A nurse in a busy emergency department has ninety seconds to establish whether the man in front of her is allergic to anything. A housing officer has a family in the lobby and one interview room. A claims adjuster has a caller who switched into Tigrinya three sentences ago. In all three cases somebody has to decide, immediately, whether to dial an interpreter or start a video call, and the decision is usually made on whatever the organisation happens to have set up rather than on what the situation needs.
That is a shame, because the two modes fail in genuinely different ways. Video remote interpreting services are not simply the telephone with pictures, and the telephone is not a degraded video call. Knowing which one suits the moment is one of the cheapest quality improvements a service can make.
What the telephone still does better
Speed is the obvious advantage. A phone line connects in seconds, needs no app, no bandwidth and no working camera, and it functions in a corridor, an ambulance, a doorway or a basement where a video signal will not hold. For short, factual exchanges, confirming an appointment, taking a medication list, checking an address, it is very hard to beat.
There is a subtler benefit too, and practitioners raise it more often than administrators expect. Audio only can improve candour. A patient discussing a termination, a tenant explaining domestic violence, an applicant admitting they cannot read their own language: some conversations go better when the interpreter is a voice rather than a face in the room, particularly in small communities where the interpreter might be a neighbour.
Where video earns its cost
The moment the exchange depends on anything visible, the phone becomes a liability. Demonstrating how to use an inhaler, pointing at a diagram, assessing whether a distressed person is oriented, watching a signature being given: these need eyes. Sign language, obviously, requires video absolutely, and this is the single most common reason organisations install the technology in the first place.
Video also helps in long, layered conversations. Turn-taking on a phone line is genuinely difficult with three or more participants, and interpreters routinely report that half their effort in a multi-party call goes on traffic control rather than language. Seeing who is about to speak removes most of that. The overview of language interpretation modes is a useful primer for anyone writing a policy on this, because the vocabulary matters when you come to write a contract.
Consecutive, simultaneous and the mistake in between
Mode is not the only variable. Remote work is overwhelmingly consecutive, meaning the interpreter waits for a pause and then renders what was said. That roughly doubles the length of an appointment, which is fine for a fifteen minute consultation and unworkable for a three hour board meeting. Simultaneous interpretation services solve the timing problem, but they need proper audio, usually two interpreters per language taking turns, and a platform built for it. Booking a consecutive interpreter for an event that needed simultaneous is the most common and most expensive scheduling error in this field. A useful account of how remote interpreting is changing global meetings sets out what that looks like in practice.
The obligations behind the choice
For public bodies and healthcare providers, none of this is purely operational. Language access requirements attach to the outcome, not the method, and an organisation that saved money on a bad connection and produced a misunderstood consent conversation has not saved anything at all. The National Council on Interpreting in Health Care publishes standards worth reading before writing an internal protocol, and a summary of the recent language access legislation gives the wider policy picture.
Choosing between a video connection and a phone call under pressure usually comes down to whether visual context is doing any real work. Sometimes it is essential and sometimes it introduces latency and failure points that a voice line avoids entirely. This examination of video remote interpreting is candid about where the format genuinely struggles.
One rule sits above all the mode selection: family members are not interpreters. Children least of all. It happens constantly because it is quick and because everyone in the room is embarrassed, and it produces the errors that end up in litigation.
Building a protocol people will actually follow
The organisations that do this well tend to keep the rule short enough to remember. Phone for anything under ten minutes and anything urgent. Video for sign language, for visual demonstration, for distress and for any conversation with three or more participants. Scheduled on-site interpreting for the longest and most sensitive appointments. Booked simultaneous provision for events.
Then make the phone number findable. The best procurement decision in the world is wasted if the sticker with the access code has fallen off the back of the handset, which is, more often than anyone would like to admit, the real reason a patient went home without understanding their discharge instructions.
Brief the interpreter, however short the call
Twenty seconds of context changes the quality of everything that follows. Say who you are, what the conversation is about, whether there is a third party in the room, and whether the news is bad. An interpreter who knows a diagnosis is coming will handle the pacing differently from one who is dropped into the middle of a sentence, and remote interpreting services are more sensitive to this than on-site work because there is no room to read.
Speak to the person, not to the interpreter, and in short complete sentences. Avoid the idioms that make sense only in your own workplace. Pause after each idea rather than after each paragraph. None of this is complicated, but it is rarely taught, and staff who have had ten minutes of training on it get noticeably better results than colleagues who have not.