Solar Site Survey vs Satellite Design: When Each Is Enough

A physical site survey costs you a truck roll, a technician’s day, and a week on the project timeline. Satellite solar design costs you an address and about an hour. The question is not which one is better. The question is which projects can safely skip the truck roll, and which projects will punish you for skipping it. Get this call wrong in one direction and you burn margin on surveys you did not need. Get it wrong in the other direction and you redesign the layout after the crew is already on the roof.

Direct answer. Satellite solar design is enough when the roof is simple, structurally recent, free of tree shading, and covered by current high-resolution imagery with LiDAR data. A physical site survey is mandatory when the roof structure is doubtful or the building is older than 15 to 20 years. It is also mandatory when trees or parapets create complex shading, or the imagery is stale or low resolution. Most residential and small commercial rooftops pass the satellite test. Most old industrial sheds do not.

This guide is written for two readers. The first is Mike, the US residential installer who runs 40 to 80 site surveys a month and suspects half of them were unnecessary. The second is Rohan, the Indian EPC founder quoting C&I rooftop projects across three states. He cannot send an engineer to every tin shed before the bid deadline. Both of them need the same thing: a defensible rule for when satellite solar design is enough, and when it is not.

What satellite solar design actually sees, and what it misses

Satellite solar design means building the PV layout, shading model, and production estimate from aerial or satellite imagery plus remote data sources. No engineer visits the site. The inputs are orthomosaic imagery, elevation data, irradiance databases, and the customer’s utility bill.

Here is what a good remote dataset gives you:

• Roof geometry. Plan-view dimensions, pitch (from LiDAR or stereo imagery), azimuth, and obstructions visible from above. Our own breakdown of satellite roof measurement for solar covers accuracy limits in detail.
• Shading environment. Trees, nearby buildings, chimneys, and HVAC units, modeled in 3D when LiDAR is available. This feeds a proper shading analysis.
• Irradiance. Site-specific Global Horizontal Irradiance from databases such as Solargis or the NSRDB, accurate to a few percent for energy modeling.
• Electrical context. Interconnection voltage and utility territory, from public records.

Here is what it cannot give you:

• Structural condition. Rafter size, purlin spacing, corrosion in a tin shed, or a sagging ridge line. Imagery sees the roof skin, not the bones.
• Roof membrane condition. Remaining life of the roof matters when you bolt a 25-year asset to it.
• Electrical room reality. Main panel capacity, breaker space, conduit routes, and meter location.
• Sub-meter vertical detail. Parapet heights, small vents, and offset crickets often blur or vanish in coarse imagery.

The accuracy ceiling is real. As Exactus Energy (2023) points out, free consumer imagery like Google Maps is fine for a first look but misses the detail a final design needs. Commercial providers with fresh aerial imagery and LiDAR close most of that gap. They do not close all of it.

What a physical site survey adds, and what it costs

A physical site survey puts a trained person on the roof and in the electrical room. It converts assumptions into measurements.

A proper survey captures five things no satellite can:

1. Structural facts. Rafter or purlin sizes, spacing, span, material condition, and attachment points. On an Indian industrial shed, this means measuring purlin sections and checking for corrosion at the roof sheet laps.
2. Roof condition. Membrane age, ponding areas, prior repairs, and whether the roof needs replacement before the array goes on.
3. Electrical facts. Panel make and capacity, available breaker slots, meter configuration, and a viable conduit path.
4. True obstruction dimensions. Parapet heights measured with a tape, not estimated from a shadow.
5. Site logistics. Access, crane reach, staging area, and safety anchor points.

The cost side matters for your P&L. In the US, a residential site survey runs $150 to $500 per visit in direct cost. Industry-observed figures for a failed or repeated truck roll push that higher. In India, an outstation industrial survey typically costs ₹8,000 to ₹25,000 once you add travel and a day of engineering time. Soft costs are the non-hardware share of a solar project, including customer acquisition, permitting, and inspection labor. They remain a large slice of residential PV pricing, as the US Department of Energy notes in its soft cost overview. Every unnecessary survey inflates that slice.

When satellite solar design is enough

Satellite solar design is enough when every input the design depends on is visible, measurable, and current in remote data. In our delivery work, six project profiles pass that test reliably.

1. Sales-stage proposals. Every project, without exception, should start remote. A 3D pre-design with shading modeling gives the customer a layout and a production number in 48 hours. Nobody wins a bid by surveying first.

2. Newer residential roofs. Homes built after 2005 in the US, with composition shingle roofs, simple gable geometry, and no trees overhanging the array area. The structure follows code, the imagery is fresh, and the AHJ accepts remote-derived layouts in most jurisdictions.

3. Large flat commercial roofs with clean imagery. Big-box retail and warehouses with membrane roofs, visible roof drains, and documented as-built drawings. When the customer supplies the structural drawings, the satellite fills in layout geometry.

4. Ground-mount screening and feasibility. Land feasibility work is remote by nature. Irradiance, slope, access, and grid proximity all come from desktop data. Our site survey and land feasibility service runs the first pass remotely for exactly this reason.

5. Portfolio re-quotes. When you quoted a site two years ago and the imagery confirms nothing changed, re-running the design remotely beats sending a crew again.

6. Net-metering and DISCOM pre-applications in India. Format drawings for net-metering approval need the layout and SLD, not a structural report. Remote data covers it, as the process in our solar site survey in India guide shows.

Verdict. If the design decision you are making is commercial, satellite solar design is enough almost every time. Should we bid this, at what price, with what yield: remote data answers all three. If the decision is physical, remote data is a starting point, not an answer. Will this roof hold this array for 25 years: only a survey answers that.

When a physical site survey is mandatory

Some risks live below the resolution of any imagery. Send the truck when any of these conditions apply.

Old roofs. A roof older than 15 to 20 years, or any roof where the customer cannot tell you the installation year, needs eyes on it. Bolting a 25-year PV asset to a roof with 5 years of life left creates a removal-and-reinstall bill later. Industry-observed costs for removal and reinstallation run $3,000 to $7,000 on residential systems, which erases the survey savings many times over.

Doubtful structure. Tin sheds in Gulf countries and Indian industrial estates are the classic case. Many were built for wind load only, with no reserve for added dead load. A 540 Wp module plus mounting adds roughly 12 to 15 kg per square meter. On a corroded purlin system, that margin does not exist. You cannot see purlin corrosion from orbit. Structural verification standards such as IS 875 Part 3 for wind load require real section data. STAAD Pro structural calculations need measured inputs to mean anything.

Tree shading. A single large tree near the array field creates hour-by-hour shading that coarse imagery models poorly. LiDAR helps, but deciduous canopies change between leaf-on and leaf-off captures. When trees are within throwing distance of the array, measure them on site or fly a drone. The physics is unforgiving, as our guide on shadow analysis for solar rooftops explains.

Irregular parapets and obstructions. Parapet walls cast winter shadows far longer than their height suggests. A 1.2 m parapet at 25 degrees latitude throws a shadow over 2.5 m long in December morning sun. If the parapet height is not measurable in the imagery, the row spacing in your layout is a guess.

Complex or unknown electrical. Main service panel upgrades, subpanel locations, and conduit paths through finished buildings all need a physical look. A satellite has never seen the inside of an electrical room.

Stale or low-resolution imagery. Parts of India, Africa, and rural US counties have imagery that is 2 to 5 years old. New construction, roof changes, and tree growth make it unreliable. If the imagery date is not within 12 to 18 months, treat remote measurements as unverified.

Site survey vs satellite design: the decision table

Use this table as the first-pass filter. It maps the conditions that decide the question to the right answer for each design stage.

Site condition	Sales-stage design	Permit / IFC design	Verdict
Residential roof, built after 2005, simple geometry	Satellite	Satellite + photos	Remote is enough
Residential roof, age unknown or pre-2000	Satellite	Physical survey	Survey mandatory
Flat commercial roof, as-built drawings available	Satellite	Satellite	Remote is enough
Flat commercial roof, no drawings, irregular parapets	Satellite	Physical survey	Survey mandatory
Industrial tin shed, any age	Satellite	Physical survey + structural check	Survey mandatory
Trees within 15 m of array field	Satellite (LiDAR)	Drone or physical survey	Survey mandatory
Ground mount, feasibility stage	Satellite	Geotech + topo survey later	Remote for screening
Imagery older than 18 months at the site	Satellite (flagged)	Physical survey	Survey mandatory
Any project, customer proposal	Satellite	Not applicable yet	Remote is enough

Two patterns stand out. First, every row starts with satellite for the sales stage. Second, the permit and IFC stage splits on structure, shading, and imagery age, not on project size.

The 5-Gate Remote Design Filter

Over hundreds of rooftop projects delivered from our Surat and Ahmedabad offices, we have compressed this decision into five pass/fail gates. A project that passes all five gets designed remotely. A project that fails any one gate gets a survey, either physical or drone. We call it the 5-Gate Remote Design Filter.

Apply it next week on live bids. Score each site in two minutes during qualification. Pass all five and the design starts the same day. Fail one and you book the survey before the customer signs, which keeps the change-order conversation off the install schedule.

The middle path: drone surveys and hybrid remote data

The satellite-versus-survey framing hides a third option that covers most of the gray zone. A drone survey captures sub-inch roof geometry and true obstruction heights in a 20-minute flight, without anyone climbing the roof. As Scanifly (2023) argues, satellite imagery supports a preliminary design but not a final array design, and drone capture is their answer to the gap.

The hybrid stack looks like this in practice:

1. Satellite pass. Free or cheap. Qualifies the site, kills bad leads, builds the proposal.
2. Customer photo pass. The customer shoots the main panel, the meter, and the attic or roof underside from a phone. This closes most electrical and structural unknowns on simple homes.
3. Drone pass. For complex roofs, heavy shading, or unknown parapets. Sub-inch accuracy, 3D model, no ladder.
4. Physical survey. Reserved for structural doubts, old roofs, and electrical complexity. The truck rolls only when the first three passes leave a real question open.

A note on the vendor debate. Tool vendors argue for the tool they sell. Aurora’s remote design flow proves satellite works for volume residential, and drone vendors prove drones win on complex roofs. Both are right about their segment. Wattmonk (2025) makes the balanced point: remote tools build the initial quote, and a physical or high-fidelity digital survey confirms the true scope. EnergyScape Renewables (2026) lands in the same place: satellite imagery helps initial estimates but is not accurate enough for final design and permitting. For the residential end, EagleView (2022) advises satellite imagery for a quick first look, with physical measurement before final design.

What the survey decision does to your project P&L

This is the part that matters to the EPC owner. The survey decision is a margin decision.

Run the numbers on a US residential installer doing 50 surveys a month. At $300 average direct cost per survey, that is $15,000 a month. If remote pre-design filters out 40 percent of those sites as either unfit or easy enough to design from imagery and customer photos, you save $6,000 a month, or $72,000 a year. That number is larger than most installer’s software budget.

For an Indian C&I EPC quoting 20 rooftop projects a quarter across Gujarat and Maharashtra, outstation surveys at ₹15,000 each cost ₹3,00,000 a quarter. A remote-first stack cuts that to surveys only on the projects that fail a gate, typically a third of the pipeline. The saving, roughly ₹2,00,000 a quarter, pays for the design work itself.

Now the other side of the ledger. Skipping a needed survey on one 200 kW industrial shed can produce a redesign after procurement, a mounting system that does not fit the purlin layout, or a structural retrofit. Industry-observed costs for that class of mistake run from ₹1,50,000 to well over ₹10,00,000 once rework, delay penalties, and remobilization land. One mistake erases a year of survey savings.

How Heaven Designs helps

The survey decision only works if your design partner can execute both sides of it. A satellite-only shop will tell you every roof is remote-friendly. A survey contractor will tell you every roof needs a visit. We run both tracks, so the project dictates the method. Send us an address and a utility bill, and you get a 3D pre-design with shading in 48 hours. Send us a doubtful tin shed, and we tell you to survey it first, then build the IFC pack from the survey data.

• Solar 3D Pre-Design: Satellite and LiDAR-based 3D model with shading analysis in 48 hours. No site visit. Built for the sales stage.
• Solar Rooftop Detailed Engineering Design: Full IFC pack: GA, SLD, structural, BOQ, mounting. Works from survey data, drone models, or verified remote inputs.
• Site Survey & Land Feasibility Services: Remote feasibility screening plus structured survey checklists when the gates say go.
• Download a sample deliverable: A redacted pre-design and an IFC pack, so you can judge the difference yourself.

If you want a second set of eyes on a specific site, contact us with the address. We will run the 5-Gate filter on it for free and tell you which track it belongs on.

FAQ

Can you design a solar system from satellite imagery alone?

Yes, for sales-stage design on most simple rooftops. Satellite and aerial imagery plus LiDAR data give you roof geometry, pitch, azimuth, obstructions, and shading. For final permit and construction design, satellite alone is enough only when the structure is recent or documented, shading is simple, and the imagery is fresh. Old roofs, tree shading, and unknown structures still need a physical or drone survey.

Is a site survey required before installing solar panels?

There is no universal legal requirement for a physical site survey. What the AHJ, DISCOM, or lender requires is accurate design inputs, and those inputs decide the question. Many US AHJs accept permit plan sets built from aerial measurement data. Structural doubts, old roofs, and complex electrical situations make a survey practically mandatory even where no rule demands it.

How much does a solar site survey cost?

In the US, a residential site survey costs $150 to $500 in direct expense, and more when you count scheduling delay. In India, an outstation industrial survey typically runs ₹8,000 to ₹25,000 including travel and engineering time. Remote satellite design costs a fraction of that and delivers in 24 to 48 hours, which is why a remote-first workflow saves $6,000 or more per month for a busy installer.

How accurate is satellite roof measurement for solar?

With fresh high-resolution aerial imagery and LiDAR, plan-view measurements land within a few percent of tape-measured values, which is enough for layout and production estimates. Free consumer imagery is less reliable, and Exactus Energy (2023) recommends it only for a first look. Accuracy collapses when imagery is stale, resolution is coarse, or vertical details like parapet heights matter.

When is a drone survey better than a physical survey?

A drone survey wins when the open questions are geometric: obstruction heights, complex roof planes, tree canopy positions, and parapet dimensions. It captures sub-inch data in one short flight without roof access risk. A physical survey still wins when the questions are structural or electrical, because a drone cannot see purlin corrosion, decking condition, or the inside of a main panel.

What is checked during a physical solar site survey?

A proper survey covers five areas: structure (rafter or purlin sizes, spacing, condition), roof condition (membrane age, remaining life), electrical (panel capacity, breaker space, conduit path), true obstruction dimensions, and site logistics (access, staging, anchoring). Each of these converts a design assumption into a measurement before procurement.

Can satellite solar design work in India, where LiDAR coverage is thin?

Yes, with adjustments. Stereo satellite imagery and customer-supplied photos replace LiDAR for pitch and height data on most sites. Net-metering and DISCOM pre-application drawings work well from remote data. The limits bite harder on old industrial sheds, where structure and corrosion decide the design, so those projects still need a physical survey before the IFC stage.