Where the Work Is Actually Done

Of the four facts the register needs, this is the one that most often decides the answer and the one the organisation holds least reliably. Working time protections are typically territorial: they attach to work performed in a place, largely regardless of where the contract was signed or who signs the payslip.

The workflow in “Where the Work Is Actually Done” becomes more dependable when scheduled work, actual time and later corrections can be distinguished. For teams exploring workforce analytics software, this workforce tool can provide operational context, while policy ownership, employee explanation and consequential decisions remain with accountable people.

That principle is simple to state and awkward to apply, because the place of work is not a field anybody maintains. The directory holds an office. Payroll holds a country for tax purposes. Neither is a statement about where a person physically does their job, and for a growing minority the two have come apart.

For an independent reference relevant to “Where the Work Is Actually Done”, consult the OWASP privacy-risk project. Use it to challenge assumptions about working time, privacy, recordkeeping and exception handling against the organisation’s real operating model.

The practical consequence is that a register built from the directory will be confidently wrong about some population, and the organisation will not know which one until it looks.

Why the office in the directory is not the answer

The directory entry records where somebody was assigned when they joined. It survives relocations, it survives a move to permanent remote working, and it survives a team reorganisation that moved everybody's reporting line without moving anybody.

It is also maintained by the person themselves in many systems, which means it is accurate for people who think about it and stale for everybody else. Nobody updates a directory entry because their working time regime changed; they update it when they want their new desk to be findable.

A register built on it will show a tidy distribution across the organisation's offices, which is exactly what makes it convincing and exactly why nobody questions it.

The sources that are closer to the truth

Access control records show which building somebody physically enters, which is the strongest available evidence for site-based staff and says nothing about anybody who works from home.

Payroll country is a strong signal because it is maintained for tax and social security reasons, which means somebody checked it at some point and has an incentive to keep it right. It is not identical to place of work and it is usually much closer than the directory.

Travel and expense records cover the rest: they show where people actually were, in aggregate, over a period. For frequent travellers and people on assignment they are often the only record that reflects reality.

Reconciling them

Take the three and compare them per person. Most people will match on all three, and the exercise exists to produce the list of those who do not.

Expect the mismatches to fall into a small number of patterns: the person who moved, the long assignment nobody closed, the remote worker in a different country, and the frequent traveller whose base is one place and whose working life is another. Each pattern is a population with its own question.

The reconciliation is a report rather than an investigation, and it should be run and read at the aggregate level first. A figure — this many people out of this many do not reconcile — is enough to decide whether anything further is warranted, and it avoids starting with a list of individuals.

The home working question

Somebody working from home is working in the country where their home is, which is obvious and still catches organisations that treated home working as a variation on the office rather than as a change of location.

Where the home is in the same country as the office, nothing changes. Where it is not, the organisation has a person performing work in a jurisdiction where it may have no entity, no registration and no knowledge of the rules. The working time question is usually the smallest part of that and it is the one this collection is about.

The useful distinction is between home working within the country, which is an administrative matter, and home working across a border, which is a different question entirely and needs a route to somebody qualified.

What to record, and what not to

Record the country in which the work is performed, as a field, maintained deliberately, with a date. That is enough for the register and it is the minimum that makes it usable.

Do not record anything finer. The organisation needs a country, not an address, and collecting more than the country turns a compliance record into something that will attract a different set of questions about why it is being held.

Where the answer is genuinely mixed — somebody splitting their week between two countries — record both and flag it rather than forcing a single value. A flagged row is honest; a forced one is a decision somebody made silently.

How often it changes

Less than people fear and more than the systems assume. In a stable workforce a few per cent a year; after a reorganisation or a change in remote working policy, considerably more in one burst.

Attach the question to the events that cause it — a change of address, a change of contract, the end of an assignment, a relocation request — rather than to a calendar review. The calendar review will catch the ones the events missed, and it will be a short list rather than a survey.

The alternative is the common arrangement, where the field is set at hire and never touched again, which produces a register that was accurate once.