CLVS common mistakes
CLVS · Understand the traps and how to avoid them.
Arriving at the deposition or hearing without having confirmed the room, seating layout, available power outlets, and ambient conditions, then improvising camera position and cable runs after participants are seated.
Candidates study equipment specs but treat logistics as someone else's job; the exam's pre-production items are answered from memory of gear rather than from a repeatable site-survey routine, so the planning step gets skipped in practice and mis-answered on paper.
What to do instead
Adopt a written pre-production checklist (room dimensions, witness/attorney/exhibit positions, outlet locations, window orientation, HVAC noise, backup power, parking/loading) and complete a remote or on-site survey before every job; practice applying it so questions about placement and contingency are answered from a known procedure.
Starting the record without a pre-flight verification: media capacity and formatting, battery/AC status on every device, and a time-of-day clock matched across camera(s) and the backup audio recorder.
Under time pressure the operator powers up and hits record; clock drift and mixed time sources only surface later when the transcript and video must be aligned, so the habit never gets corrected by feedback.
What to do instead
Run a fixed pre-flight sequence before the oath: format fresh media, confirm power source and spare batteries, set every device to a single reference clock and note any residual offset in the log, then verify record is actually rolling (tally light, timecode counting, media remaining) before swearing the witness.
Framing the witness so the shot is either a tight face-only close-up or so wide that the table, counsel, and unrelated activity dominate, and failing to widen or reposition when an exhibit is being marked or displayed.
Operators default to a flattering or 'cinematic' composition, or lock the camera and leave it, rather than framing for the record's purpose-identifying the speaker and capturing the exhibit being discussed.
What to do instead
Use a consistent medium close-up (head and shoulders, witness centered, eyeline positioned per standard practice) with the witness's hands visible when signing or handling exhibits; when an exhibit is marked or shown, widen or reframe to include it, then return to the standard shot, announcing nothing on the record unless required.
Leaving the camera in full auto (autofocus, auto iris, auto white balance, AGC) or using digital zoom and in-camera effects such as image stabilization gimmicks, date stamps burned in by the camera's own overlay, or picture 'looks'.
Auto modes look fine in a quick test and remove decisions under pressure, and digital zoom is mistaken for a way to reframe without moving the tripod; candidates also assume in-camera processing is harmless.
What to do instead
Shoot fully manual: manual focus set and locked on the witness, manual iris and white balance set to the room, gain off or fixed, optical zoom only with the tripod repositioned for tighter shots, and no in-camera effects or overlays-any required identification comes from the on-record slate and the log, not from the camera's menu.
Recording audio only from the camera-mounted microphone, or placing a lavalier without strain relief and clothing-noise control, and running no independent backup audio recorder.
Camera audio is the path of least resistance and usually 'sounds okay' in headphones, so the risk of a single point of failure is never felt; lav placement problems are only discovered when the recording is reviewed later.
What to do instead
Always run a primary lavalier on the witness plus an independent backup recorder with its own microphone, route the cable with strain relief and a loop under clothing, tape down the element, and confirm both sources are clean and separate before the oath is administered.
Setting audio levels once at the start and not monitoring: peaks driven into clipping during laughter or a raised voice, or levels so low that the transcript reviewer cannot hear soft passages, with no headphone check after breaks or seating changes.
Operators trust the meter they saw once and remove headphones because they are uncomfortable or because they want to listen to the room; auto-level circuits mask the problem until the damage is already on the file.
What to do instead
Monitor continuously with closed-back headphones on the recorded signal (not the room), set gain so normal speech peaks sit comfortably below 0 dBFS with a safety limiter, and re-check levels after every recess, seating change, or microphone swap; note the check in the log.
Placing the witness in front of an uncovered window or a bright wall so the face is silhouetted, or lighting the room so unevenly that a hard shadow falls across the face or the exhibit being examined.
Operators judge exposure by how the room looks to the eye rather than by contrast ratio on the recorded image, and they inherit whatever lighting the venue provides instead of controlling it.
What to do instead
Position the camera and witness to eliminate backlight (close blinds, flag or diffuse the window, use the room's existing light as fill toward the face), check the recorded image and waveform/zebra rather than the LCD alone, and bring a small battery or AC fill light to lift the face without creating a hotspot.
Failing to record a proper on-camera slate (case caption, date, time, location, participants, operator, and equipment used) and failing to capture on the record the administration of the oath, appearances, and any off-record or recess announcements.
The slate feels like redundant paperwork and participants start talking immediately; the operator responds to the room's momentum instead of controlling the opening of the record.
What to do instead
Before recording testimony, state or display the required identification information and hold it long enough to be read, record the oath and appearances in full, and whenever the proceeding goes off the record or resumes, state the time and reason on camera so the recording is self-documenting.
Maintaining only a partial log (missing start/stop times, breaks, objections, exhibit identifications, and equipment anomalies) and delivering a single copy of the file with no verified backup or documented chain of custody.
Logging competes with operating the camera in real time, and file handling is treated as an after-the-fact chore; candidates underestimate how much of a legal video practice is the record of the record.
What to do instead
Keep a contemporaneous log with time-of-day and timecode for every start, stop, recess, and notable event, duplicate and integrity-check (checksum) the original media before leaving the site, store the working copy separately, and document every transfer, format conversion, and delivery in a chain-of-custody record.
Altering the recording in post-trimming dead air, deleting a segment, color correcting, noise-reducing or otherwise 'cleaning up' the audio/video-or conversely commenting on camera, coaching a witness, and implying certification scope or authority the operator does not have.
Operators want to deliver a polished product and equate quality with editing; combined with the informal atmosphere of a small deposition, the line between technician and participant blurs, and certification marks or titles get overstated.
What to do instead
Deliver the unaltered original as the record, with any copies clearly labeled as copies and no content removed or processed beyond accepted duplication; stay silent and neutral on camera, never coach, prompt, or opine, and describe your credentials exactly by the scope of the certification held, disclosing any conflict or limitation to the retaining party.