Return to Office Mandate: A Practical Guide for Scattered Teams

By [Author Name] · [Date] · [X] min read · [Table of Contents]

TL;DR

• A return to office mandate is a hardware event before it's a culture event, and it moves kit in both directions at once.
• If you only ever issued laptops, skip most of this. You've got a desk problem, not a logistics one.
• The work splits four ways: counting what you own, deciding on duplication, moving it, and recovering it from people who quit.
• Three cohorts behave differently. People near the office, people who commute, and people who won't come at all.
• Decide one setup or two before you announce. That single choice drives most of the budget.
• Done properly, nobody spends week one hunting for a monitor that was in a storeroom all along.

You announce three days a week from the first of the month. Nobody blinks. The board paper covered desks, lease costs and an attrition estimate, and it didn't mention that 412 people are sitting at home on company monitors, docks and chairs, every one of which is now in the wrong building or about to be bought a second time.

These aren't edge cases. A company that went remote in a hurry bought kit in a hurry, usually on expenses, usually without anybody writing down where it went or who signed for it. Another ran four years of ad hoc replacements nobody logged. A third shipped chairs to 200 homes and has no idea which model went where.

Recommended Tools

Best tools for HR Operations

B BambooHR HR management system R Rippling HR & IT automation W Workday Enterprise HRIS

View all recommended tools

The real issue isn't the cost of the mandate. It's that the hardware consequences land six weeks after the announcement, when the decision is made and the budget is already set. This is what equipment lifecycle planning is supposed to solve.

When You Don't Actually Need an Equipment Programme

Stage one: manual is genuinely fine. You issued laptops and nothing else. People carry them in. There's no monitor to move, no chair to argue about, and the whole exercise is a desk count. Don't build a programme for that.

Stage two: friction appears. You're mandating three days and people start asking whether they keep the home screen. Somebody's already bought a second monitor on expenses. That's the signal, and it usually shows up in week two.

Stage three: it becomes a liability. Staff resign over the policy and leave holding kit nobody's chasing. Devices with company data sit in spare rooms. Now it's a data question, not a furniture one.

Stage four: the edge case. You hired people as remote workers with no office within commuting distance. The mandate can't apply to them and they need a written answer before the announcement, not after. Get this wrong and you'll lose them in a fortnight.

What HR Teams Actually Need from Equipment Planning

> "Do I bring my stuff back?" > Say yes or no before you announce. > Left unanswered, people assume the expensive version, and they tell each other.

> "Am I keeping the home setup?" > Usually the laptop travels and the second screen stays put. > This is the single biggest line in the budget and it gets decided by accident.

> "Who moves a 27-inch monitor on a train?" > Nobody. Plan for that. > If you expect kit back, somebody without a car physically can't comply.

> "What happens to the chair?" > Most companies leave it. Decide deliberately rather than discovering it. > Chairs cost real money and resell for almost nothing.

> "Does this change if I quit?" > Everything issued stays company property, and some people will quit. > Say it before the announcement or you'll be saying it in an exit email.

There's a sixth question nobody voices. People spent four years building home setups that are frequently better than anything the office will give them back, because they were bought one at a time against a real need rather than in bulk against an average. Say so plainly. Pretending a hot desk and a laptop riser isn't a downgrade is what turns a grumble into a grievance.

The Three Types of Equipment Position

One setup, carried. What it is: a laptop that travels, screens provided at the office, nothing replaced at home. When it's right: three days or fewer, decent office stock, cost pressure. When it fails: people who built good home setups feel robbed, and they are.

Two setups, funded. What it is: a full arrangement in both places, nothing carried. When it's right: senior or specialist roles where the kit genuinely matters. When it fails: at scale. This is what happens by default if you don't decide, because nobody volunteers to hand back a monitor.

One and a half. What it is: laptop travels, office provides screens, home keeps what it has and gets no replacement when it breaks. When it's right: almost everyone, almost always. When it fails: slowly. In three years you've got an unmaintained estate in people's houses and no plan for it.

I've watched four companies land on the third option after trying one of the first two. Getting there deliberately on day one saves about a year of argument.

The Three Cohorts, and Why One Average Is Useless

Every mandate splits your people three ways, and a single cost-per-head figure is wrong for all of them.

There are the people who'll comply and live near the office. They need a working setup in the building, and whether that means moving their home kit in, buying new, or using whatever's already sitting on the third floor depends entirely on an inventory you probably haven't done. Their cost is the most variable of the three.

Then the people who'll comply and commute a distance. They'll carry a laptop and want a desk that isn't permanently theirs, which is where the dock and monitor demand comes from. A hot desk with no screen is the single most common complaint after any mandate lands, and it converts into a purchase order within about a fortnight.

And the people who won't comply at all. Some resign, some negotiate an exception, some quietly don't show up and wait to see whether anybody notices. Every resignation in that group is a retrieval, and those cluster tightly in the eight weeks after the date rather than spreading themselves politely across the year the way your normal attrition does.

Cost per head between the first two differs by a factor of three or four in most companies I've looked at, which means an averaged figure is simultaneously too high for one group and too low for another. Size them separately. Even roughly. The number becomes defensible the moment you stop blending them.

Counting What You Own, Properly

Everyone says know your inventory. Nobody says how. So here's the method that actually produces a number you can defend in a budget meeting, rather than the vague reassurance most teams bring to one.

Start from procurement records, not the asset register. The register is downstream and it inherits every gap upstream of it. Pull every hardware purchase since the remote period started, and include things bought on expenses and individual cards, because that's where peripherals hide. What you'll usually find is that laptops went through procurement and everything else didn't, so your laptop count is roughly right and your monitor count is badly wrong.

Then sample. Don't survey. A full survey of 412 people gets you a 60 per cent response rate and three weeks of chasing, and you'll still be extrapolating at the end of it. Pick 20 people across teams, ask them directly what's physically in front of them, and you've got a usable multiplier in two days. You're after the ratio between what records claim and what exists. That ratio holds well enough across a population.

Record four things per item. What it is, who's got it, roughly when it went out, and whether it still works. That last one gets skipped constantly and it changes the plan, because a four-year-old monitor somebody has been tolerating isn't worth paying to transport anywhere.

Three things turn up almost every time. Kit issued to people who left and was never recovered, which is the one that stings. Kit bought twice for the same person because a replacement request got raised, fulfilled, and never closed against the original asset. And a whole category nobody tracked at all, usually headsets and adapters, individually cheap and collectively a number that surprises people.

Write the gap down as a figure. A sentence like "the register says 410 monitors and the sample suggests about 310 of them exist and still work" survives a budget conversation with a finance director who wants to know why you need money. "The register's probably out of date" doesn't.

The Office Has Its Own Hardware Problem

Mandate planning treats the office as a fixed thing that already exists. It usually doesn't.

A building that sat at 30 per cent occupancy for four years has been quietly stripped. Desks got consolidated. Monitors migrated to whoever needed one. Docks vanished into drawers and then into homes. Cables went first and nobody replaced them. So a floor that nominally seats 300 actually equips about 140, and you find this out in week two rather than week minus six.

Then there's the dock problem, which is specific and annoying. Every shared position needs a dock and cable that work with every laptop model in the building. Mixed estates break this immediately. A floor running three laptop generations and two operating systems needs either universal docking, which costs more per desk, or signage nobody reads. Standardise the dock rather than the laptop. It's cheaper and it works.

And count the things that aren't desks. Meeting room screens and cables cause more first-week frustration than anything else, because the first thing a returning workforce does is hold meetings, and those rooms have been dormant for years.

Exceptions, and Keeping Them Fair

Every mandate generates exception requests. How you handle them decides whether the policy reads as reasonable or arbitrary, and the hardware side is a real part of that.

Accessibility comes first and deserves separating from everything else. A meaningful number of people have a home setup that functions as an adjustment: a particular chair, a specific screen arrangement, software configured for them, a quiet room. Requiring attendance without addressing that asks somebody to work without something they need. Routing it through the same queue as a general preference is how this goes wrong, and it goes wrong often. Send these straight to whoever handles adjustments, and treat the equipment question as part of the adjustment rather than as a desk allocation.

Caring responsibilities and commute come second. They're genuine, they're not hardware questions, and they produce one anyway. Somebody granted two days instead of three still needs a working setup at home as well as a desk in the building, which means the exception carries an equipment cost that nobody counted when it was approved. Record it.

Role-based exceptions come third. Some jobs can't be done in an open office and some can't be done outside one. Deciding these by role rather than by individual request removes a lot of case-by-case argument and holds up better when challenged.

One rule saves enormous difficulty. Don't require anybody to return kit while their exception request is open. Taking a monitor from somebody who's asked for an adjustment and hasn't had an answer is remembered long after the policy changes.

Sequencing the Announcement

The order matters more than the wording, and the usual order is close to the worst available.

What normally happens: the policy goes out, questions arrive, and the hardware answers get issued over the following fortnight as they're decided. That reads as improvisation because it is, and it undermines the parts you planned well.

Put the decisions that affect individuals in the same communication as the policy. People need four things at once. The date. Whether it applies to them. What happens to the kit they're holding. Who to ask. Anything missing gets filled in by speculation within a day, and the speculated version travels faster than your correction and is always worse.

Give the exception route in the same message, with a name on it. Announcing a change and promising a process later generates a fortnight of private anxiety among exactly the people most likely to have a legitimate case.

Office logistics can wait. Which desk, which floor, what the booking system is. Those matter on the day and not before, and trying to settle them before the inventory is finished is precisely how teams end up announcing seating arrangements they have to reverse a fortnight later.

Set one date for the hardware position and hold it. Three revisions over six weeks does more damage than one position some people dislike.

How to Choose: Five Questions Before You Talk to Any Vendor

What do you actually own, and where? Not what the register says. If you can't produce a list with a confidence level attached, that's your first job and it's bigger than the mandate. Most registers are wrong by a quarter.

What's already in the building? Walk the floor. Offices that sat at 30 per cent occupancy for four years have been quietly stripped, and they've also got monitors in cupboards nobody logged. Both are true at once.

Assigned desks or shared? This is the biggest single lever and it gets decided on culture grounds without anybody pricing it. Shared desks at three days a week need roughly 60 per cent of headcount in setups. Assigned needs 100.

What's your retrieval capacity this quarter? Not a percentage. Devices per week. If a mandate produces 23 extra leavers in two months and you comfortably handle five a month, you have a queue before you have a problem.

Who owns the hardware side? Mandates get announced by leadership, planned by facilities, paid by finance. The equipment work lands on HR and IT, who in most companies found out about the policy at the same moment as everybody else and now have six weeks to move several hundred items. Name somebody first.

Six Platforms Worth Knowing

RemoAsset Best for: distributed teams that need kit moving in both directions across borders. Why companies choose it: procurement and retrieval in 80+ countries, which matters when a mandate produces inbound office moves and outbound resignations in the same eight weeks. Pricing is published, from $1,499 a year on the Subscription tier for 50 to 5,000+ employees, including a 60 per cent cut on retrieval charges. It's our own product, which you should weigh accordingly. Where it struggles: there's no network discovery, so it won't find hardware you forgot you owned, which is exactly what a mandate exposes. No software licence management either. And only the Subscription tier publishes a figure, with pay-as-you-go quoted instead.

Workwize Best for: companies that want kit held and redeployed rather than shipped from source each time. Why companies choose it: own warehouses, which is a genuine structural advantage when returned equipment needs somewhere to live before it goes out again. Where it struggles: no published rate card. The figures on its pricing page are savings claims rather than prices, so you can't model it without a conversation. Its live domain is goworkwize.com now.

Firstbase Best for: teams building a fleet from scratch rather than inheriting one. Why companies choose it: the subscription shape is clean and the supply side is strong. Where it struggles: it publishes nothing on price. The model assumes it supplied the kit, which makes it awkward when you've got 400 devices already scattered across homes and you want them collected, not replaced.

GroWrk Best for: companies that want scoping flexibility rather than a fixed package. Why companies choose it: it'll shape the service around what you actually need. Where it struggles: no rate card. Its pricing page runs a competitor cost estimator, and those numbers are its own estimates rather than anybody's published prices. Don't carry one into a board paper.

allwhere Best for: teams handing the whole problem over. Why companies choose it: managed service, low internal lift. Where it struggles: publishes nothing, so there's no way to compare before a call. That's a real cost when you're trying to budget a mandate in three weeks.

Deel IT Best for: companies already employing international staff through Deel. Why companies choose it: equipment bundled with employment, one relationship. Where it struggles: weaker standalone, and no published pricing. If Deel isn't already your employer of record it's a strange place to start.

The Decision Table: Where Each Approach Actually Fits

Most teams sit in two or three of these rows simultaneously. Start with whichever one is actively on fire.

What Comes Back, and What to Do With It

A mandate that pulls kit out of homes creates a second problem roughly a month later, and it arrives as physical objects rather than as a line in a plan. Boxes turn up. Nobody decided what any of it is for, so it goes wherever there happens to be space.

The instinct is a cupboard, and that's how you end up with 80 four-year-old monitors nobody will issue and nobody will throw away. Returned kit splits three ways, and each way needs a decision made before the first box lands rather than after the fortieth.

Redeploy covers anything with useful life that matches what you currently issue. That's the valuable pile. It genuinely offsets purchasing, but only if somebody checks, cleans and records each item as it arrives rather than stacking it against a wall to deal with later. Kit that goes into storage unchecked is kit nobody trusts enough to issue, so it stays there until it's obsolete.

Hold as spares covers working items a generation behind. A buffer of ready machines earns its place, especially in the weeks after a move when somebody's laptop dies and the normal lead time is unacceptable. Cap it though. A spare pool without a ceiling becomes a storeroom.

Dispose covers everything too old or broken to issue. Decide the route in advance, with data handled properly, because the default route is a cupboard and a cupboard isn't a route.

The decision nobody makes is who owns the pile. Facilities assume IT, IT assume procurement, and in reality it belongs to whoever is accountable for the asset register being correct at the end of the year. Name them before the returns start. Ten minutes. Prevents a year of drift.

One financial detail. Equipment written down over three or four years that comes back in year two still carries book value, and scrapping it early has an accounting consequence somebody in finance would rather be told about than discover. Have the conversation first.

Measuring Whether It Worked

Mandates get judged on attendance, which is easy to count and tells you almost nothing about whether the hardware side held up.

Three measures are worth taking and they're all cheap. First, how many people could actually work on their first day in the building, asked directly at the end of that week rather than inferred from ticket volume. New arrivals behave like new starters. They work around problems instead of reporting them, so a quiet week isn't evidence of a smooth one.

Second, projected spend against actual. If you budgeted 300 monitors and bought 418, your inventory was wrong and the useful question is which part of it. That number also tells you whether to trust the register next time.

Third, retrieval completion on that quarter's leavers, measured as devices actually recovered rather than requests raised. This is the number most likely to be bad and least likely to be looked at, because it lands in a different team's reporting from the mandate itself.

Don't measure satisfaction with the equipment in month one. People who dislike the policy will rate the kit badly and people who like it will rate it well, and you'll learn about the policy rather than the hardware.

Take all three at week six and again at month four. Twice is enough. The gap between the two readings is where the real story sits, because most mandates look chaotic at week six and the majority of that noise resolves on its own without anybody intervening. Whatever hasn't resolved by month four needs a decision rather than more time.

The Multi-Country Version

Everything above assumes one country. Across several it changes shape rather than scale, and running one plan everywhere produces outcomes that look arbitrary from inside the markets you got wrong.

Kit can't simply move. Hardware issued in one country and returned to an office in another crosses a border, and that's a customs event with a declared value and paperwork, not a courier booking. Teams that treat an international return as a domestic one find out when a package is held and somebody asks for a commercial invoice they've never produced in their life.

What you issued differs by market too, usually without anybody deciding it should. Local purchasing quietly produces local keyboard layouts, local power supplies and whatever that supplier happened to have in stock the month somebody placed the order. So "bring your monitor in" is a clear instruction in one country and meaningless in another where nobody ever got one. Same policy. Different reality.

And whether you can require attendance at all isn't uniform. Ask locally. That's a question for advice in each market every single time rather than a policy you write once centrally and apply everywhere without checking. The hardware consequence is simply that you may end up with different cohorts in different countries, and the equipment plan has to follow that rather than lead it.

Plan per market, report centrally. That's the opposite of the instinct and it's the arrangement that actually survives contact with six different customs regimes and six different supplier relationships. Set the principle centrally, let each market decide what moves, keep one register. One logistics plan across six countries is a plan that's wrong in five.

If you're retrieving from several countries at once, that's the point where doing it yourself stops making sense. The volume is bounded and the complexity isn't, which is exactly the shape of problem worth paying somebody else to own.

One Thing Worth Doing Before Anything Else

If you take one action from this piece, make it this one, because it costs an afternoon and routinely removes a third of the projected spend.

Walk the building and count. Monitors, docks, chairs, cables. Not what the facilities inventory says, which was last accurate in 2021 and has been copied forward ever since by people who assumed somebody else was checking it. What's physically there, in cupboards, under desks, in the room everyone forgot about behind the kitchen.

Offices that ran at thirty per cent occupancy for four years are strange places. Things vanished. Taken home by people who needed them and never logged it, mostly. Other things accumulated instead, because every departing employee's desk got cleared into storage and nobody ever once went back through any of it. Both happened at the same time. That's why guessing fails in both directions.

I watched one company budget for 300 monitors and then find 190 of them already sitting in the building, which turned a purchase order into an afternoon of carrying things upstairs. Another budgeted for none at all, assuming the place had been stripped bare, and turned up 240. Neither had looked first.

Do it before the announcement, not after. Once the policy is out you're buying under time pressure against a date you've already published, and that's the most expensive way to acquire anything.

The Cost of Choosing the Wrong Approach

The visible cost is the unbudgeted purchase order, and honestly it's recoverable. Buying 200 monitors in a hurry costs somewhat more than buying them on a planned cycle with a negotiated rate, and that difference is annoying rather than serious. Fine.

What hurts is the kit that never comes back. Devices held by people who left during a contested policy, in homes nobody visited, with a retrieval process that was swamped in exactly that quarter. Those quietly leave the register, and by the time anyone notices it reads as ordinary attrition in the asset numbers rather than as a consequence of the mandate. I've seen a company write off 31 devices this way and attribute none of them to the policy that caused it.

Then there's the revision problem. A mandate announced as three days becomes two within a year more often than people expect. Anybody who funded a complete second setup for 400 people has spent that money against a position that no longer exists.

So the diagnostic question is this. Are you solving a counting problem, a logistics problem, or a people problem? The answer changes which of the six platforms above is even relevant to you, and in my experience most teams assume it's the second when what they actually have is the first.

The Retrieval Spike Nobody Resources

The attrition number in the board paper is a percentage. What lands on your desk is a queue, and those two things don't feel remotely similar when one of them is sitting in your inbox.

Resignations after a mandate cluster hard. People wait for the policy to be confirmed, give notice within the same three or four weeks, then work out similar notice periods, so the devices all come back inside one narrow window rather than trickling in the way they normally do. A process that handles five retrievals a month without anybody noticing will not absorb 23 in a fortnight. And the failure isn't dramatic. Kit just sits in spare rooms for months because nobody had the capacity to arrange collection, and somewhere around month four it quietly stops being recoverable at all.

Leavers here are also less cooperative than usual, which nobody plans for. Somebody who resigned over a policy they disliked has no particular reason to chase a courier slot or sit at home waiting for a collection window that may or may not arrive. Recovery rates on this cohort run measurably worse. Plan on your ordinary rate and you'll be short.

Then there's the timing collision. The same small team is usually responsible for setting up the building and recovering devices from people who are leaving, in the same eight weeks, and one of those two is going to give. It's always retrieval. The office has a date in the calendar and a laptop in somebody's spare room doesn't.

Resource it as a separate workstream with its own owner, or accept the losses and write them into the budget honestly. Pretending the existing process will cope is the version that costs the most.

What People Carry, and What They Won't

Worth a short section because it's the detail that quietly decides compliance.

A laptop in a bag is fine. A dock and cables in the same bag is fine, grudgingly. A 27-inch monitor is not, and a chair is laughable. Yet mandates routinely ask for all four back as though they're the same category of object, written by somebody who hasn't tried it.

The practical split is simple enough. Anything under about two kilos that fits in a bag will come in. Anything larger needs you to either collect it or leave it where it is, and collection from 200 homes is a logistics project with a real budget rather than a line in an announcement. Most companies discover this when the first person asks how, and the honest answer turns out to be that nobody thought about it.

Decide per item type rather than per person. Laptops and peripherals come in. Monitors get collected on a schedule or stay where they are. Chairs stay almost without exception, because shipping one across a country costs more than the chair is worth and selling it secondhand recovers somewhere close to nothing at all.

Write that down in the announcement. Three lines. It removes about half the questions you'd otherwise field in week one, and it stops people inventing a more demanding version of the policy than the one you actually wrote.

When You're Ready to Move Beyond Spreadsheets

You'll know. The signal isn't headcount, it's that three people are maintaining three different versions of what the company owns, and none of them reconciles. Add a mandate on top and the gap between those versions becomes the thing standing between you and a budget you can defend.

RemoAsset sits at the point where the moving matters more than the counting: kit going out to new starters, coming back from leavers, crossing borders in both directions, with one record of where everything is. That's the maturity curve it's built for, and it's our own product so weigh that accordingly.

If that's roughly where you are, it's worth a closer look.

A last word on the thing that makes all of this harder than it needs to be.

Mandates get written by people thinking about culture and collaboration, which is reasonable, and the hardware lands on people who weren't in that conversation. That gap is where most of the cost comes from. Not from bad decisions, just from a decision made without the one input that would have shaped it.

So if you're reading this before an announcement goes out, the single most useful thing you can do is get the inventory number in front of whoever is writing the policy. One figure. What we own, where it is, what moving it costs. It'll change the policy more than any argument about culture will, because it's the only part of the discussion with an invoice attached.

Frequently Asked Questions

Know where every device is, in every country, without asking anyone.