How to investigate a UFO sighting

1. Record the observation before choosing an explanation

A useful first account describes what someone could actually see or hear. “Three white lights moved east over the roof” gives a researcher more to work with than “an alien craft appeared.” Keep both the witness’s interpretation and the observable details, but put them in separate parts of your notes.

Write down when the account was recorded as well as when the event reportedly happened. A recollection written years later is not the same kind of record as a note made that evening. Preserve uncertainty: an approximate time should remain approximate, and an unknown direction should remain unknown.

• Date, local time, time zone and how each was established.
• Observer position, viewing direction, duration and obstructions.
• Shape, colour, apparent motion, sound and changes over time.
• What was directly observed, what was inferred and what was learned afterwards.

2. Keep original material and its context

Keep an untouched copy of any original photograph, video or recording you are given permission to retain. Work from a duplicate when cropping, brightening or annotating. Record who supplied it, when it was received and whether the file has already passed through a messaging service or editing tool. A screenshot of a video is a different research object from the original video.

For an online account, save its publication date and source address alongside your notes. If the page quotes another report, follow that reference before treating the two pages as separate corroboration. Two websites can contain the same story without representing two independent observations.

3. Search details separately, then combine them

Begin with the strongest descriptive detail rather than the most dramatic interpretation. Try the location on its own, then a phrase such as “humming sound” or “bright lights.” Inspect the passages that matched before combining details. Suggested terms are starting points, not a promise that matching records exist.

Use Anomaly Graph’s search to add separate terms and compare their overlap. Check whether a phrase appears in the actual account, in a later discussion or in a list of explanations. A highlighted mention of “missing time” does not necessarily mean the witness reported it: the passage might explicitly deny it.

Record variants you deliberately tried, such as “hum” and “humming,” and keep the original query. Broader wording may find useful leads, but it may also introduce accounts describing different experiences. A relevance match is a reading aid, not an identification of the phenomenon.

4. Ask what would distinguish the alternatives

Make a short list of plausible explanations and the evidence each would require. For an apparent moving light, the useful questions may concern viewing direction, camera movement, distance cues or whether other observers saw the same thing. Do not assume you can calculate speed from an image when distance is unknown.

A comparison becomes stronger when it could change your mind. Note which observation would support an explanation, which would count against it and what information is missing. “Not enough information” is a legitimate finding. It is more informative than presenting an untested possibility as a conclusion.

5. Finish with a small, actionable research brief

Choose a few records worth following rather than forwarding an enormous result list. For each, include the matched passage, its source, the date and location precision, and the next unanswered question. Explain why the record matters to this investigation and what would make it a poor comparison.

For any field follow-up, seek permission before visiting private land or contacting a witness. Do not publish private addresses or assume someone wants to be approached because an old report names them. The aim is a respectful, testable investigation—not a persuasive story built from coincidences.

• What is the narrow question?
• Which passages support the comparison?
• Are the accounts independent?
• Which alternative explanations remain?
• What specific next observation or document would help?