The lights come back on. The final guest leaves. Equipment cases close—but the investigation is not finished. What happens next determines whether an unusual moment becomes useful evidence or simply another unexplained clip.
Public investigations are remembered for the hours spent inside the location. The quieter work begins afterwards: accounting for the equipment, protecting the recordings, rebuilding the night and testing every promising moment against the ordinary.
This is not the dramatic part of paranormal investigation. It is the part that gives the drama context.
Pack down without losing the record
A careful pack-down is more than putting devices back into cases. Each recorder, camera, trigger object, memory card and power supply should be accounted for. Unexpected damage, flat batteries, moved equipment or interrupted recordings should be noted while the circumstances are still fresh.
The same applies to the location. Temporary equipment should be removed, furniture returned as agreed and any issue reported to the venue contact. Respect for the building continues after the session ends.
Notes made at this stage can later explain why a camera stopped, why a device changed position or why part of a recording is missing. Memory is not a reliable substitute for a contemporaneous record.
Protect the original files
Recordings should be transferred as soon as practical and protected before anybody begins trimming, amplifying or converting them. The original is the closest available record of what the device captured; review and enhancement belong on separate working copies.
The Scientific Working Group on Digital Evidence describes data integrity as a responsibility that continues throughout the evidence lifecycle. Its current collection guidance recommends documenting file details, creating working and archive copies, and using cryptographic hashes to establish a baseline that can later help verify whether data has changed. Read SWGDE 18-F-002-2.0.
A hash does not prove that a recording is paranormal or authentic in every sense. It has a narrower and valuable purpose: comparing the file with a previously calculated digital fingerprint.
Build one investigation record
A night may produce recordings from several cameras, audio devices and investigator phones, alongside written notes and observations. Viewed separately, each tells only part of the story.
The review record can bring together the original filenames, devices, locations, operators, approximate times and known interruptions. A shared event—a spoken marker, a door closing or a distinctive sound—can help align recordings when device clocks do not agree.
The aim is not to create certainty where none exists. It is to preserve enough context to understand what happened before, during and after a reported event.
Review the ordinary before the extraordinary
Potential evidence should be approached as an observation, not a conclusion. A voice-like sound, light anomaly or equipment activation earns investigation—not an automatic paranormal label.
Reviewers should check movement, speech, clothing noise, handling, traffic, wildlife, airflow, plumbing, vibration, reflections, infrared illumination, radio transmission and equipment behaviour where relevant. Video may reveal the source of an audio event. A second recorder may show that a supposed response travelled from another room.
An explanation does not spoil the investigation. It removes a weak claim and protects the credibility of anything that remains.
Keep enhancement transparent
Amplification, equalisation and noise reduction can make a quiet passage easier to examine. They can also alter the balance of a recording, emphasise artefacts and shape what a listener expects to hear.
SWGDE's digital-audio guidance recommends obtaining the original where possible, conducting examinations on a working copy and documenting relevant decisions. It also notes that processing applied before examination may limit what can be achieved later. Read SWGDE 20-A-001-2.0.
If an enhanced version is shared, the audience should be told what changed. An edited presentation copy should never quietly replace the source recording.
Write down disagreement
Evidence review does not require every investigator to hear the same words or accept the same explanation. In fact, disagreement is information.
Ambiguous audio is stronger when reviewers hear it without subtitles or a suggested phrase. Their responses can be recorded independently before being compared. If interpretations differ substantially, the clip should be described as ambiguous rather than forced into a single reading.
Useful conclusions may include explained, likely contamination, ambiguous or unexplained. None of these terms turns an observation into proof of a paranormal cause.
Report only what the evidence can carry
A responsible report separates what was observed from what is inferred. “A voice-like sound appears after the question” describes the recording. “A spirit answered” claims an identity and cause the recording alone cannot establish.
Guidance for digital-audio authentication says conclusions should be supported by the analysis, should not be overstated and should avoid claims of absolute certainty. Read the SWGDE authentication guidance.
Public presentation should retain enough context for people to form their own view: the unlabelled clip first, the surrounding recording, who was present, whether the event was heard live, what other devices captured and which explanations were tested—subject always to privacy and venue agreements.
Close the investigation properly
The final task is not finding a dramatic clip. It is leaving a defensible record of what was captured, how it was handled, what was tested and what remains unknown.
Some events will be explained. Some will remain too uncertain to interpret. A small number may resist the reasonable explanations available at the time.
“Unexplained” is not a failure, and it is not proof. It is an honest boundary around the information we have.
The investigation does not end when the building falls quiet. It ends when the record is protected, the alternatives are tested and the conclusion says no more than the evidence allows.
Expect Nothing. Question Everything.