Deciding Which Roles Can Be Remote

Organisations usually answer this at the level of the job title, which is why the answers are arbitrary and resented. The useful analysis is one level down.

Analyse activities, not roles

Almost no role is entirely remote-capable or entirely not. Take any job and list what the person actually does in a week, then ask of each activity: does this degrade at distance, and by how much?

A laboratory technician cannot run the assay remotely. They can write the report, review data and attend the planning meeting remotely. The role is not remote; a meaningful share of the work is.

That distinction matters because it produces workable arrangements instead of a blunt yes or no.

Four honest constraints

Physical presence required. Handling equipment, materials, patients, physical inventory. Not negotiable and not worth pretending otherwise.

Supervised access. Some regulated environments require work to happen on controlled premises. This is a real constraint, though it is often asserted more broadly than the regulation actually requires. Check the rule rather than the folklore.

Dense ambiguous collaboration. Work where the output emerges from a group repeatedly interrupting each other. Early product definition, crisis response, complex negotiation. This genuinely degrades at distance — not to zero, but noticeably.

Apprenticeship. People learning a craft by watching skilled practitioners work. This is the constraint organisations most consistently underestimate, and it applies mostly to juniors rather than to a role in general.

Three constraints that are usually excuses

"We need to be able to see them." This is a management capability problem stated as a work requirement. If you cannot tell whether someone is doing their job without watching them do it, the measurement is the problem.

"Clients expect it." Sometimes true. Frequently an assumption nobody has tested. Ask the clients.

"The culture." Culture is a set of behaviours, not a floorplan. If it only exists when people are in a room, it is not culture, it is atmosphere.

The apprenticeship problem deserves care

This is the one constraint that is real, underrated and unevenly distributed.

Experienced practitioners generally work well remotely — they have the context, the network and the judgement already. People early in a craft acquire those largely by observation, and remote work removes most of the observation.

The answer is not to refuse remote work for juniors. It is to replace incidental learning with deliberate structure: pairing on real work, explicit review as a scheduled activity rather than an afterthought, a named person responsible for their development, and defined checkpoints.

That structure costs management attention, which is why it usually does not happen, which is why juniors struggle remotely.

How to make the decision defensibly

Write the activity list for the role. Actual activities, from someone doing the job.

Mark each as location-independent, degraded at distance, or requiring presence.

Total the presence-requiring share. If it is small and concentrated, you have a hybrid role with defined in-person days for specific activities. If it is large and distributed through the week, you have an on-site role.

Say which activities drive the requirement. "This role is on-site three days because of X, Y and Z" is defensible. "This role is on-site" is a rule people work around.

Revisit when the work changes. Roles evolve; the analysis should not be permanent.

Consistency matters more than generosity

The largest source of resentment about remote arrangements is not restriction. It is inconsistency — two people doing similar work under different rules, with no explanation anyone can articulate.

A restrictive policy applied evenly is tolerated. A generous policy applied unevenly corrodes the organisation, because every arrangement becomes evidence about who is favoured.

If you cannot explain why two similar roles have different arrangements, you have a problem regardless of which arrangement is better.