IMonitor Software Blog

Most buying mistakes in this category are not technical. They are decisions taken before anyone looked at a demo: what the tool is allowed to see, who will be able to read it, and what happens when the project is reviewed in a year. Here are twelve checks, grouped by the order in which they will bite you.

Legal and organisational checks

1. Can you state the purpose in one sentence?

If the sentence needs a comma to hold two purposes, you have two projects. Split them, because each needs its own scope and its own notice.

2. Is the notice written before the installation date?

A notice written after go-live is a reconstruction. Write it first, then configure the tool to match it, not the other way round.

3. Do employee representatives have a role?

In several countries they have co-determination or consultation rights on monitoring technology. Confirm the local rule in your jurisdiction before the purchase order, not after.

4. Is there a named person accountable for the data?

Not a department: a person. Someone must be able to answer “who looked at this record, and why” without asking around.

Technical checks

5. Can each module be switched on independently?

Screenshots, keystroke content, website categories, file tracking and time records should not be one switch. If they are, you cannot honour a narrow notice.

6. Does it keep working offline, and what happens to the data?

Laptops travel. The agent should record locally and upload later, with a defined cap, and the local cache should be encrypted or cleared after sync.

7. Is there a real self-service view for employees?

Being able to see your own records changes the tone of the whole deployment. If the vendor treats this as an add-on, the product was designed for one-way surveillance.

8. How does uninstall and offboarding work?

Ask for the documented removal procedure, including what happens to records when an employee leaves and when a department is closed. “Delete the folder” is not a procedure.

Data governance checks

9. Are access rights granular, and is every view logged?

The most common trust failure is an unnecessary view, not a malicious one. Per-role permissions plus an access log make the question answerable.

10. Is retention enforced by the system or by people?

Prefer a product that deletes on a schedule. Retention that depends on someone remembering is retention that does not happen.

11. Can you export and then leave?

You should be able to export your own records in a usable format and terminate without losing them. Ask what the export contains before you sign.

Vendor checks

12. Will they say no to a feature you asked for?

The most useful signal in this category: a vendor who explains why a request is disproportionate, instead of shipping it. If every request is met with “we can do that”, the tool will eventually be configured into a problem.

What good looks like on day one

• Time records on for everyone; every other module off.
• A one-page notice that names the data, the viewers and the retention period.
• Two named people with access, and a log that shows their views.
• A calendar reminder for a six-month review, with the authority to switch things off.

That configuration is deliberately boring. It is also the one that survives an audit, a staff meeting and a change of manager.

Frequently asked questions

This article is general information and does not constitute legal advice. Confirm the rules that apply in your jurisdiction before deploying monitoring software. Last updated: 23 October 2026.

Buyers usually arrive at this question after a bad meeting. Someone asked for “monitoring”, someone else heard “time clock”, and the conversation ended without a decision. The two are related, but they answer different questions, and confusing them leads either to over-buying or to a tool that cannot do the job.

Here is the distinction in plain terms, four scenarios that decide it, and what to configure in each case.

The difference in one line each

Where they overlap

Most modern tools do both, which is exactly why the confusion exists. An agent that records working hours can also record application usage; a monitoring product can produce a timesheet. The overlap does not mean the two are interchangeable. It means the configuration is where the decision lives.

Four scenarios that decide it

1. You must document working time

In several jurisdictions employers are required to record working time, and the record must be reliable and retained. That is a time-tracking problem. Buy for accuracy and export, not for screenshots.

2. You had a data incident

A file left the company, or someone is suspected of copying client data. That is a monitoring problem, and it should be scoped to the incident: specific devices, a defined window, a named investigator. Broad always-on capture is not the proportionate answer.

3. You suspect the work is not happening

This is the most common case and usually the one where monitoring is the wrong first move. Before recording anyone, check the five operational indicators — delivery against commitments, time stuck waiting, focus time, where requests come from, rework count. If those show a real problem, it is usually a process problem, not an effort problem.

4. You need both, for different groups

Field staff on billable hours, plus an IT team that needs incident forensics. That is legitimate, and it argues for one platform with per-group configuration rather than two products with separate data sets.

What to configure, depending on the answer

• Time tracking only: enable hours, breaks and project attribution; leave screenshots, keystroke content and website categories off; give employees a self-view of their own timesheet.
• Incident scope only: enable the relevant modules for specific devices for a fixed period, log every access, and set an automatic end date.
• Both: keep the groups separate so that payroll data never sits next to forensic captures, and so that a viewer of one cannot see the other.

The legal asymmetry worth remembering

Recording working time is often an obligation. Monitoring content is almost always an intrusion that must be justified. That asymmetry should drive the defaults: hours on, screenshots off, and a written reason before anything else is switched on.

Frequently asked questions

This article is general information and does not constitute legal advice. Consult a professional on employment law and data protection before deploying either capability. Last updated: 23 October 2026.

“We cannot see productivity in remote work, so we are going to install monitoring.” The sentence still comes up regularly. In practice it does the opposite of what is intended: monitoring does not raise productivity. It raises the volume of records. And the more you record, the more the team works for the record.

This article sets out five indicators that show how a distributed team is actually doing, and how to run them. It is not an argument against monitoring. It is an argument about what to measure first.

Why “not seeing” becomes a problem

The discomfort comes from lost information. In an office, passing someone in the corridor told you something: how they are doing, whether they are stuck, who they are talking to. Remote work zeroes that out. The fastest way to fill the gap is to look at the screen, and monitoring becomes the default option.

There are other ways to fill it, and they are usually cheaper, faster and less resisted.

Five indicators instead of monitoring

1. Delivery against commitments

Not “how many hours did you work”, but “did the committed thing ship on the committed date”. The simplest indicator, and the least misinterpreted.

2. Time stuck waiting

How long a task sits while someone owes a reply. This is not about individual effort: it shows where the organisation blocks. Monitoring never sees it.

3. Focus time outside meetings

The share of the calendar occupied by meetings. Too high, and individual work is fragmented. The calendar is more accurate than any screenshot.

4. Where requests come from

Who asks whom. Every question concentrated on one person, or someone who never asks, are early signals. What matters is the skew, not the volume.

5. Rework count

How often the same work is redone. A high count points to an unclear instruction or specification. More monitoring does not help, because the cause sits upstream.

Running it in 30 minutes a week

1. Start of week, 15 minutes: confirm what ships this week; each person commits to one thing.
2. Midweek, 5 minutes: collect only what is blocked. Look at the blockage, not the person.
3. End of week, 10 minutes: put commitment and result side by side. No blame; it is input for next week.

With those three pieces, most teams get out of the “we cannot see” feeling. Cost: zero. Time: half an hour.

When recording is genuinely justified

• Working time with legal effect: a separate subject, with its own basis.
• Investigating a specific data incident: scope limited to that case.
• Evidencing hours for contractors: where the contract requires it.

All three share a shape: one purpose, a bounded period. Not “everything, just in case”.

If you do deploy, in this order

1. Run one month with the five indicators only.
2. Identify what remains uncovered. Do not deploy on a feeling.
3. Record only that area, for a defined period.
4. Document the data, the retention period and who has access.
5. At the end of the period, check whether the indicators actually improved.

Frequently asked questions

This article is general information and does not constitute legal advice. Consult a professional on employment law and data protection. Last updated: 23 October 2026.

When companies tell employees that work computers will be recorded, the obstacle is rarely the intention. It is the wording. A notice that says “the company may monitor systems as necessary” answers none of the three questions people actually ask: what is seen, who sees it, and how long it is kept.

This page gives a notice you can copy, and the points that usually go wrong when adapting it. It is a template, not legal advice: the fit with your internal rules and local law needs a professional’s eye.

Seven elements the notice must contain

1. Purpose: working time, information security, or process improvement. One is better than three.
2. Scope: which devices, which departments, how many.
3. Data recorded: logon and logoff times, application usage, screenshots yes or no.
4. Data not recorded: personal email, personal cloud. Saying this explicitly changes how the notice is received.
5. Who can access: by role and function, not “management”.
6. Retention period: how long, and what happens afterwards.
7. Contact point: where questions and objections go.

The template

Copy the text below and replace only the bold bracketed parts.

What usually goes wrong when adapting it

• “As necessary” instead of a list. That phrase opens the scope without limit. Enumerate the data types.
• Leaving out the exclusion list. It is the first thing people ask about. Naming the personal sphere explicitly gives the rest of the text credibility.
• Naming the viewing department only. Add function and level.
• Retention without deletion. “Three months of retention” is not enough; add “then deleted”.
• Conflict with the employee handbook. If the data listed differs from the handbook, the whole text becomes attackable.

How to deliver it

1. Document the internal rule first, and consult employee representatives where required.
2. Give everyone the same text, in a meeting rather than only by email.
3. Hold a session per department, with time for questions.
4. Collect acknowledgement, stating that refusal carries no detriment.
5. After go-live, open self-service so people can see their own records.

Frequently asked questions

This article is general information and does not constitute legal advice. Consult a professional before changing internal policies or processing personal data. Last updated: 23 October 2026.

Most companies that ask about employee monitoring software are not really asking about software. They are asking a management question: how do we know the work is getting done when we cannot see it. The software decision comes after that question, not before.

This guide covers what to decide before installing anything, what to record and what to leave alone, and how to introduce it without losing the team. It is general information, not legal advice: the rules differ by country and by state, and a local professional should confirm the specifics.

Three decisions to make before you install

Everything else follows from these, and getting them wrong is what makes monitoring projects fail.

1. One purpose. Productivity improvement, information security, or working-time verification. Pick one. If you cannot say which, you are not ready to install.
2. A written scope. Which devices, which departments, which hours, which level of detail. “All company computers, all the time” is a scope, but it is rarely the one you need.
3. A retention period. How long the records stay, and what deletes them. If nobody can answer this, the data will stay forever by default.

What to record, and what to avoid

One question decides every row: can the purpose be met without this data? If a less intrusive option exists, that option is the one to choose.

Informing the team

1. Turn the three decisions into a document that names the data, the people with access, and the retention period.
2. Deliver the same text to everyone, in a meeting rather than an email, because it allows questions.
3. Collect acknowledgement, and state clearly that refusing does not lead to any detriment.
4. Let people see their own records. This single step changes the tone of the entire project.

Legal checkpoints that apply almost everywhere

• Purpose limitation: monitoring must serve a stated purpose, and data collected for one purpose should not quietly serve another.
• Proportionality: the least intrusive means that achieves the purpose.
• Transparency: people know what is collected, why, and who can see it.
• Access control: a small, named group can view records, and each view is logged.
• Retention limits: a defined period and actual deletion at the end of it.
• Employee representation: in several countries, works councils or unions have co-determination or consultation rights. In Germany, works councils have a codetermination right on monitoring technology; in France, the CNIL expects proportionality and prior consultation; in Italy, the workers’ statute restricts remote control; in Brazil and Korea, data protection law governs the processing. Confirm the local rule rather than assuming.

A rollout order that avoids conflict

1. Start with time records only: no screenshots, no keystroke content.
2. Fix the viewing side first: who can look, and on what occasion. Most trust damage comes from an unnecessary view, not from the tool.
3. Open self-service before adding features.
4. Add only what you can explain, one department at a time if needed.
5. Review after six months and switch off whatever no longer serves a purpose. Being able to switch off is the strongest argument in the rollout conversation.

Frequently asked questions

This article is general information and does not constitute legal advice. Consult a professional before changing internal policies or processing personal data. Last updated: 23 October 2026.