Start with the decisions the data must support
Write down the questions that designers, contractors and owners expect the survey to answer. These might concern the relationship between floors, a connection to an existing structure, or the available space around equipment. Identify the features that must be measured and the areas that are only useful as context.
For each question, agree an output and an acceptance method. A designer reviewing overall building geometry may need a registered cloud and sections. A contractor preparing a particular connection may need a focused survey with additional checks. Treat those as different requirements in the brief.
Make access and coverage part of the scope
A room list and marked site plan help turn an abstract scanning request into a fieldwork plan. Include occupied spaces, roof access, service areas, external boundaries and interfaces with adjoining structures. Record who can arrange access and whether work must occur around occupants or operations.
Ask the survey team to identify areas likely to be obstructed before mobilising. Agree how furniture, stored materials and temporary works will be handled. If an area cannot be reached, decide whether to reschedule it, capture it by another method, or deliver an explicitly incomplete record.
Check the assembled survey
Scans from different positions need to be brought into a common spatial framework. Registration software can report the fit between scans or targets, but the project still needs a review of the assembled result. Leica describes registration reports that include statistics for individual constraints, which makes them useful supporting evidence.
The review should cover the critical interfaces identified in the brief. Ask for checks across floors, building wings and transitions between interior and exterior capture. A visually convincing overview is insufficient when the decision depends on a small area that may have weak coverage.
Package information for the receiving team
Confirm the receiving software and intended workflow before the final export. Agree whether the team needs the complete dataset, smaller zones, drawings, or a model created from the measurements. Include a small sample handover early enough to resolve compatibility problems without delaying the main delivery.
Every package should explain its units, coordinate reference, survey dates and revision. Add a coverage plan and a short register of exclusions. If simplified or reduced files are supplied for everyday viewing, identify the fuller record from which they were derived and explain where it is retained.
Keep the record useful as work progresses
Name an owner for the survey baseline and define how later changes will be recorded. A cloud captured before demolition cannot establish conditions after walls are removed. Ask project teams to reference the survey revision when they raise queries or issue design information.
Acceptance should end with a practical use check. Can the recipient open the files, locate the relevant area, understand the limitations and make the agreed measurements? Resolve missing information while the capture team still has the site context. This is more valuable than treating file delivery as the end of quality control.
Common questions
Does a large point cloud mean the survey is accurate?
No. File size mainly tells you how much data is present. Assess coverage, control, registration evidence and checks relevant to the intended use. A large dataset can still omit a critical surface or contain an alignment problem.
Should every project receive a BIM model?
Only when the planned work needs one. A registered cloud, targeted sections or measured drawings may answer the questions. Specify a model separately, including the elements and information it must contain.
Further reading
This article was rewritten in October 2026 around the subject of the original Geobiro post. The original publication date is retained; the update date identifies the current text.