When to Buy Something

Attendance software is cheap enough that the question is rarely cost. It is whether the problem you have is one a product solves.

Real signals

Two people approved for the same week more than once.

A dispute about a balance you cannot settle from records.

More than one approver, with no shared view of what has been approved.

Part-time and variable hours, where entitlement arithmetic has become genuinely fiddly.

Statutory calculations you know you are getting wrong.

More than one site, so a wall calendar cannot be the shared view.

Any one of these repeatedly is enough.

Signals that are not

Someone said a spreadsheet is unprofessional. It is not, and at fifteen people a good one outperforms a badly configured product.

A vendor's compliance warning. Ask which specific requirement you are failing; usually the answer is vague.

Wanting reports nobody has asked for.

Growth that has not happened yet. Buy for the size you are plus a year, not for the size you imagine.

A single bad month, which is usually a process failure rather than a tooling one.

What the product should do

Bookings and approvals in one place.

Balances calculated, including pro-rating.

A shared calendar showing who is off.

Self-service so people see their own balance.

Sickness recorded with dates, without a diagnosis field you feel obliged to fill.

An export, so leaving is possible and payroll can be fed.

Six things. Anything beyond them is a reason to look harder at the price.

What it should not do

Monitor activity, take screenshots, or score productivity. These appear in products in this category and no attendance purpose requires them.

Automatically trigger disciplinary steps from an absence count.

Require a medical reason as a mandatory field, which turns an ordinary record into health data.

Check the defaults, because in several products these arrive enabled.

Cost, honestly

Per person per month, which for a small team is modest.

Plus your time setting it up: entitlement rules, people, historical balances.

Plus the annual work of rolling the leave year and checking calculations.

And the switching cost later, which is small if the export works and irritating if it does not.

Test the export in the first month, not in year three.

The order that works

Run the spreadsheet properly for six months first.

Write down the five questions you keep having to answer.

Trial two products against those five questions, with your own awkward part-timer.

Then decide, and the decision may still be to keep the sheet.

Try the export first

Before committing to anything.

In the trial, export your data and open it somewhere else.

Check it contains balances, history and approved future leave, not just a list of names.

Time how long it takes.

Whatever is missing now will be missing when you leave, and that is the moment when there is no leverage to ask for it.

Turn the principle into a test

For an example that can make this requirement testable, consult open the workflow reference. Treat the page as a starting point rather than proof: reproduce the workflow with real roles, exceptions and permissions.