely powerful — it's why eRoof uses this same technology as the first step in its own flow. A satellite pass can identify roof planes, estimate tilt and azimuth, and model shading from nearby buildings or large trees at a broad scale.
What it can't do is see your roof the way an installer standing on it will. Imagery resolution, capture angle, and update frequency all limit how precisely a satellite-based model can represent the physical, present-day condition of your specific roof. That's the root of nearly every estimate-vs-installed gap.
The Real Reasons Satellite Estimates and Installed Systems Differ
Roof Obstructions Satellite Imagery Doesn't Resolve
Vents, plumbing stacks, chimneys, skylights, HVAC units, and satellite dishes all take up roof space a panel could otherwise occupy. Depending on image resolution and capture angle, small obstructions can be underrepresented or missed entirely in an automated model. A real site assessment, or a detailed 3D model built from more complete roof data, accounts for every one of these — which almost always reduces usable roof area, and therefore panel count, versus a first-pass estimate.
Shading That Only Shows Up at Certain Times of Day or Year
Satellite-based shading analysis typically models shade from large, fixed structures and trees visible in the image at capture time. It's much harder for any remote model to fully capture seasonal sun-angle changes, a neighbor's tree that's grown since the last image update, or shading from features on the roof itself, like a dormer shading an adjacent plane in winter. Real installs are shaped by shading patterns across the full year, not a single snapshot.
Panel Derating and Real-World Equipment Losses
The nameplate wattage on a solar panel is measured under standardized lab conditions. In the field, panels lose some output to heat, age, manufacturing tolerance, and connection resistance — a well-understood phenomenon called derating. A rough estimate that applies a generic efficiency assumption will differ from a design that models derating for the specific panel and inverter equipment actually being installed.
Inverter and System Losses
Every solar system loses some energy converting DC power from the panels into usable AC power, plus additional losses in wiring runs and connections. These losses vary with inverter type (string vs. microinverter), wire length, and system layout — all details that only become concrete once a full design is drawn, not when a satellite tool is estimating at the address level.
Orientation and Tilt Assumptions vs. As-Built Reality
A satellite model estimates roof pitch and compass orientation from imagery, which is a reasonable approximation but not a survey-grade measurement. Small errors in either can shift projected output. A 3D design built with more precise roof geometry narrows this down considerably.
Weather, Soiling, and Local Climate Patterns
Dust, pollen, snow cover, and regional weather patterns affect real-world output over a year in ways a generic regional solar-resource estimate can only approximate. This is a known limitation of any remote estimate and one reason installed performance is reported as a range, not a single guaranteed number.
Permitting-Driven Design Changes
Local jurisdictions often require fire setbacks, specific rapid-shutdown equipment, structural load considerations, or rooftop walkway clearances. These requirements can force a redesign that reduces panel count or shifts placement — changes a pre-permitting satellite estimate has no way to anticipate. For more on how this affects what homeowners are told upfront, see how Google Solar AI is changing who gets approved.
Common Causes of the Estimate-vs-Install Gap
| Cause | Why It Happens | What Narrows It |
|---|---|---|
| Roof obstructions (vents, chimneys, skylights) | Imagery resolution/angle limits | Detailed 3D roof modeling |
| Seasonal or growing shade | Single-capture snapshot | Multi-angle, updated modeling |
| Panel derating | Lab ratings vs. field conditions | Equipment-specific design |
| Inverter/wiring losses | Layout-dependent, not address-dependent | Full system design |
| Tilt/orientation error | Imagery-based approximation | Precise roof geometry |
| Weather and soiling | Regional averages, not site-specific | Local climate data in design |
| Permitting requirements | Jurisdiction-specific rules | Permitting-ready design step |
How a 3D-Verified Design Step Narrows the Gap
This is the core idea behind eRoof's approach: start with the same satellite-based estimate everyone else uses, then go further before you ever talk to a contractor. eRoof builds a full 3D model of your specific roof, places panels around actual obstructions, applies equipment-specific derating and inverter-loss assumptions, and checks the layout against permitting-relevant constraints — all before a contractor shows up.
The output isn't just a nicer estimate. It's a permitting-ready design that contractors can review and claim as a bid, so the number you see is grounded in your actual roof geometry rather than a generic model. To understand how to evaluate the numbers you're given at any stage, read a solar estimate like a pro is a good next step, and if you're wondering whether layout itself matters this much, see the panel placement mistake that kills solar ROI.
From Design to Installed System
Closing the estimate-vs-install gap on paper is only half the job — the other half is making sure what gets designed is what actually gets built. Once a design is finalized, equipment sourcing can connect through Stockup, and installation itself can be handled through ProDone, keeping the same verified design intact from roof model to finished install rather than leaving room for a second round of surprises on install day.
For homeowners comparing a quick satellite estimate against a full in-person quote, it's also worth understanding how instant roof estimates are replacing in-person quotes — and where each approach still has real value.
Solar Estimates and Grid Participation
A more accurate design doesn't just matter for your roof — it matters for how your system participates in the broader grid. Systems that are modeled accurately from the start are easier to integrate into programs like virtual power plants, where predictable, verified output matters for dispatch planning. You can learn more about how this works through Humdo's virtual power plant solutions, and how a solar system fits into a broader connected home through smart property IoT.
Frequently Asked Questions
Is a satellite solar estimate accurate enough to make a decision from?
It's accurate enough for a first-pass comparison of whether solar makes sense for your roof, but it's not a substitute for a detailed, roof-specific design. Treat it as a starting point, not a final number.
Why does my final panel count end up lower than the first estimate?
Usually because a detailed design accounts for obstructions, setbacks, and permitting requirements that a satellite-only pass can't fully resolve from imagery alone.
Does eRoof still use satellite imagery like Google Solar AI?
Yes — the satellite-based estimate is the first step. eRoof then builds on it with a full 3D roof design so the final numbers reflect your actual roof rather than a generic model.
How long does it take to go from estimate to a permitting-ready design?
The 3D design step is built to run in minutes, not weeks, so you can move from an initial estimate to something a contractor can act on quickly.
Can contractors just use the eRoof design instead of doing their own site visit from scratch?
Contractors can claim AI-approved designs as bids, using the verified 3D model as their starting point rather than rebuilding the analysis independently.
The Bottom Line
The gap between a satellite solar estimate and a real install isn't a sign that either number is wrong — it's a sign that they're answering different questions at different stages of the process. A satellite estimate tells you whether it's worth looking further. A full 3D design tells you what will actually get built. Closing that gap early, before you're deep into contractor quotes, is what a permitting-ready design process is for.
Ready to see how your own roof holds up beyond the satellite estimate? Explore how a connected, verified solar design fits into your broader home setup with Humdo's smart property IoT solutions.