Churn

A site's population turns over continuously. People arrive for a phase, a week, or a day, and the records have to cope with that as the normal state rather than as an exception.

For teams turning churn into a repeatable record, view the platform provides a useful operational reference for time, attendance and reporting workflows. It should be assessed alongside Chartered Institute of Building so the software choice remains grounded in site practice and applicable guidance.

What the register shows

How many distinct people have been on site across the project, against how many are there today.

On a large project the first number is several times the second, and most organisations have never produced it.

First-day and last-day distributions, which show whether people stay for phases or for days.

And the proportion who appear once, which is usually larger than expected and is the group with the least induction and the highest risk.

Why churn matters here specifically

Every new person is an induction, a competence check and a register entry.

Which means the administrative load is proportional to churn rather than to headcount, and a site of two hundred with high churn costs more to control than one of four hundred that is stable.

The first-day risk is real: unfamiliar site, unfamiliar layout, unfamiliar hazards, and in most incident statistics a disproportionate share of injuries.

And short-duration people are hardest to reach with anything, including toolbox talks, changes to arrangements and the things learned from incidents.

The one-day visitor problem

Somebody who comes for a single day to do one job.

Full induction is disproportionate and no induction is wrong.

The workable answer is a short briefing plus an escort or a supervised work area, which transfers the site knowledge to somebody who has it.

And a record, because they were present and would need finding.

What churn does to competence records

A card verified in March for somebody who left in April and returned in September has been verified for a different period.

Which argues for verification at access rather than at enrolment, and for a status that expires with the card rather than with the project.

Re-verify returning people rather than reactivating an old record, which takes a minute and prevents a specific and common failure.

Reading the pattern

Churn concentrated in one subcontractor is a supply problem at their end, and it will show up as incidents and as poor work before it shows up commercially.

Churn concentrated in one trade is usually a market condition.

And a rise across the site frequently follows something: a welfare problem, a payment delay down the chain, a programme change that cut hours.

Which makes the churn figure an early indicator of things nobody reports.

Reducing the administrative cost

Enrolment once per person across all your sites rather than per project, where the same companies recur.

Pre-arrival data entry by the subcontractor.

And a returning-worker path that is quicker than a first enrolment while still re-verifying the things that expire.

These three convert most of the churn cost, and none requires different hardware.

The measure

Distinct people across the project against current headcount.

Proportion present for one day only.

First-day incidents as a share of all incidents, which is the figure that justifies attention to induction.

And churn by subcontractor, which is the one that predicts other problems.

Getting information to a changing workforce

A toolbox talk reaches whoever is present that morning.

Which on a high-churn site is a minority of the people who will work there that month.

So anything important has to be in the induction as well, and the induction has to be updated when it changes — which is the phase-change discipline again.

And the briefing board at the gate reaches everybody once a day, which makes it more valuable than it looks and worth keeping current rather than papered over with notices from March.

The last week of a project

Churn peaks at both ends, and the closing weeks are the worst: snagging trades arrive, the main workforce leaves, and the register is being maintained by people whose job is ending.

Which is when access control is most likely to lapse and when the site is most congested with small unfamiliar crews.

Plan for it: keep the enrolment resourced to the end, and keep the roll call capability until the gate closes for the last time.