How to choose an attendance system for multiple branches
06.09.2026
When choosing an attendance system for multiple branches, the question asked most often is "which software is good". In practice the decision is settled by something else: which part of the record breaks first as the branch count grows.
The answer is almost always the same - comparability. Below is why, and what the selection criteria should be.
What branch count changes
In a single office the record is one person's job and they know their own format well. When a second branch opens, a second file appears; at the third, a third.
After a while the differences surface: one writes "09:05", another "9.05"; in one, lateness counts after ten minutes, in another after fifteen; in one the break is inside working time, in another it is not.
The numbers add up but do not compare. This is the most dangerous state, because the report exists and looks correct - yet a decision based on it can be wrong.
A notable detail: the problem depends not on headcount but on the number of files. For a single office of two hundred a spreadsheet stays bearable; for six branches totalling sixty people it does not.
Structural requirements
The first thing to require from the system is structural support - at least three levels: branch, department and section.
It looks like a technical detail, but it directly determines whether the reporting is usable. Without structure the report is either too general - "42 late arrivals company-wide" - or too granular, a separate row per employee. Neither serves a decision.
With the structure in place, every recording point attaches to its branch and the same data reads at three levels: company, branch, section. In practice the second is the cut used most.
Permission levels
The second requirement is who sees what.
A branch manager seeing company-wide data is usually neither necessary nor helpful. The model that works:
- Branch manager - their branch.
- Department head - their team.
- HR - the overall picture.
- Accounting - the approved final figures only.
In a system without permission levels one of two things happens: either everyone sees everything, or the data is distributed through files HR builds by hand. The second is a return to the original problem - manual work.
Comparable reporting
Genuinely comparing branches has one condition: the rules have to be centralised.
If the lateness threshold, the break rule, the rounding step and the schedule templates are held in one place, the indicators are calculated by the same logic. Then the sentence "branch three has twice the lateness of branch four" carries meaning.
When each branch keeps its own rules, the same sentence is meaningless - the numbers sit side by side but do not measure the same thing. This is the most common analytical error in multi-branch companies.
The practical test question: are the lateness figures of two branches calculated by the same rule? If the answer is "I don't know", comparison is not yet possible.
Adding a new site
This is the biggest practical difference between the models, and it often settles the decision on its own.
In a device-based setup every new branch is a project of its own: order, wait, cabling, installation, testing, acceptance. That is both a budget line and a delay to the opening date.
Where the recording point attaches to the structure, a new branch is essentially a row to add - the code is printed, the point is created in the structure, employees are added.
If two or three new sites are planned over three years, this difference becomes the most volatile line in the cost calculation. The full method is on the attendance system costs page.
Selection criteria
- Are there branch, department and section levels?
- Can permission levels be built by role?
- Are the rules centralised, or kept separately in each branch?
- How many days does adding a new site take?
- Is the report available as a branch cut and in comparison form?
- Can notifications be sent scoped to a branch?
All six need a concrete answer during a demo. "Yes, that's possible" is not enough - it should be shown.
A worked example: from six branches to one report
A pharmacy chain had six branches and sixty-two employees. Each branch manager filled in their own spreadsheet and sent the file to head office on the 2nd.
At head office two people merged those files. The job took up to three days, and every month the same question came back: why is lateness so low at branch three?
The answer only surfaced a year later: at that branch the manager recorded lateness after twenty minutes, while everywhere else it was ten. For a year the company had been looking at figures that were, in fact, not comparable.
Three things changed in the new setup. The rules were centralised - the lateness threshold, break rule and schedule templates were held in one place. The structure was built so every recording point attached to its branch. Permission levels were set: the manager saw their branch, HR saw everything.
The result: the file-merging work disappeared entirely. But the company flagged something more important - branch comparison became meaningful for the first time, because the figures were now calculated by the same rule.
When a seventh branch opened, the extra work took one day: a row was added to the structure, the code was printed, employees were connected.
What the branch manager needs
In multi-branch setups the most overlooked party is the branch manager. The system is built for head office and the manager becomes just a data source - which is the most common reason such rollouts fall out of use.
In practice the manager needs three things:
- A daily picture - who is in today and who is not.
- Notifications scoped to their branch - a shift that has not opened, lateness.
- The ability to correct - leaving a trace, but without having to go to head office.
The third causes the most argument. In practice, closing off correction entirely does not work: the manager asks HR for the correction by phone, and no trace is created anyway. A correction that leaves a trace is both fast and verifiable.
The next step
In a multi-branch company the rollout starts with one branch - not the simplest, but the one whose rules change most. A month of parallel running exposes the gaps and creates a template for the rest.
Calculate the QRGate price for your branch count, or first look at what it does.
