Dictating site inspection notes on site keeps facts apart from conclusions. Note what you see in plain words, then put the opinion below it. Start with location and weather so each note carries its own moment.

Dictating site inspection notes starts before the camera and continues after the drive, and the fragments that survive cross-examination are the ones captured while standing at the property. Before the site visit, dictate the plan: the address, the access conditions, the areas to inspect, the person to meet, and the questions the matter needs answered at the property. The plan is the log's frame, and it is the document that later explains why the visit was limited to the east side, because a limitation that was planned is a methodology choice, and a limitation discovered during the deposition is an embarrassment. The plan is dictated before leaving the office, where the schedule is known.
At the site, the log captures what the eyes record. The sequence is location, condition, and the person who pointed, repeated for each observation point, and the dictation is spoken on the spot rather than transcribed from notes in the car. The on-the-spot dictation has a timing value that cannot be recovered later: the light, the weather, the sound, the order of what was seen, the moment when the condition changed as it was looked at. The drive-home version of the same observations is a summary that has already dropped the detail it did not think mattered, and the dropped detail is the one the other side's expert will have recorded.
After the drive, the log is closed with the missing items: what the site did not show, what the person at the site could not answer, and what the next visit needs. The closing list is the part of the log that reads like a professional document rather than a diary, because it shows the visit's boundaries honestly. An investigation log that states its gaps is a log a court trusts; the log that presents a complete narrative of an incomplete visit is the log that the cross-examiner turns into the exhibit that costs the finding. Dictate the closing list in the car, while the drive is still the visit's context.
Each observation gets three spoken components at the spot. The location is the anchor: the building and floor, the room or the area, the distance from a fixed reference such as the loading dock, and the direction of view. Location details are the ones that get challenged first, since a finding attached to the wrong place is a finding the expert can explain away. The condition is what the eye sees: the crack's length and width in the terms the measuring tools used, the stains, the state of the surface, the visible state of the equipment. The condition is dictated in the vocabulary of the observer, and the observer records what was measured, not what was estimated.
The person who pointed is the third component and the one that connects the observation to a source. The visitor, the site supervisor, the engineer, or the tenant who indicated the area becomes the link between the fact and the person who can testify about it, and the log names the person and the person's role at the moment of the pointing: the facility manager stated the drain had not been cleaned since the spring. The source is not evidence of the condition by itself; the source is evidence of where the knowledge came from, which is the question the deposition will ask first.
The on-the-spot sequence also captures the direction of the observation. The log states who was at the observation point, what was seen, and from where, since a condition that reads differently from the parking lot than from the plant floor is a condition with a perspective problem. When the inspection is done in the presence of the opposing party or the opposing expert, the log records their presence and their comments, because a condition observed in the other side's company is a condition whose description will be compared to theirs. The spot note is the shortest document in the file and the one most often quoted, because it is the only one written next to the object.
The sentence pattern that shields findings separates what was seen from what was concluded, and the separation is spoken in the note itself. The pattern is two sentences in a fixed order: the fact sentence, which states only the observation in the words a camera would support, and the inference sentence, which begins with the marker that labels it: inference, and then the conclusion. The crack in the concrete is three meters long. Inference: the crack is consistent with a settlement at the foundation. The fact can be attacked only if it is wrong; the inference can be attacked as an argument, and an argument that is labeled is an argument that looks like an honest method rather than a hidden bias.
The pattern's protection works because the two sentences can be separated under questioning. The witness confirms the fact easily, then offers the inference as the professional view it is, and the cross-examiner who tries to blur the two is interrupted by the log itself, which is already sharp about the distinction. The log that mixes them gives the expert the phrase the whole case turns on: my notes say the building is settling, when the note actually says the crack is three meters. The difference is the difference between an observation and a diagnosis, and the pattern is what keeps the diagnosis from dressing as an observation.
The pattern also disciplines the drafter's own reasoning. The inference rule is followed by the check: state the fact, state the inference, then state the alternative inference that the same fact supports, and record the alternative in the note. The alternative is what makes the primary inference credible, because a finding that admits its rival is a finding that was made on evidence rather than on preference. The alternative inference is normally one sentence, and it is the sentence that the report's reader, or the judge, or the expert for the other side, tests first. The test passes when the log has already shown it.
The photos are the site's visual memory, and they are only as good as the log's links to them. The link is simple: the dictation states the photo number at the moment the photo is taken, in the same note as the location and the condition, so the observation and the image share one record. The camera in the note is the picture frame in the log, and the two cannot be separated by the next reader. Dictate the link at the capture moment, not at the end of the visit, because the caption parked in the barrel of the photo file is a caption that was written by the person who never saw the site.
The linked log is what closes the distance between the finding and the exhibit. When the report cites the crack, the citation goes to the log line, and the log line goes to the photo, and the photo shows the crack in the location the note describes. An exhibit that cannot be traced to one note and one image is an exhibit that the other side can call a reconstruction. The trace is the chain for visual evidence, and it is built by the spoken link at the moment of the shot rather than by the best guess at the time of the exhibit list.
The photo pass also records what the image does not show. The log states the photo's context: taken at 9:40, in natural light, from the northeast corner, including the guardrail in the frame. The context is what lets the expert confirm the photo's evidentiary value weeks later, when the memory of the light has gone. The note also flags the absence: the photo does not show the far side of the berm, and the flag is the sentence that keeps the exhibit honest. The image plus the context plus the absence is the complete visual record, and the complete record is the one that survives the objection that the photograph is incomplete.
Thirty days later, the site notes get their cold read, and the read is against the questions the deposition will ask. Open the notes, set aside the memory of the visit, and read only what was written, asking the sequence: where was this, what was seen, who said it, and what did the note say was inferred. The cold read is the test of the note's self-sufficiency, and its product is the list of sentences that the next reader cannot reconstruct without the visit. Those sentences are the note's gaps and they are the places where the deposition will push.
The cold read usually surfaces three classes of inadequacy: the observation that lacks a measurement, the inference that lacks its alternative, and the location that lacks a reference. Each class gets one line in the follow-up list, and the follow-up list, dictated the same day, becomes the plan for the return visit if the matter requires one. The gap list is the honest product of the thirty-day pass, and it is worth more than the confidence of a note that reads smoothly, because a smooth note is a note that has already begun to interpret its own content. The pass converts the note from a record into evidence about the record.
The read against the deposition questions closes by testing the note's first line, the one most likely to be asked: what did you see that day, and what did you conclude from it. If the note answers both parts separately and cites the photo and the person for each, the note is ready. If the note answers only one, the answer is the line the deposition will spend twenty minutes on. The re-read is also the moment to fix the note's format, and the fix is dictated as a correction rather than a rewrite, so the original line and the corrected line both exist. The corrected line is the note's honest history, and the honest history is the strongest part of the file.
On the spot, for every observation. The drive-home version is a summary that has already dropped the details it did not think mattered, and the dropped detail is the one the other side will have recorded. On-the-spot dictation keeps the observation attached to the location, the light, and the person who pointed, and the timestamps make the notes contemporaneous, which is the fastest defense to a reconstruction accusation.
Use the two-sentence pattern. The fact sentence states only what a camera would support, and the inference sentence starts with the marker inference and then states the conclusion. Then add the alternative inference the same fact supports. The labeled separation means the fact can be confirmed easily under questioning while the inference stands as the professional view it is, and the alternative makes the primary inference credible.
Because the person is the link between the observation and the source of knowledge, and the deposition will ask where the knowledge came from. The log names the person clearly, the facility manager or the site supervisor, at the moment of the pointing, so the note carries the attribution rather than the claim. Attribution also shows the observation was a documented exchange, not the inspector's later reconstruction.
Dictate the photo number into the note at the moment the photo is taken, in the same line as the location and the condition, and describe the context: time, light, angle, and what is in the frame. Flag what the image does not show. A finding then cites the log line, the log line cites the photo, and the exhibit chain is traceable from report to image.

Record it at the spot, link the photos, and re-read cold in thirty days.
Use it across macOS, Windows, iOS, and Android