WFH monitoring best practices

Rolling out monitoring to a remote team is mostly a management exercise. The install takes minutes; the part that determines whether it works happens before and after.

What should you decide before choosing a product?

Write down the question you need answered, in one sentence. Then pick the lightest configuration that answers it. Most companies that arrive asking for screenshots need one of the two levels below it.

The honest test. If you cannot write a sentence explaining what decision level 4 lets you make that level 3 does not, you do not need level 4. "It would be nice to see" is not a purpose, and it will not survive a conversation with your team or a data-protection reviewer.

What goes in the monitoring policy?

One page, in plain language, that an employee can actually read. It should state:

• What is collected — the specific list, not a category.
• What is never collected — this reassures far more than the first list.
• Why — the business purpose, in one sentence.
• Who can see it — which roles, and that access is logged.
• How long it is kept — a number of days.
• What the employee can see — ideally everything about themselves.
• Which devices — company machines only, stated explicitly.

If your vendor will not give you the first two lists in a form you can paste into that policy, that is worth noticing before you buy. Ours is the privacy page, written to be circulated.

How should the rollout be sequenced?

1. Tell people first, with the policy attachedNot a line in an all-hands. A written note they can read twice and ask about.
2. Run a pilot with one team, screenshots offTwo weeks. This surfaces the real edge cases — split shifts, meeting-heavy roles, deep work with no input — while they are still cheap to fix.
3. Review the pilot data with the pilot teamShow them their own numbers and ask what the data gets wrong. It always gets something wrong, and hearing it now is cheaper than hearing it from everyone later.
4. Set retention before you widenA number of days, written into the policy and configured in the product. See data retention.
5. Roll out, then keep discussing the outputData nobody ever discusses is surveillance. Data that changes how work is planned is a tool. The difference is entirely in what managers do next.

What are the mistakes that actually cost teams?

• Installing first and explaining afterwards. Unrecoverable. The team will conclude you tried to get away with it.
• Using the activity percentage in a performance conversation. It measures typing. Doing this once teaches everyone to game it, and from then on the data is worthless.
• Monitoring everybody because one person is a problem. Deal with the one person. Surveillance as a substitute for a difficult conversation is visible to everyone.
• Indefinite retention. Nobody looks at a screenshot from eight months ago. You are holding risk for no benefit.
• Giving managers more access than they need. Scope it, and log it.
• Never showing employees their own data. The single cheapest thing you can do for adoption, and the one most products get wrong.

What does good look like after three months?

• Hours per client are accurate enough to price the next contract from.
• Somebody noticed a person was overloaded before they resigned.
• The team is not talking about the tracker, because it stopped being an issue.
• Nobody has needed to open a screenshot in weeks.

If instead you are having arguments about activity percentages, the level was set too high. Turn something off.

Related