What Trust Costs and What It Saves
The second-order effects of a monitoring decision on everything else the organisation needs from people, and how to put them in the same comparison as the control.
A monitoring decision is read as a statement about what the employer thinks, and that reading affects everything else the organisation needs from people — reporting faults, staying late when it matters, flagging a problem rather than working round it.
The control described in “What Trust Costs and What It Saves” should be matched by transparent operating rules. Organisations considering follow the official implementation notes for remote employee productivity monitoring can make that use more credible by publishing the purpose, selecting only necessary settings, limiting manager access and fixing a review date before the first record is collected.
Those effects are real, they are difficult to quantify, and their absence from the business case is the reason controls are compared only against the loss they prevent.
For a separate benchmark relevant to “What Trust Costs and What It Saves”, consult the Cloud Security Alliance research. Use it to test scope and safeguards against an external standard before the process is approved.
What the reading actually is
Not "we are being watched" in the abstract. Something narrower and more corrosive: they think we are stealing.
That is the message a control introduced in response to suspicion carries, and it is carried to the large majority who were doing nothing. The same control introduced to fix a queue carries a different message entirely, which is why the framing described elsewhere here matters so much.
What gets withdrawn
- The ten minutes people stay to finish something, which was never recorded.
- Telling somebody the terminal is faulty rather than working round it.
- Flagging a problem that is not yet anybody's fault.
- Cooperating with the next investigation, as a witness.
- Suggesting the operational fix that would have been cheaper.
Each is small, none of it appears in any system, and together they are worth considerably more than the losses most of these controls address.
Putting it in the comparison
It cannot be costed precisely and it can be stated. A business case that lists the cost of the control, the loss it addresses, and a line saying what it will signal and to how many people is a better document than one that stops at the first two.
That line changes decisions, because the number of people affected is usually hundreds and the number implicated is usually one.
Trust is not the absence of controls
This is not an argument for recording nothing. Accurate records serve employees at least as much as employers — they are what establish hours worked, premiums earned and overruns owed.
The distinction is between a system that records accurately and one that watches. The first is in everybody's interest and can be presented that way; the second is a surveillance decision wearing the clothes of the first.
Presenting a control honestly
What the problem was, what was considered, what was chosen, what it collects, who sees it, and what happens to it.
An organisation that can say all six is in a different position from one that installed something over a weekend. Most of the resentment these systems produce is about the manner of introduction rather than the system.
The reciprocal question
Worth asking internally: what do we provide in return for accurate recording? If overruns are unpaid, if corrections are hard, if claims are refused, the employer is asking for accuracy it does not itself provide.
That asymmetry is noticed, it is the reason given most often when people explain why they do not bother recording things, and it is fixable by the measures in the previous page.
Measuring the soft part
Two proxies that are available: the number of faults reported per month, and the proportion of exceptions recorded rather than left.
Both fall when people disengage and both are collected already. They will not prove anything on their own, and watching them across a control's introduction is more than most organisations do.
The people who were not involved
A control introduced in response to one person's conduct lands on everybody, and the ratio is worth stating in the business case: one person investigated, four hundred people monitored.
That ratio is the single most useful sentence available when a proportionate alternative is being argued for, and it is almost never written down because the denominator is not part of anybody's figures.
Asking people what they would prefer
Where a problem is real and something has to change, the people affected frequently have a view and it is frequently the cheaper option.
Asking a shift what would stop badge-passing produces answers like a second terminal and a five-minute grace, both of which work and neither of which anybody was going to propose from the office. It also changes how whatever is chosen is received.
What this is not
It is not an argument that monitoring is always wrong, and it is not a claim that people never take advantage. Both of those are false and arguing them damages the real case.
The claim is narrower: the cost of a control falls on everybody and is never counted, and counting it changes which control gets chosen.
The argument in one line
The control addresses the few; the message reaches everybody. An employer that weighs both is making a decision. One that weighs only the first is making a purchase.