The Complete Survey Project Workflow

Every professional drone survey follows the same sequence. Skip a step and you end up with bad data, unhappy clients, or both. Here is the complete workflow from first contact to final delivery.

Step 1: Define the Project Objectives

Before you touch a drone, nail down what the client actually needs. “We need a map” is not an objective. Push for specifics:

• What are you measuring? (Distances, areas, volumes, elevations)
• What accuracy do you need? (Centimeter-level for engineering, meter-level for general overview)
• What format do you need the deliverables in? (GeoTIFF, DXF, PDF report, point cloud)
• What is the site area and terrain type?
• What is the timeline?

Write this down. Email it back to the client for confirmation. This prevents scope disputes later.

Step 2: Site Assessment

Review the location before you show up. Check satellite imagery for terrain features, obstacles, and access points. Identify the nearest roads and safe takeoff areas.

Check airspace classification. Mining sites and construction areas sometimes sit near controlled airspace, which means you need LAANC authorization before flying.

Review any site-specific safety requirements. Industrial sites often require PPE, safety briefings, and escort personnel.

Step 3: Flight Planning

Use mapping software (DroneDeploy, Pix4D Capture, Litchi) to plan your flight:

• Draw the survey boundary
• Set altitude based on required ground sampling distance (GSD)
• Configure overlap: 75-80% front overlap, 65-70% side overlap for general mapping
• Add crosshatch lines for tall structures or complex terrain
• Set the camera to capture at regular intervals based on speed and altitude

Plan battery changes. Calculate how many flights you need to cover the full area. Bring extra batteries.

Step 4: Ground Control Points

For surveys requiring centimeter-level accuracy, place ground control points across the site before flying.

Place 5-10 GCPs distributed across the survey area, with at least one near each corner and one in the center. Measure their positions using an RTK GNSS rover. Mark them with visible targets (chequerboard panels or spray-painted crosses on pavement).

If the client does not need survey-grade accuracy, you can skip GCPs and rely on the drone’s built-in GPS. Just be clear about the accuracy tradeoff in your proposal.

Step 5: Fly the Survey

Launch from a flat, clear area with good GPS reception. Monitor the drone throughout the flight. Watch for:

• Battery levels (land with 20% reserve)
• Wind gusts that could blur images
• Obstacles not visible in your flight planning software
• GPS signal quality (at least 15 satellites for reliable geotagging)

If a flight is interrupted (battery swap, weather), note exactly where it stopped so you can resume from the correct location.

Step 6: Data Processing

Upload your images to photogrammetry software. The processing pipeline:

1. Image alignment: Software matches features across overlapping photos
2. Point cloud generation: Aligned images produce a dense 3D point cloud
3. Mesh creation: The point cloud becomes a triangulated mesh
4. Orthomosaic: Camera perspectives are corrected to produce a flat, measurable map
5. DEM/DSM generation: Elevation data is extracted from the 3D model

Processing time depends on image count and computer power. A 500-image survey might take 2-4 hours on a capable workstation.

Step 7: Data Analysis

Once processing completes, extract the specific deliverables the client requested:

• Measure distances and areas on the orthomosaic
• Calculate stockpile or cut/fill volumes from the DEM
• Generate contour lines for topographic maps
• Create cross-section profiles for road or pipeline planning

Verify your measurements against known reference points. If you placed GCPs, check that your model aligns with their surveyed coordinates within the expected tolerance.

Step 8: Quality Control

Before delivering anything, run these checks:

• Compare your orthomosaic against satellite imagery for gross alignment errors
• Verify GCP residuals are within acceptable range (under 2-3 cm for RTK surveys)
• Check for gaps in coverage, especially at the edges of the survey area
• Confirm volume measurements by cross-checking with a second method or known quantities

If something looks wrong, re-fly the area. Delivering bad data destroys trust faster than delivering late.

Step 9: Deliver Results

Package your deliverables professionally:

• Orthomosaic as GeoTIFF for GIS users, JPEG for general viewing
• DEM as GeoTIFF for elevation analysis
• Point cloud as LAS/LAZ file for 3D visualization
• Volume reports as PDF with clear methodology and confidence intervals
• KMZ file so the client can view results in Google Earth

Include a brief report summarizing the survey parameters: date, weather conditions, number of images, GSD, processing software, and accuracy results.

This workflow becomes second nature after 10-15 projects. The first few will feel slow. That is normal. Speed comes from repetition, not from skipping steps.