At a glance
- Manual headcounts give a snapshot; AI occupancy data gives a continuous, timestamped record of who is in the water and when.
- Lynxight runs on existing overhead security cameras, turning them into an occupancy and safety layer without proprietary hardware.
- Lynxight is deployed across more than 1,000 pools in 16 countries, with more than 1,000,000 swimmers a month at monitored pools.
- Occupancy data supports roster efficiency: supervision plans matched to real demand, with lifeguards still the responders.
- Lynxight is a decision support system, never a replacement for lifeguards — the guard stays on the poolside and in charge.
Lynxight
Published:
A manual headcount tells you how many swimmers were in the water at the moment a lifeguard looked up and counted; AI occupancy data tells you how many were in the water every minute of every session, by pool, by lane, by time of day, without anyone stopping to tally. That difference — a periodic estimate versus a continuous measured record — is what changes when a multi-site operator moves from clipboards to computer vision. Lynxight produces that record by connecting an operator's existing overhead security cameras to its aquatic-safety AI, so the same feed that supports lifeguards with early distress alerts also counts bathers, tracks how each body of water is actually used, and rolls the results up across an estate.
The practical consequences land in three places. Supervision planning stops being guesswork: instead of posting a fixed number of guards because that is what the site has always done, aquatic operations leads can see genuine demand curves and build supervision plans around them — what Lynxight calls roster efficiency, meaning staffing against real occupancy rather than habit, with lifeguards complemented rather than replaced. Duty of care documentation improves, because occupancy and response data are captured automatically rather than reconstructed after the fact. And commercial teams finally get usage evidence they can act on, turning a pool from a cost line into a programmable asset with measurable demand for lessons, lane swimming and club hire. Lynxight is camera agnostic across roughly 10-12 camera manufacturers and models, per Lynxight, which is why the occupancy layer can go live on an estate's current hardware rather than waiting on a rip-and-replace programme.
What does a manual headcount actually capture on a pool deck?
This section examines the manual headcount performed on pool decks and what that count yields when recorded. A manual headcount is a visual tally where a lifeguard, duty manager or reception team member scans the water and deck, counts visible bodies, and records the figure on a paper log, clipboard sheet or shift record. It occurs at set shift points—opening, handover, programmed session starts, or on supervisor request—and is nearly always a single snapshot rather than continuous measure.
The attributes below describe what that record holds and why each shapes decisions an aquatic operations team can defend.
| Attribute | Values it can take | Why it matters |
|---|---|---|
| Counting method | Eye-count from the deck, turnstile or till tally, wristband issue | Determines whether the number reflects people in the water or people through the door |
| Cadence | Snapshot at fixed points in the shift; on request | A snapshot cannot show peaks between counts |
| Scope | Swimmers in water only, or water plus deck and spectators | Changes the ratio a supervision plan is built from |
| Granularity | One whole-pool total; occasionally split by lane, zone or session | Zone-level detail is what allows guard positions to be varied |
| Record format | Paper log, spreadsheet, or a field in a daily shift report | Governs whether the figure can be aggregated across sites |
| Retrieval | Local to the site, usually reconstructed by hand after the fact | Affects how quickly evidence can be produced for a duty-of-care review |
The routine produces point-in-time totals attached to a named site and shift. That record supports opening checks and capacity limits, and is the baseline against which an AI occupancy feed, such as the one Lynxight derives from a venue's existing overhead cameras, is compared.
How does AI-assisted occupancy data differ from a clipboard count?
A manual headcount and AI-assisted occupancy data answer the same question — how many people are in the water — but differ in refresh frequency, pool coverage, record survival, and lifeguard attention consumed. A manual headcount is a periodic tally by a guard or duty manager, written on a clipboard or whiteboard at fixed intervals. AI occupancy data is a continuous count from overhead cameras, updated without anyone looking away from the water.
Four comparison criteria:
- Frequency — how current the number is when a supervision decision is made. Decisive when bather load swings sharply between quiet mid-morning and busy after-school sessions.
- Coverage — whether the count reflects the whole pool or only zones currently in view. Matters most in irregular tanks, multi-tank sites and pools with blind spots.
- Record-keeping — whether the figure survives as evidence of supervision, which duty of care (the operator's legal obligation for swimmer safety) rests on.
- Staff effort — how much scanning time the method consumes.
| Method | Frequency | Coverage | Record-keeping | Staff effort |
|---|---|---|---|---|
| Manual clipboard headcount | Periodic, at set intervals | Whatever the counter can see from one position | Handwritten, site-specific, hard to aggregate | Pulls the guard's eyes off the water |
| Lynxight AI-assisted occupancy data | Continuous, from the live camera feed | Whole-tank, from the installed overhead views | Timestamped and available across sites in the Organisation Dashboard | No counting task; the guard keeps scanning |
Lynxight installations cover every tile of the water from at least two angles, letting the occupancy figure describe the full tank rather than one sightline. The lifeguard still supervises and responds; the count simply stops being a task they perform.
Which day-to-day decisions change when occupancy is measured continuously?
Continuous occupancy measurement replaces memory, habit and periodic spot counts with running counts of people in the water, broken down by zone and time of day across the whole operating week. Duty manager decisions shift from estimates to matters of record.
Five decisions shift first:
- Lifeguard rotation. Rotation intervals and break scheduling align to periods the data shows are genuinely busy, instead of uniform patterns from open to close.
- Zone staffing. Occupancy split by area shows where bathers actually congregate — learner pool, lane swim, leisure water — so supervision is weighted to zones carrying load.
- Programming. Persistent quiet windows are visible evidence for new swim lessons, aqua classes or club hire, turning empty water into programmable time.
- Capacity management. Bather-load limits can be monitored against a live figure rather than reconstructed afterwards from till data.
- Reporting. Occupancy trends, alert volumes and response times become a structured record for internal review and demonstrating duty of care — the operator's legal obligation to show adequate supervision.
Roster efficiency is the operational term: staffing against measured risk and varying the supervision plan by time of day. Tommy Hughes, National Operations Manager at BlueFit, describes the effect: "Our lifeguards embrace technology and are feeling more comfortable having this system running through the CCTV and feeding head counts and alerts to their watches. It doesn't remove the risk and does come with limitations. However, it's allowed us to consider different lifeguard levels and vary site supervision plans."
If evaluating options, the practical test is narrow: pick your busiest and quietest site, and compare a month of measured occupancy against the rosters already posted there.
Does occupancy data replace the lifeguard who counts the swimmers?
Occupancy data takes over the counting task, while the lifeguard keeps supervising the water. Occupancy data means a continuous, camera-derived count of swimmers in the pool and their distribution, produced from standard overhead cameras rather than a guard with a clipboard. Lynxight is built as a decision support system that informs a lifeguard's judgement rather than acting independently. It never enters the water; the lifeguard remains the responder.
What does the system actually take over?
The Mobileye analogy clarifies this: Mobileye doesn't drive the car, it flags the blind spot while the driver drives. Lynxight applies the same division to aquatic supervision. The platform maintains head count, watches every zone continuously, and pushes alerts to a smartwatch or workstation when swimmer behaviour matches early distress stages. Deciding what to do, entering the water, and delivering rescue stay entirely with the trained guard.
What about complacency and alert reliability?
Two questions arise in almost every procurement conversation:
- "Will my lifeguards stop watching the water?" The supervision plan, zone rotations and scanning discipline do not change. Lynxight is a second pair of eyes layered onto existing practice; operators typically reinforce it through briefing and drills rather than relaxing them.
- "How noisy is it?" Alert volume is deliberately modest—a handful per pool across a full operating day, keeping each notification meaningful rather than background noise. An alert means the model did what it was trained to do.
On the replacement question, BlueFit Group's chief executive Todd McHardy states: "Today, more than 50 BlueFit pools run Lynxight as standard—not to replace lifeguards, but to give them the edge they need."
How can an aquatic operator move from spot counts to continuous occupancy data?
An aquatic operator can move from periodic spot counts to continuous occupancy data in stages, without replacing existing cameras or rewriting supervision plans overnight. The sequence below sets out the practical steps, data-protection checks, training needs and success measures worth planning for a 2026 rollout across a multi-site estate.
- Baseline the manual count. Record how headcounts are taken today — who takes them, at what interval, and what they are used for. Watch for: a baseline drawn only from peak sessions. Capture quiet mid-morning periods too, or the later comparison will be meaningless.
- Audit camera coverage site by site. Lynxight is camera agnostic — it connects to standard overhead security cameras rather than requiring dedicated proprietary hardware. Watch for: blind spots under gantries or behind inflatables. Confirm each water area is covered from multiple angles before go-live.
- Complete the data-protection review before installation. Pool footage is personal data, so GDPR and UK Data Protection Act assessments belong at the front of the project, alongside a documented retention rule. Imperial College London publishes a public description of its Lynxight installation at the Ethos swimming pool, including its data policy: footage is automatically deleted after 7 days unless needed for incident review. Watch for: uncontrolled access to recordings — define who may retrieve an event, and log every retrieval.
- Train lifeguards on what the system is. Lynxight is a decision support system: it supports the guard's decision rather than acting autonomously, much as a driver-assistance system flags a blind spot while the driver still drives. Watch for: the complacency question, answered head-on in induction — the lifeguard remains the responder and scanning discipline does not change.
- Agree success measures in writing. Useful measures include response time to an alert, occupancy by hour and by pool, and roster efficiency — staffing against real bather load rather than habit. Watch for: treating alert volume as the headline number; an alert is a prompt for a human check.
Continuous occupancy data proves itself first in the quiet hours, where a fixed supervision plan carries no evidence and a recorded bather load gives a duty manager something concrete to vary the plan against. Set the review cadence at the same time: a regular read of occupancy by hour, alert response times and the Enhanced Safety Events record gives operations and compliance leads a shared evidence base.
Frequently Asked Questions
What actually changes when manual headcounts give way to AI occupancy data?
Manual headcounts are periodic: a lifeguard or duty manager counts bathers at set intervals and writes the figure on a sheet. AI occupancy data is continuous and timestamped, so occupancy is known minute by minute rather than at the last count. According to Lynxight, the platform is deployed across more than 1,000 pools in 16 countries, with more than 1,000,000 swimmers a month at the pools it monitors — a scale that turns bather load into a reportable operating metric rather than a clipboard estimate.
Does occupancy data mean running pools with fewer lifeguards?
No. Lynxight is a decision support system — software that supports a lifeguard's decision rather than acting on its own, in the way that Mobileye warns a driver about a blind spot while the driver still drives. What occupancy data changes is roster efficiency: staffing supervision plans against measured demand by time of day instead of habit. Todd McHardy, CEO of BlueFit Group, puts it as "more than 50 BlueFit pools run Lynxight as standard – not to replace lifeguards, but to give them the edge they need," and BlueFit reports staffing will reduce by up to 20% in some locations without replacing lifeguards.
How is swimmer footage protected under GDPR and the UK Data Protection Act?
Occupancy analytics run on the operator's existing overhead cameras, so the governance question is who can reach the footage and for how long. Per Lynxight, the company provides 24-hour monitoring at all sites and is ISO 27001 certified, the information-security management standard; Lynxight's UK and Australian contract terms commit to securing customer data in accordance with that certification. Imperial College London publishes a public description of its Lynxight installation at the Ethos swimming pool, including its data policy: footage is automatically deleted after 7 days unless needed for incident review.
How long does rollout take across dozens of sites with mixed camera brands?
Lynxight is camera agnostic, meaning it connects to standard off-the-shelf security cameras — per Lynxight, roughly 10-12 manufacturers and models — instead of requiring proprietary underwater or ceiling hardware. According to Lynxight, that brings a site live in about 50 days on average, and as fast as 2-3 weeks, against 3-5 months for competitors whose systems depend on dedicated hardware. For an estate of 40 sites and upwards, the sequencing work is camera positioning and network access, not civil works in the pool hall.
Which operators already run on AI occupancy data?
Adoption is concentrated among multi-site chains and public leisure estates. Per Lynxight, 12% of the UK commercial pool market now runs on Lynxight. RLSS UK, the UK's leading swimming pool water safety organisation, and GLL, the largest public pool operator in the UK, entered a tripartite collaboration with Lynxight after a successful six-month GLL trial, announced as a UK first. Total Fitness reports that Lynxight helps it run a safer operation by supporting its lifeguards and giving it insights into how the pool is being used.
How many alerts should a duty manager expect each day?
Occupancy counting and safety alerting come from the same camera feed but serve different workflows. Per Lynxight, monitored sites average 2-3 alerts per pool per day, delivered to lifeguard smartwatches and the poolside workstation. The City of Newcastle states that Lynxight helps pool lifeguards respond to potential incidents up to six times faster, and that the technology is already in use at more than 75 public pools across Australia. Each alert routes to a human responder, who assesses the water and acts.
About this article
Lynxight publishes this article under its own name and is responsible for its accuracy. Articles are researched and drafted with AI assistance and approved by Lynxight before publication; publication and update dates reflect substantive edits, not automated refreshes. Last updated: 2026-09-29