Drone Inspection Services: Roof, Solar and Infrastructure: Complete Guide (2026)

In Australia, CASA separates the pilot from the business. The Remote Pilot Licence qualifies you as an individual, and training typically runs from about 1,250 Australian dollars at the cheap end to 3,500 dollars for a full course with assessment. The Remote Operator Certificate authorises an organisation to carry out commercial operations. There are excluded category pathways that permit some low-risk commercial flying without a ReOC, but they are narrow, and the landowner exclusion specifically requires that you accept no payment. If you plan to trade as a drone service provider in Australia, budget for both the RePL and the ReOC, and know that the ReOC is a real administrative commitment, with a nominated Chief Remote Pilot and annual renewal.

Everywhere else, find your national aviation authority and read the primary source yourself. Trade blogs go stale fast.

Insurance, which is the second gate and the one people skip

So the certificate question is settled, wherever you happen to fly. What decides whether a building owner actually lets you over their property is the cover you carry when you ask.

In the United States the FAA does not require drone insurance for commercial operations. That is a legal fact with almost no commercial weight. Almost no serious client will let you on site without a certificate of insurance naming a liability limit, and many will want to be named as an additional insured. Expect one million US dollars of liability as the standard ask. Published rates put an annual one million dollar liability policy starting around 750 dollars, with brokers commonly quoting 600 to 800 dollars; hourly on-demand cover runs roughly 10 to 15 dollars per flight hour, and monthly plans sit in the 40 to 65 dollar range. Two million dollar limits start around 1,131 dollars a year. Hull cover for the aircraft itself is separate and priced as a percentage of insured value.

In the United Kingdom insurance is mandatory. The CAA requires operators flying commercially to hold third party insurance compliant with assimilated Regulation (EU) 785/2004 and the Civil Aviation (Insurance) Regulations 2005. The CAA casts the commercial net wide: being paid to take photographs or video, being paid to carry out surveys, delivery, farm or estate work, and flying in an educational setting all count. The minimum third party liability for aircraft up to 500kg is 750,000 Special Drawing Rights, an IMF-defined unit, so the sterling equivalent moves with exchange rates. Hobby policies bundled with model flying club membership generally exclude commercial use, and some exclude the acts of war cover that 785/2004 requires, so read the policy wording and ignore the marketing.

Practical advice: buy annual liability cover before you quote your first job. Clients ask for the certificate when they book, and if you have to say the cover is on order, you lose the job.

If a client asked for your insurance certificate on the phone tomorrow, could you email it within the hour? That is the test, and passing it costs you around 750 dollars a year in the US.

Equipment and what it really costs

With the paperwork behind you, the next money leaves your account for hardware. This is where the $1,500 end and the $15,000 end of the range go their separate ways.

There is a wide gap between the minimum kit that can do visual roof work and the kit that opens up the better paid thermal and mapping jobs.

At the visual end, a current prosumer aircraft with a decent sensor, obstacle avoidance and a mechanical or good electronic shutter will do a competent residential roof survey. Expect 1,200 to 2,500 US dollars or roughly the same in pounds for the airframe, plus a fly-more style bundle for spare batteries. You need at least three batteries; two will not get you through a commercial day, and battery health fades faster than people expect.

At the thermal and enterprise end the numbers jump. Dealer listings for the DJI Mavic 3 Thermal in 2026 sit around 6,800 to 8,000 US dollars depending on the reseller and bundle. Newer enterprise thermal platforms such as the Matrice 4T were listed by US dealers at roughly 7,100 to 7,400 dollars new in mid-2026. The RTK module for the Mavic 3 Enterprise series runs about 1,028 dollars, the enterprise controller about 1,399 dollars and a battery kit about 799 dollars. That is easily a 10,000 dollar kit before software.

Other costs people forget: a tablet or laptop that can process photogrammetry, cloud processing credits at roughly 99 to 199 dollars per property for high-accuracy 3D models, a landing pad, a high-visibility vest and cones, a hard case, spare propellers, ND filters, and mobile data for airspace apps. Add a modest website and a Google Business Profile.

There is a significant buying complication if you are in the US. On 22 December 2025 the FCC added all foreign-produced uncrewed aircraft systems and UAS critical components to its Covered List following the FY2025 NDAA Section 1709 deadline. The practical effect, as summarised by law firms covering the action, is that drones already authorised before that date stay usable and you may keep flying aircraft you already bought, but new foreign-manufactured models cannot receive FCC authorisation and therefore cannot be lawfully imported or sold in the US. If you are buying in the United States, expect supply of new Chinese-manufactured models to tighten and prices on existing stock to behave oddly, and if you intend to bid for federal or federally funded work, check procurement language on country of manufacture before you buy anything. A second airframe as a spare is easier to justify now than it was.

Thermal, and whether it is worth the money

Thermal is where the margin is, and also where amateurs get found out. A thermal camera produces a picture that looks impressive and means nothing unless you understand emissivity, reflected apparent temperature, heat transfer and the environmental conditions that make a scan valid. Solar inspection in particular needs clear skies, enough irradiance and low wind, which narrows your working window a great deal.

If you want to sell thermal work, get the qualification. Infraspection Institute sells a Level 1 sUAS Thermography distance learning course at 1,695 US dollars, with the certification exam a further 375 dollars, covering roughly twenty hours of instruction in infrared theory, heat transfer, equipment selection, standards compliance, image analysis and report generation. The Infrared Training Center runs a four-day Level I course in classroom, live online and onsite formats and lists sUAS-specific tracks, though it quotes prices on request instead of publishing them, so ask before you assume. Some universities and continuing education providers run applied UAS thermography courses at lower cost; check that whatever you buy leads to a recognised certification, since a certificate of attendance carries little weight.

The commercial argument is simple. Adding thermal to a roof inspection adds roughly 100 to 200 dollars per job on published rate guides. Solar work is priced per megawatt and opens a whole different class of client to you. If you only ever do visual roof surveys, you will always be competing on price.

So when would you buy the thermal kit: now, on hope, or later, against clients who already book you? The second answer is cheaper, and the minimal setup at the end of this page assumes it.

What people actually charge

That is what the kit costs you. Now for what a day of flying it is worth to somebody else.

Treat every published rate guide as a rough indication. Almost none of them disclose their sample. That said, several independent 2026 sources cluster tightly enough to be useful to you.

United States residential roof inspection: 150 to 400 dollars for a standard visual survey with a basic photo report, rising to 250 to 600 dollars when photogrammetry processing and a measurable 3D model or orthomosaic are included. Thermal typically adds 100 to 200 dollars.

United States commercial roof inspection: 400 to 1,200 dollars depending on roof area and number of structures, with large or complex buildings quoted above 1,500 dollars. Some providers quote multi-building industrial portfolios in the thousands.

United Kingdom residential roof survey: guidance ranges from about 150 to 750 pounds, with a typical detached or semi-detached property photography-only survey landing between roughly 250 and 500 pounds. UK day rates for roof inspection work are reported in the 400 to 1,400 pound band depending on deliverables and how hard the job is to fly.

Hourly framing, which some clients prefer: roof and building inspection 200 to 500 dollars per hour, thermal and infrared 300 to 700 dollars per hour, construction and survey work 250 to 600 dollars per hour.

Solar: residential and small commercial arrays 150 to 400 dollars per inspection. Utility-scale is priced per megawatt. The 300 to 500 dollars per MW band is widely repeated, but every published instance of it traces back to drone-training companies selling courses to would-be operators, and none to an operator or asset owner disclosing what was actually paid. Treat it as an unverified industry estimate, expect lower per-MW figures on very large sites, and get a real quote from an asset owner before you build a plan around it. A 5MW community solar farm at 400 dollars per MW is a 2,000 dollar job that takes most of a day plus a day of analysis.

Wind turbine blade inspection: commonly 300 to 600 dollars per turbine for standard visual work, rising to 800 to 2,000 dollars where advanced sensors or detailed engineering analysis are involved. This is repeat, high-volume work, and it is dominated by established contractors and hard to break into on your own.

Pricing advice that survives contact with reality: avoid quoting by the hour to a client who has never bought this before, because they will fix on the twenty minutes of flying. Quote a fixed price per property or per asset, with a defined deliverable, a defined turnaround and a stated re-visit policy for weather. Set a minimum call-out that covers your travel. Charge separately for rushed turnaround, because storm-season clients will pay for it.

What is the furthest you would drive for a 150 dollar survey before it stops being worth your morning? Put that distance into your minimum call-out today.

With the thermal add-on at roughly 100 to 200 dollars a job, a few roof surveys in a month could start covering what the kit costs to keep. If inspections reach $400 a month, the floor of the income figure here, what remains after travel and insurance could help the car payment go out without a second look at the balance. Set the minimum call-out so every flight pays for its own trip.

Push that further to the upper end of the range. Steady roof and solar work could mean the mortgage paid from flights you chose to take. It could mean afternoons free for your kids because you set the schedule yourself. Every job starts with a certificate most people never bother to earn.

Finding the first client

You have the aircraft, the cover and a price you believe in. None of it earns you anything until someone calls, and this part decides whether they do.

Your first client will almost certainly come from outside the marketplaces. Broker platforms such as Zeitview, FlyGuys and Droners.io exist and do pay, and they are a reasonable way to get reps and build a portfolio, but the platform sets the rates, missions in busy areas are claimed within minutes, and one long-running observation in the sector is that pilots who make a living get the large majority of their revenue from direct clients and only a minority from platforms. Use platforms to learn the workflow, and build the business elsewhere.

The direct route that works is unglamorous. Pick one buyer type and go to them. Roofing contractors are the easiest first target because the value to them is immediate: you take a ladder climb out of their estimating process and shorten their quote turnaround. Walk in with a finished sample report on a property you have already surveyed, and leave the price list in the car. Offer the first survey free or at cost in exchange for permission to use it as a case study, and say plainly that this is a one-time introductory arrangement.

Loss adjusters and independent claims firms are your second target. They are harder to reach and slower to onboard, but they bring volume in storm seasons and they care about report consistency more than price. Expect to be asked for your insurance certificate, your licence, and sometimes evidence of a documented safety management approach before you get a job.

Property managers, housing associations, church and heritage building trustees, facilities managers at industrial estates, and commercial solar installers round out the list. Local authority and utility work usually runs through procurement frameworks and is out of realistic reach in your first year as a solo operator.

Practical channels that bring enquiries: a Google Business Profile with genuine local service area targeting and real photographs, a website with one page per service and one page per town you serve, LinkedIn outreach to named estimators and adjusters (generic company inboxes rarely answer), and turning up in person at regional roofing, facilities management and solar trade events. Searches for phrases like drone roof survey in a named town are low volume and extremely high intent.

Which roofing contractor in your town would you walk into first, sample report in hand? Write the name down, and survey a property for that pitch this week.

A realistic ramp

Month one to two is administrative. You study, sit the exam, apply, wait for vetting, register the aircraft, buy insurance and build a report template. Revenue is zero, and outgoings are between roughly 1,500 and 15,000 depending on how much kit you buy.

Month two to four you fly practice jobs. Survey your own house, family properties and any building whose owner will let you, and build a portfolio of six to ten finished reports. Start approaching contractors. Expect your first paid job somewhere in this window, and expect it to be small.

Month four to nine, if your outreach is steady, a part-time operator in a reasonable market might reach two to six jobs a week. At an average net ticket of 250 to 350 that is roughly 2,000 to 8,000 gross a month before travel, insurance amortisation, software and tax, though the lower half of that band is far more common in the first year.

Month nine to eighteen is where your business either specialises or stalls. Operators who add thermal, win a retainer with one or two contractors, or land a seasonal insurance panel arrangement tend to settle into steady work. Operators still selling generic aerial photos tend to plateau at a few hundred a month and quietly stop.

Seasonality is severe and rarely discussed. You cannot fly in high wind, heavy rain or fog. In the UK and northern US that removes a real share of working days from November through February. Storm seasons bring spikes of insurance work. Plan your cash flow around a lumpy year.

If January brought you two flyable days, what would pay your bills that month? Answer that before you leave a salaried job for this.

What goes wrong

Weather cancellations wreck schedules. Build a re-visit policy into every quote and charge for a second mobilisation beyond the first.

Airspace refusals. A property sits under controlled airspace or inside a restricted area, and the authorisation does not come through in time. Check airspace when you quote; the morning of the job is too late.

Underpricing. The single most common failure. New operators price against the twenty minutes of flight time and forget travel, insurance, kit depreciation, processing time and the fifteen per cent of jobs that need a return visit. Work out your real cost per job before you set a rate.

Report quality. Clients who receive an unstructured folder of images do not come back. Standardise your template on day one: property details, date and conditions, flight log summary, overview images, per-elevation images, annotated defects with locations, and a plain-language summary.

Data protection. You are photographing private property and often catching neighbouring properties and people along the way. In the UK this engages UK GDPR, and you should check the current Information Commissioner guidance on surveillance and drones before you start. Have a privacy notice, a retention period and a lawful basis. In the US, state and local privacy and trespass rules vary widely, and some municipalities have their own restrictions even though airspace is federal.

Scope creep into regulated advice. You can document what you observe. Unless you separately hold the qualification, you are no structural engineer, chartered surveyor or licensed roofing contractor, and your reports should never read like structural opinions or claim settlements. Describe conditions and defects factually, state your limits in the report, and refer the client to a qualified professional for diagnosis. In the UK, calling your output a survey when a client reasonably takes that to mean a RICS survey is a real commercial risk. Say what you did and what you left out.

Equipment loss. Drones fly into things. A fly-away, or a crash into a client's roof, without hull cover and adequate liability can end your business in an afternoon.

Every month you wait, another pilot near you passes the exam and introduces themselves to the roofer you meant to call. The certificate costs the same next year. The local relationships will cost far more, because by then they belong to someone else. Book your study time tonight and pick an exam date you can see on the calendar.

Who should skip this

If you cannot stand regulatory paperwork, do not start. The flying is a small fraction of the job, and the compliance load never goes away.

If you need income within a month, do not start. Licensing alone takes several weeks, and the first paid job usually comes months later.

If you live somewhere with dense controlled airspace, a small building stock or an established incumbent operator with contractor relationships, be honest with yourself about the market before you spend ten thousand on kit.

If you are uncomfortable selling, do not start. This is a business-to-business service sold by direct approach. Nobody finds you by accident.

If you are hoping for passive income, look elsewhere entirely. This is a service business with a per-job time cost, seasonal demand and physical site attendance.

If you plan to fly without insurance because your jurisdiction does not legally require it, you will be shut out by most commercial buyers and exposed to ruinous liability.

Regulatory changes worth watching

The FAA published a proposed rule on 7 August 2025 in the Federal Register to normalise beyond visual line of sight operations, which would become a new Part 108. It remains a proposed rule. Commentators have offered various timelines, and you should treat none of them as authoritative until a final rule is published, with implementation expected to lag publication by a year or more. If and when it lands, BVLOS would change the economics of linear infrastructure inspection substantially, and it would likely favour larger operators with formal safety management systems over solo pilots like you.

In the United States, the FCC Covered List action of 22 December 2025 will keep reshaping equipment availability and procurement language through 2026 and beyond.

In the United Kingdom, class marking and Remote ID obligations phase in further on 1 January 2028, and EU C-class recognition ends on 31 December 2027. Judge any drone you buy now against those dates as well as against today's rules.

When beyond visual line of sight rules land, the operators with logged hours, insurance history and client lists will be first in line for the larger contracts. Those things take time to build, and nobody can rush them later. If you want to be ready when the rules change, the clock started already.

A minimal viable setup

If you want the cheapest defensible start, this is roughly it. Sit the national licence exam. Register your aircraft. Buy annual third party liability cover at the limit your target clients ask for, typically one million in the US and a 785/2004-compliant policy in the UK. Buy a current prosumer aircraft with three batteries. Build one professional report template. Survey ten buildings for free to fill a portfolio. Approach twenty local roofing contractors in person with the portfolio and a fixed price list. Add thermal capability and a thermography qualification only once visual work brings in steady revenue, because the thermal aircraft is your single largest expense, and it is far easier to justify against existing customers than against hoped-for ones.

Which ten buildings could you survey for free this month, starting with your own? A list of ten addresses is the first real step in this whole plan.

Quoting a job, and the deliverable nobody specifies

Pricing in this work usually goes wrong at the scoping stage, well before you pick a rate, because clients ask for an inspection and mean a report.

The flight is the smallest part of the job. A roof inspection might take twenty minutes in the air and several hours in total once you count travel, site assessment, airspace checks, data transfer, processing and the written output. Quoting against flight time is the most common way operators price themselves below cost.

Specify the deliverable in writing before you agree a price. What the client receives is the product: raw images, an annotated set, a written report with findings, a 3D model, a thermal overlay, or a comparison against a previous survey. These are wildly different amounts of work, and clients often assume the most elaborate one is included.

Agree the number of revisions. Reports attract requests for more annotation, re-flights of areas of interest, and reformatting for a third party. One included revision, with further work chargeable, keeps your commitment from becoming open-ended.

Price the mobilisation separately. Travel, site access arrangements, and the chance of a wasted trip because of weather are real costs. A call-out charge protects you from the client who books a flight in a week with a poor forecast and then cancels on the morning.

Decide who owns the data. Clients often assume they own everything captured; operators often assume they keep the imagery. Both positions are defensible, and your contract should say which applies, especially if you may want to use the imagery as a portfolio piece.

The operators who make this profitable sell an inspection report with a drone attached. That is a professional service with a defensible price. Selling drone flights with a report tacked on puts you up against anyone who bought the same aircraft.

Inspection reports with a defined deliverable and a stated re-visit policy are what keep clients rebooking each season. Should that work build to the $7,000 top of the monthly range and hold across a lumpy year, with expenses paid and savings behind you, the weekday job could become optional and the flying days could be the ones you pick. Few operators get there quickly, so quote the report clearly and let the repeat work build.

Weather, and the constraint that shapes the whole business

Flying depends on weather in a way clients do not instinctively understand, and managing their expectations is most of your scheduling problem.

Wind usually stops you before rain does, and the wind at roof height can differ a lot from the wind at ground level. Light conditions matter for image quality, and thermal work has its own requirements around surface temperature differences that limit the time of day as well as the day.

Three practices keep this from turning into a client-relations problem.

Book provisionally with a stated weather policy. Agree a named backup date at booking, and a clear statement of who decides and by when. The decision should be yours, because the client cannot judge flyability and will always want to go ahead.

Never fly a marginal day to keep a client happy. The output is poor, which brings a re-flight you pay for, and the risk to the aircraft and to people below rises exactly when your judgement is being bent by not wanting to reschedule.

Cluster jobs geographically. Since good days are the limit, the operators who earn well are the ones who can fill a good day with three jobs in one area, where others spend it on one job across a county. That argues for building density in a smaller radius instead of accepting work anywhere.

The seasonal pattern follows from all this and belongs in your income model: a temperate-climate operator has far fewer flyable days in winter, and the annual figure is lower than the monthly figure multiplied by twelve. Plan for that, and the quiet months will feel expected when they come.