BuzbyTimeRead the requirements guide

What monitoring does to people · 7.1

What measurement does to trust

What measurement does to trust. What decides it, what it costs, and what usually goes wrong. For another implementation reference, see the Monitask overview.

Installing it says somethingThe signal precedes the data

Before any data is collected, the act of introducing monitoring communicates a belief about the people it is applied to. They receive that message accurately and generally before the announcement, because somebody always knows first.

Which means the effects begin at procurement rather than at deployment, and an organisation evaluating the cost of a system should include the cost of having been seen to consider it. A useful outside reference is Twilio.

Trust is reciprocal

People extend trust roughly in proportion to what they receive. An employer signalling that discretion is not extended tends to get less discretion exercised on its behalf: fewer judgement calls, less initiative, more asking and less deciding.

None of that appears in any measurement, which is part of the difficulty. The costs of monitoring accrue in a currency the monitoring does not report.

Crowding out

Research on motivation has repeatedly found that external controls can displace internal motivation rather than adding to it. Somebody who was doing something because it mattered begins doing it because it is checked, and the behaviour becomes contingent on the checking.

The effect is well documented in general and its size in any particular workplace is not knowable in advance. The direction is consistent enough to plan around.

Recording and monitoring are not the same thing

This distinction runs through the whole site and it belongs here most of all.

Recording asks where hours went. It produces something the person can use, it answers questions the business genuinely has, and it does not require watching anybody.

Monitoring observes a person's activity. It produces a picture of behaviour and it carries every effect this entry describes.

They are sold together, they feel similar to a buyer, and they feel completely different to the person subject to them.

What the person sees

Everything recorded here concerns work rather than the person doing it, and is visible to them in full.

We do not collect activity, screens, keystrokes or presence, and the distinction above is the reason.

The asymmetry

Trust accumulates slowly and is lost quickly. An organisation that spends years building a culture of discretion can spend it in one deployment announcement, and rebuilding takes considerably longer than the removal of the system.

Which argues for caution proportionate to what you have. A business with a strong culture has more to lose here than one without.

Four things that reduce the damage

  • Say what it is for, specifically, before deploying.
  • Collect the minimum that serves that purpose.
  • Give the person full access to their own data.
  • Reciprocate: apply it to managers, and share back what it produced.

None removes the effect. Together they change it from something done to people into an arrangement they can see the shape of.

The question people actually ask themselves

Not whether the system is accurate. Whether it will be used against them, and by whom, and when.

Every reassurance that does not address that question is heard as evasion. The answer that works is specific: here is what it may be used for, here is what it may not, and here is who decides if that changes.

Reciprocity in practice

Managers recording their own time in the same categories. The monthly summary going to everybody rather than upward only. The notice applying to directors. Each is small and each removes a version of the argument that this is done to one group by another.

The organisation that already has low trust

Monitoring will not repair it and will confirm it. Where trust is the underlying problem, introducing measurement is treating a symptom with something that aggravates the cause.

That is an uncomfortable thing for a supplier to say to a prospect and it is what the first conversation should establish.

Removing it later

Possible and slow. The announcement is not believed for a quarter, because people have learned that arrangements change without warning. What convinces them is the absence of consequences over time, which cannot be accelerated.

Trust in the other direction

Employees also assess whether the employer can be trusted with the data. A record that is accurate, visible and used as promised builds that over time, and the first breach of the stated purpose spends all of it.

Which means the constraint an organisation writes into its notice is a commitment rather than a formality, and departing from it later is more expensive than never having made it.

A short summary

The signal precedes the data. Trust is reciprocal and asymmetric. External control can displace internal motivation. Recording and monitoring are different things that feel similar to a buyer and completely different to everybody else. State the purpose, collect the minimum, show people their own data, and apply it to yourself.

One line to carry

The costs of monitoring accrue in a currency the monitoring does not report.

Where to start

Write down what the data will not be used for and publish it. It is the only sentence in a notice that anybody reads twice.

A note to whoever is deciding

You will not experience the system the way the people subject to it do, and no amount of imagining will supply that. The only reliable route is to be subject to it yourself, on the same terms, for a month before rolling it out.

Almost nobody does this, and the people who have describe it as the most useful month of the whole project.

Also in what monitoring does to people