A 90-day playbook to launch a compliant drone program

Before the clock starts

A 90-day program launch is achievable — but only if two prerequisites are in place before Week 1 starts. First, budget approval must be secured and hardware procurement can begin. Second, the program owner is named: a single person accountable for the program's regulatory compliance, safety record, and output quality. Without a named owner, the 90-day clock never really starts.

The scenario below assumes a single-site inspection program, in-house Part 107 operation (not outsourced to a DSP), and standard VLOS (visual line of sight) operations. Programs requiring BVLOS authorization should add 6–18 months for the waiver process, running in parallel with but not blocking the VLOS program launch.

Weeks 1–2: Airspace and site assessment

The first two weeks are entirely about regulatory and site intelligence. No hardware should arrive before you know what you are allowed to do with it.

Airspace classification. Use the FAA's B4UFLY app or the ArcGIS-based FAA DroneZone to map airspace classes at your site. Class G airspace (uncontrolled airspace below 1,200 feet AGL in most areas) allows VLOS commercial operations with Part 107 certification but no prior authorization. Controlled airspace (Classes B, C, D, and E surface) requires authorization — either via LAANC (Low Altitude Authorization and Notification Capability, the FAA's near-real-time authorization platform) for eligible areas, or a manual waiver for areas outside LAANC coverage.

LAANC eligibility check. Not all controlled airspace is covered by LAANC. Check whether your site is in a LAANC-enabled facility map. If it is, UAS operators can receive near-real-time authorization through apps such as AirMap, Kittyhawk, or DroneLogbook. If your site is not LAANC-enabled, you will need to file a Part 107 airspace authorization application with the FAA, which can take days to weeks depending on complexity.

Obstruction and hazard survey. Walk the site and document: obstacles at flight altitudes (power lines, antennae, cranes), no-fly zones (fuel storage, active construction, utility corridors), and preferred landing zones. Note any temporary flight restrictions (TFRs — short-term airspace reservations for special events, VIP movements, or emergencies) that may affect recurring missions.

Deliverable: A site airspace dossier — airspace class, LAANC status, obstruction map, and preliminary operating altitude and geographic boundaries for the program.

Weeks 3–4: Certification and procurement

Part 107 certification. The FAA Remote Pilot Knowledge Test (the written exam required to obtain a Part 107 Remote Pilot Certificate) is administered at FAA-approved testing centers. Registration costs $5; the exam fee is typically $175. Preparation typically requires 20–40 hours of self-study using FAA study materials or commercial prep courses. The certificate is issued within 10 business days of passing.

Certify at minimum two people. Naming a single pilot as the program's sole certified operator is a risk the previous article in this series documents at length — one personnel change grounds the program.

Hardware procurement. Purchase the aircraft, payloads, and accessories based on the platform class decision. Reference A decision framework for choosing a platform class if that has not been completed. Procurement lead times for enterprise platforms vary from days (direct order for in-stock models) to 6–8 weeks for custom configurations or platforms with export licensing requirements.

Key procurement checklist items:

• Aircraft, controller, and batteries (minimum 3 batteries per aircraft for field efficiency)
• Payload appropriate to mission type (RGB, thermal, multispectral, LiDAR)
• Spare props, landing gear, and common consumables
• Carrying/transport cases rated for the aircraft
• Tablet or monitor for controller display in field conditions

Insurance. Initiate the UAS insurance application in Week 3. Hull and liability riders require the aircraft serial numbers (coordinate timing with procurement). Allow 1–2 weeks for underwriting. Do not fly before insurance is confirmed in writing.

Weeks 5–6: Software and data pipeline setup

The data pipeline is the part of the program most frequently configured incorrectly or not configured at all until after the first flights. Configuring it before the first flight means the first flight produces immediately usable data.

Photogrammetry platform. Select and provision a photogrammetry software platform. Enterprise platforms in this space produce orthomosaics (georeferenced, stitched aerial map images with uniform scale and accurate ground measurements), 3D point clouds, and digital surface models from overlapping drone imagery. Evaluate options on: GSD accuracy claims, supported input formats (image metadata, RTK/PPK correction data), output formats (GeoTIFF, LAS, OBJ), cloud vs on-premises processing, and integration with your downstream system (GIS, CAD, asset management).

Most platforms offer a trial period. Use Weeks 5–6 to process sample datasets (the vendor will have training datasets) and verify the output format integrates with your receiving system before you have real flight data depending on it.

File naming and storage architecture. Define a standard before the first flight:

• Folder hierarchy: YYYY-MM-DD / Site-Name / Mission-Type / Raw / Processed / Report
• File naming: YYYYMMDD_SiteName_MissionType_PilotInitials_FlightN_ImageNNNN
• Storage location: enterprise cloud (not a personal account) with defined access controls
• Backup protocol: at minimum one redundant copy, not on the same device as the primary

Downstream integration. Map the data flow from processed output to the consuming system. Who receives the report? In what format? Who has access? If the program feeds an asset management system, GIS platform, or EHS reporting tool, define the integration now — not after six months of data has accumulated in an orphaned folder.

Weeks 7–8: Operations manual and SOPs

The operations manual is the document that turns a collection of equipment and people into a program. Without it, every decision is made ad hoc. With it, a second pilot, a new program manager, or an external auditor can understand how the program operates.

A compliant commercial UAS operations manual typically covers:

• Program scope: Defined geographic areas, mission types, and exclusions
• Personnel: Named pilot-in-command, visual observers if applicable, maintenance authority
• Aircraft and equipment registry: Serial numbers, registration numbers, maintenance schedule
• Pre-flight checklist: Aircraft condition, airspace authorization, weather minimums, communications
• In-flight SOPs: Altitude limits, speed limits, no-fly zones, abort criteria
• Emergency procedures: Lost link response, aircraft failure, injury protocol, incident reporting
• Maintenance schedule: Battery cycle tracking, motor inspection intervals, firmware update log
• Data handling: File naming, storage, retention, access controls, client delivery protocol
• Incident log: Required documentation for any incident or near-miss

The FAA does not prescribe a specific format for a commercial UAS operations manual, but it is a required element of any BVLOS waiver application and a practical necessity for any program that expects to onboard a second pilot, operate at a client site, or face an insurance claim review.

Deliverable: A signed, dated operations manual version 1.0, stored in the program's document management system with version control.

Weeks 9–10: Pilot proficiency and simulation

Part 107 certification demonstrates knowledge, not stick-and-rudder proficiency. Especially for pilots who are new to a specific platform, structured practice before the first operational mission reduces crash risk and improves data quality.

A structured pre-operational training protocol:

1. Simulator time. Most enterprise platforms have a simulator mode. Practice standard mission patterns (grid survey, orbit, waypoint mission) in simulator before flying live aircraft.
2. Open-field orientation flights. Fly the aircraft in manual mode at an open, unconstrained site. Practice all emergency maneuvers (low battery return-to-home, manual landing, obstacle avoidance override).
3. Mission-type practice. Fly the specific mission pattern your program will use (façade inspection orbit, grid survey, patrol route) at the practice site before flying at the operational site.
4. Data quality check. Process imagery from the practice flights through the photogrammetry pipeline. Verify GSD (ground sample distance, the real-world area represented by one pixel) meets the program standard. Adjust flight altitude, overlap percentage, or speed to correct shortfalls.

Minimum proficiency standard before operational flights: three full practice missions with zero incidents and processed imagery meeting the defined GSD standard.

Weeks 11–12: First operational mission and verification

The first operational mission is a controlled deployment, not a full-scale production run.

Pre-mission checklist:

• LAANC authorization confirmed for the flight window
• Weather reviewed (wind below aircraft operating limit, precipitation clear, visibility adequate)
• Insurance confirmed active for the aircraft serial number and site
• Operations manual reviewed by PIC (pilot in command) for this mission
• Emergency contact list current (site safety, first responder, aviation authority for incident reporting)
• Battery state of charge verified, spares staged
• Camera/payload calibrated, memory cards formatted

During the mission: Complete the pre-flight checklist in full. Log actual take-off and landing times. Note any deviations from planned flight path or unexpected airspace encounters.

Post-mission protocol:

1. Download and back up all raw imagery within 24 hours of the flight
2. Process imagery through the photogrammetry pipeline within 3 business days
3. Deliver output in the defined format to the defined recipient
4. Complete the mission log entry in the operations manual
5. Document any deficiencies, near-misses, or improvement observations

Verification benchmark: The first mission is successful when the output data meets the defined quality standard, the consumer system receives and can use the output, and the mission log is complete. If any element fails, document the gap and address it before the second mission — not after six more flights of the same error.

90-day milestone checklist

Completing all nine milestones in 12 weeks is achievable with a focused program owner. Programs that slip are usually waiting on hardware delivery or insurance underwriting — start both in Week 3, not Week 6.

What comes after Day 90

The 90-day playbook produces a running program: certified, insured, with a data pipeline and an operations manual. Scaling from one site to multiple sites, or from VLOS to BVLOS autonomous operations, requires additional work — but it builds on this foundation.

For the procurement decision that precedes this playbook, see The drone vendor RFP: questions and red flags, which covers the technical and compliance questions to ask before selecting the platform and software your 90-day program will be built on.