Most companies never answer that question — they pick a number of days and apply it to everyone. Here's a cleaner way: decide per role, using what the job physically requires, not what feels safer.
Your return-to-office policy isn't a culture decision. It's a hiring decision, and you've already made it.
Draw a 45-minute radius around your office. That circle is now your entire talent pool, for every role you're hiring. Every engineer outside it, including the great onece, is invisible to you.
That's a legitimate trade. Some roles genuinely need hands on hardware. Some teams really are better in a room together. But it should be a decision, not a default. Right now most RTO policies are one blanket rule applied to twelve different jobs, and the hiring pipeline quietly pays for it.
If you mandate five days onsite and still can't fill the role after four months, the market isn't broken. Your radius is.
Start with one question
Does this role touch physical things — racks, cabling, devices, badge access, a specific client site?
If yes, some onsite presence is real. If no, the seat is a habit, not a requirement. That single test resolves most of the argument before anyone mentions culture.
Role by role
|
Onsite is genuinely real |
Onsite is assumed but rarely required |
|
Field techs and smart hands |
Tier 1 and Tier 2 remote support |
|
Data centre and hardware work |
NOC monitoring and after-hours coverage |
|
Device imaging and depot |
Cloud and infrastructure engineering |
|
Anything tied to a client's location |
Scripting and automation |
|
Closed-network and air-gapped environments |
Most security operations work |
The left column is a staffing constraint. The right column is a preference — which is fine, as long as you know which one you're paying for.
Then ask what the office is actually for
If the answer is “so juniors learn from seniors,” say that — and schedule it. Named days, named people, a real reason to travel. Two intentional days beat five accidental ones, and candidates can tell the difference in the first interview.
The seat and the job are not the same thing
A pattern worth naming: teams lose the right engineer over a commute, a school run, a partner's job in another city — then file it as a compensation problem. It was a geography problem wearing a compensation costume.
The work itself hadn't moved. Same tickets, same escalations, same systems, all reachable from anywhere with a decent connection. What changed the outcome for the teams who got this right was separating two ideas: the seat is where the work is done; the job is whether the work gets done, on time, covered, with someone accountable for it.
Once you separate those two, the person who couldn't relocate is suddenly hireable again.
Run the exercise this week
01 List every open and filled IT role. All of them, not just the ones you’re hiring for.
02 Mark each one physical or not physical, using the touch test.
03 For every non-physical role, write down what the office is for. If you can’t, drop the requirement.
04 Re-post the roles that changed. Watch what happens to applicant volume.
Build your team in 30 days
If the exercise leaves you with a role that doesn't need a seat — that's the part we solve. Ibex builds dedicated L1 and L2 support, after-hours coverage without overtime, and NOC, cloud and SecOps talent that was never inside your commute radius to begin with.
info@goibex.it · 844-737-1477
