Skip to content
LobbyExpress
Search

How-to

Measuring whether your visitor system works

Five practical measures, from time to badge to roll-call accuracy in a drill, with how to collect each one using records you already have and a stopwatch. Set your own baseline; ignore anyone else's benchmark.

By LobbyExpress Editorial

Most visitor systems are judged on anecdote. A host complains that their guest waited too long, a receptionist says the kiosk is fiddly, a fire drill goes badly and everyone has a theory why. Nobody can say whether things are better or worse than six months ago, because nobody measured.

Measurement does not need a data team. The system already records most of what you need, and the rest can be gathered with a clipboard, a stopwatch and a fire drill you were going to run anyway.

This how-to sets out five measures that tell you whether the system is doing its job, how to collect each one, and how to use them. It deliberately gives no target figures. Published benchmarks rarely match your building, your visitor mix or your definitions; your own baseline is the one that counts.

Step 1: decide what “working” means

A visitor system has a few jobs: get people through reception smoothly, tell hosts their guests have arrived, and know at any moment who is in the building. The measures below map to those jobs. Before collecting anything, agree with facilities, security and reception which of them matter most to you, so the results lead to decisions rather than a report nobody reads.

Each measure sits at a different point in a visit. The timeline below shows where.

Example visit timeline with five measurement pointsAn illustrative timeline of a single visit with six moments: host invites the visitor, visitor arrives, badge issued, host notified, host meets the visitor, and visitor signs out. Brackets show where each measure sits: pre-registration happens before arrival; time to badge runs from arrival to badge; host response time runs from notification to the host meeting the visitor; sign-out completeness depends on the final sign-out; and roll-call accuracy depends on the whole on-site period.EXAMPLE VISITInvitedArrivesBadgeissuedHostnotifiedHost meetsvisitorSignsoutPre-registrationinvited before arrival?Time to badgeHost responseSign-out recorded?Roll-call accuracyis the live list right at any moment while they are on site?
Example only, not to scale. Each measure looks at a different part of the visit, so a problem in one rarely shows up in the others.

Step 2: collect the five measures

Measure What it tells you Where the data comes from
Time to badge How long reception takes Stopwatch sample at the desk
Pre-registration rate Whether hosts use the system System records
Host response time Whether notifications work Timestamps or observation
Sign-out completeness Whether the live list can be trusted System records
Roll-call accuracy Whether the list works in an emergency Fire drill reconciliation

Time to badge

Definition: from the moment a visitor reaches the desk or kiosk to the moment they hold a badge or have completed sign-in. Write down where the clock starts; “joins the queue” and “reaches the kiosk” give different answers.

How to collect: most systems record when sign-in finished but not when the visitor walked in, so sample it by hand. Stand near reception at a busy period and a quiet one, and time a set number of visitors each. Note pre-registered and walk-in visitors separately, since they usually differ.

Pre-registration rate

Definition: the share of visits that a host created before the visitor arrived, out of all visits in the period.

How to collect: most systems can export visits with a creation time and a sign-in time. Count visits where the first is earlier than the second. Break the figure down by department or site; a low rate in one team is a training conversation, not a system fault.

Host response time

Definition: from the notification being sent to the host meeting their visitor, or acknowledging the notification if the system supports it.

How to collect: if the system records acknowledgements, export them. If not, ask reception to log the time a host appears for a sample of visits over a week. Long waits often point to notifications going to the wrong channel, or to hosts who are not in the building that day.

Sign-out completeness

Definition: the share of visits that ended with a sign-out by the visitor or reception, rather than being closed automatically at the end of the day or left open.

How to collect: export visits for a period and count those with a genuine sign-out. Check how your system marks automatic closures, so they are not counted as sign-outs. This is the measure that most directly predicts whether your evacuation list will be right.

Roll-call accuracy

Definition: in a drill, how closely the system’s on-site list matched the people actually at the assembly point.

How to collect: at your next fire drill, have wardens reconcile the live list against the people present, and record three counts.

Reconciling the roll-call list in a drillTwo overlapping circles. The left circle is people on the system’s on-site list; the right circle is people actually at the assembly point. The overlap is accounted for. The left-only area is people on the list but not found, usually visitors who left without signing out. The right-only area is people found but not on the list, usually visitors who never signed in, which is the more serious gap.On the system listAt the assembly pointAccountedforOn list,not foundleft withoutsigning outFound,not on listnever signed inWastes search timein a real emergencyMore serious:nobody wouldlook for them
Record all three counts after every drill. The left-only group points to sign-out habits; the right-only group points to people bypassing sign-in, and deserves attention first.

Run the reconciliation the same way each time, and agree it with your fire risk assessor so the drill still meets its main purpose. The aim is to test the list, not to slow down the evacuation.

Step 3: baseline, change, re-measure

A single measurement tells you little. Its value comes from comparison with your own earlier figures. Take a baseline before making any change, then change one thing at a time so you know what made the difference.

The measurement cycleA four-step cycle shown as connected boxes: define the measure and write it down; take a baseline over a set period; change one thing, such as a new reminder or a moved kiosk; then re-measure the same way over the same length of period, and either keep the change or revert it. An arrow returns from the last step to the second, starting the cycle again.1DefineWrite the measuredown precisely2BaselineMeasure over aset period3Change one thingA reminder, a movedkiosk, new training4Re-measureSame method, thenkeep or revertThe new figure becomes the next baseline
Changing several things at once makes the result impossible to interpret. One change per cycle is slower but tells you what worked.

Some changes are worth testing this way: a sign-out prompt on the badge, a kiosk moved nearer the exit, a reminder to hosts the day before a visit, a second notification channel. Each one should show up in a specific measure. If it does not, consider reverting it.

Watch for distortions

Any measure can be gamed once people know it is watched. If reception is asked to raise sign-out completeness, the quickest route is to sign everyone out at five o’clock, which improves the number and makes the evacuation list worse. Pair each measure with a check on the behaviour behind it, and explain to staff why the measure matters rather than only that it is being tracked.

Seasonal patterns matter too. Compare like with like: a week of contractor works against another, a quiet holiday period against another quiet period.

Step 4: report a few numbers, regularly

Pick the measures that matter most to you and report them monthly to whoever owns the visitor system, with a short note on anything that moved. Five numbers on one page, with the same definitions every month, will do more than an elaborate dashboard that nobody trusts.

Report sign-out completeness and roll-call accuracy to whoever is responsible for fire safety on site, as well as to facilities. A drop in either is a safety issue, and should be treated with the urgency of one.

If you run several sites, use identical definitions everywhere before comparing them. A site that times “reaching the kiosk” and another that times “joining the queue” are not measuring the same thing.

Where to go next

For the full drill procedure behind roll-call accuracy, read running an evacuation roll-call. If you are standardising measures across a portfolio, see rolling out visitor management across multiple sites. And if your measures suggest the system itself is the problem, our guide to writing a visitor management RFP and the directory are the places to start looking for a replacement.