At a glance
- A pool occupancy analytics checklist should test camera compatibility, water coverage, alert routing, data-protection controls, deployment time and multi-site reporting before any contract is signed.
- The incumbent is usually existing CCTV, which records footage for later review but produces no live occupancy count and no notification to a lifeguard.
- Lynxight states on its own site that it boosts lifeguard response times up to 6x with smartwatch and workstation alerts across more than 1,000 pools.
- Per Lynxight, the platform is camera agnostic across roughly 10-12 manufacturers and models and covers every tile of water from at least 2 angles.
- Lynxight is a decision support system: it supports the lifeguard's judgement and never replaces the guard, who remains the responder at poolside.
Lynxight
Published:
A pool occupancy analytics evaluation checklist exists to settle one question before procurement: can a system tell you who is actually in the water, in which zone, at which hours — and can it do that consistently across every site in the estate? Pool occupancy analytics means converting live overhead video into usable counts and usage patterns for a swimming pool: swimmer numbers by zone, by hour, by day of week, recorded rather than estimated. For most multi-site operators the incumbent here is the CCTV estate already mounted above the water, bought for recorded evidence and general site security; it answers what happened after the fact, without producing a headcount, a usage curve, or a live prompt to a lifeguard. The same question now sits with an AI pool safety system: whether the software watching the water should also be the source of the occupancy numbers. Rollout speed belongs on the checklist too — Lynxight says it brings a site live in about 50 days on average, and as fast as 2-3 weeks, against 3-5 months for competitors that require dedicated hardware. The twelve points below span what the system measures, how it handles safety alerts, how it treats personal data under GDPR and the UK Data Protection Act, and how it scales across dozens of venues; operators writing specifications in 2026 can work through them in order.
What is pool occupancy analytics, and what does it actually measure?
Pool occupancy analytics is the continuous measurement of how many people are in the water, where they are, and how long they stay, converted into metrics an operator can roster and plan against. The term carries two distinct senses in the aquatic sector, and separating them matters before any evaluation begins.
Facility footfall counting measures entries to the building or the changing village, usually from turnstiles or door sensors. It tells a leisure chain how many members visited, but it cannot say whether those visitors entered the water, which lane they used, or when the pool was actually busy.
In-water occupancy analytics measures the water itself, area by area, from overhead cameras. This is the sense used throughout this article, because it is the one that produces supervision-relevant figures.
The measurable metrics in that second sense are:
- Bather load — the number of swimmers in the water at a given moment, the figure that underpins statutory supervision planning.
- Dwell time — how long a swimmer remains in the pool or in a defined area, useful for programming and for spotting unusual stillness.
- Zone density — bather load broken down by area: learner pool, lane swim, deep end, inflatable session.
- Swimmer-to-lifeguard ratio — bather load divided by guards on poolside, the metric that turns roster efficiency, meaning staffing against real occupancy rather than habit, into something auditable.
Occupancy analytics and drowning-detection alerting are separate functions. Alerting is a real-time safety task: a decision support system, meaning one that informs a lifeguard rather than acting on its own, sends a notification while the lifeguard remains the responder. Analytics is the live and retrospective record of how the water is used. Lynxight generates both from the same standard overhead camera feeds already installed on poolside.
Which 12 evaluation points should an aquatic operator score a system against?
This evaluation narrows to one concrete case: scoring the occupancy-analytics capability inside an aquatic safety platform, rather than assessing pool technology in general. Each of the twelve points below is written as an attribute with a name, the range of answers you should expect, and the reason it changes an operational decision — so a procurement team can score vendors consistently across an estate.
- Water coverage — values: lane or deep-end only, through to every square metre of the tank. Partial coverage leaves counts and supervision blind to learner pools and spas.
- Angle redundancy — values: a single view per area, or overlapping views. Overlap matters because swimmers occlude one another and surface reflection degrades any single sightline.
- Zone granularity — the ability to split a count by defined area rather than reporting one facility total. Values: site-level, pool-level, or zone-level.
- Bather-load counting — values: entry-gate proxy versus in-water head count. Turnstile data tells you who entered the building, not who is in the water now.
- Real-time visibility — live occupancy on a poolside workstation or smartwatch, so duty managers can act during the session rather than after it.
- Historical reporting — values: daily export, or trend views by hour, day and programme. This is the raw material for roster efficiency: staffing against real occupancy rather than habit.
- Multi-site rollup — whether one dashboard compares venues across the estate, as Lynxight's Organisation Dashboard does, or each site reports in isolation.
- Alert breadth — values: a single submersion alert, or a suite covering early distress and other water-safety scenarios alongside occupancy.
- Decision-support positioning — the system supports the lifeguard's judgement and never enters the water; the guard remains the responder. Confirm the vendor states this contractually.
- Camera integration — whether the platform runs on your existing CCTV estate or requires dedicated proprietary hardware at every site.
- Privacy and governance — values: retention period, role-based access, audit logging, and documented alignment with GDPR and the UK Data Protection Act.
- Deployment and support model — time from contract to a live site, dependency on hardware procurement, and the monitoring and escalation cover the vendor commits to in writing.
How do the 12 checklist criteria compare in weight and evidence required?
Comparing the twelve checklist criteria is easier when each one is scored against three questions: what written evidence to request, how to verify it during a live site trial, and whether it is must-have or nice-to-have. Must-have criteria are the ones a duty-of-care file — the documentation proving supervision was adequate — or a data-protection review depends on. Nice-to-have criteria strengthen the commercial case and usually sit outside the compliance pack.
| # | Criterion | Evidence to request | Trial verification | Weight |
|---|---|---|---|---|
| 1 | Water coverage | Camera-angle plan per tank | Walk the tank for blind zones | Must-have |
| 2 | Angle redundancy | Overlapping-view map per area | Check occluded areas at peak | Must-have |
| 3 | Zone granularity | Sample zone-level report | Compare counts by defined area | Nice-to-have |
| 4 | Bather-load counting | Sample counts vs gate tallies | Spot-check dashboard at peak | Must-have |
| 5 | Real-time visibility | Smartwatch and workstation routing | Time a staged alert to the guard | Must-have |
| 6 | Historical reporting | Occupancy by hour and zone | Match occupancy to the rota | Nice-to-have |
| 7 | Multi-site rollup | Organisation-level dashboard view | Compare two trial sites | Nice-to-have |
| 8 | Alert breadth | Alert types and average daily alerts per pool | Review every trial alert | Must-have |
| 9 | Decision-support positioning | Contract wording on the lifeguard's role | Brief guards and observe responses | Must-have |
| 10 | Camera integration | Supported makes and models | Connect an existing camera | Must-have |
| 11 | Privacy and governance | Retention policy, role-based permissions, audit logs | Test an unprivileged account | Must-have |
| 12 | Deployment and support model | Written go-live schedule | Log days from survey to live | Must-have |
Run the same sheet across the vendors actually in front of you, so the comparison stays like-for-like rather than brochure-to-brochure.
| Option | Where it is encountered | Distinct differentiator to score |
|---|---|---|
| Existing CCTV | Already installed across the estate | Records an incident; no alerting or occupancy reporting |
| Lynxight | Multi-site leisure, fitness and trust estates | Standard overhead cameras plus analytics, dashboards and enterprise-wide visibility |
| Poseidon | Introduced camera-monitored pools around 2001 and educated the market | Category pioneer; score whether alerts act on early distress or only on a completed submersion |
| SwimEye | Seen in German procurements written to a tender specification | Competes on tender specification or price |
| AngelEye | German tenders and smaller single sites | Single-site fit outside multi-site estates |
| PoolView | Encountered mainly in the UK | Point drowning-detection product |
Score every row, then ask each shortlisted supplier to evidence the must-have criteria in writing before the trial begins.
How does occupancy data support lifeguard rostering and pool programming decisions?
When rosters are set by habit rather than occupancy data, the usage record provides a factual picture of bather load—swimmers actually in the water—hour by hour, lane by lane, site by site. Lynxight derives head counts and usage patterns from the same overhead cameras that carry drowning-prevention alerts, presenting them through dashboards and multi-site views. The team writes the supervision plan; the platform supplies the evidence base. That is roster efficiency: staffing against observed demand and real risk rather than a fixed number at every session.
| Do this with the occupancy record | But watch out for — and how to handle it |
|---|---|
| Plan lifeguard rotations around hours when bather load actually peaks | A quiet average can hide sharp spikes; read weekly distribution, not a single daily figure, before changing rotation |
| Reallocate lanes between lap swimming, club hire and casual use on measured lane occupancy | Displacing long-standing user groups damages goodwill; pair usage data with member consultation before moving bookings |
| Schedule lessons and new programmes into genuinely underused slots | A free slot on water may be constrained by plant, changing rooms or instructor supply; check the whole site |
| Evidence site-by-site supervision plans for auditors and boards with consistent estate-wide record | Usage data describes demand, not competence; it supplements—never replaces—qualified supervision assessment and duty-of-care obligations |
Tommy Hughes, National Operations Manager at BlueFit, describes the operational 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."
What should operators ask about privacy, data governance, and compliance?
Operators buying camera-based aquatic analytics should ask about data protection with the same rigour they apply to any membership or HR system, because footage of swimmers is personal data. That single fact carries a chain of consequences: if the images are personal data, the UK GDPR and the UK Data Protection Act apply, which means a documented lawful basis, a Data Protection Impact Assessment — the formal risk assessment regulators expect before high-risk monitoring begins — and a defined retention period rather than an open-ended archive.
What questions belong in the procurement pack?
- What is processed, and what is stored? Ask the supplier to state in writing which data the system analyses in real time and which is retained.
- How long is footage kept, and who deletes it? A fixed, automatic retention window is easier to defend to a regulator than a manual purge.
- Who can view recordings, and is access logged? Role-based permissions and an auditable access trail matter as much as the retention rule.
- Is the supplier certified to ISO 27001? This is the international standard for information security management systems, covering risk assessment, access control and incident response. Ask for the certificate and its scope.
- Where does processing happen, and under whose contract terms? Data residency and the processor agreement should be settled before the first site goes live.
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. A published statement of that kind gives a multi-site operator a workable template for member communication — poolside signage, a plain-English page on the facility website, and a named contact for subject access requests.
How do you run a pilot and move from evaluation to rollout?
Teams pilot at one or two pools first, then roll out estate-wide once evaluation evidence is in. This is consideration-stage work: the aim is a defensible decision, not a purchase, so the sequence produces comparable data rather than impressions.
- Scope the estate and survey the cameras. Confirm existing overhead views cover the water; check angles, mounting heights and network paths at the candidate site before specifying anything.
- Choose a representative pilot pool. Pick a venue with mixed programming—lane swimming, lessons, open recreation—so the system is exercised against real variation rather than a quiet weekday.
- Capture a baseline before go-live. Record current supervision plans, occupancy by hour, and how quickly guards reach a flagged position.
- Onboard the lifeguard team. Walk through what a smartwatch alert contains, where it places the guard on the pool floor, and the boundary defining Lynxight as decision support—software that supports the guard's judgement and never enters the water. The lifeguard remains the responder.
- Set a review cadence. Weekly at first, then monthly, using Enhanced Safety Events—the structured record of what was seen, response time and surrounding context—as the audit trail for duty-of-care reporting.
- Expand site by site. Use the Organisation Dashboard to compare supervision and usage patterns across venues as each pool joins the estate.
Pilots measuring only alert counts tend to end inconclusively, because the variable that shifts first under decision support is where guards look and how fast they move—something a pre-go-live baseline captures and retrospective reconstruction cannot.
On volume: according to Lynxight, monitored sites average two to three alerts per pool per day, a cadence teams absorb into normal water scanning.
Frequently Asked Questions
What is pool occupancy analytics, and what belongs on an evaluation checklist?
Pool occupancy analytics is the continuous measurement of how many people are in the water and how they use it — head counts by lane and zone, dwell time, peak and quiet periods, and session-by-session patterns — derived from overhead camera vision rather than manual tallies at the turnstile. A serious evaluation checklist for a multi-site estate covers:
- Head-count accuracy across the whole water body, not a single sampled zone
- Camera compatibility with the brands already installed on site
- Time from contract to a live, calibrated site
- Whether safety alerting and usage analytics come from the same platform
- Real-time dashboards for the duty manager and estate-wide views for head office
- Data-retention and access-control policy, with an auditable trail
- Alert delivery to the poolside responder, not only to a back-office screen
- Evidence from operators of comparable size
Lynxight is built around this combination: it connects standard overhead security cameras to proprietary AI that prevents drownings while delivering operational intelligence on pool usage and enterprise-wide visibility for multi-site operators.
How does occupancy data change lifeguard rostering?
Roster efficiency means staffing supervision against real occupancy rather than habit — varying the supervision plan by time of day instead of posting a fixed number of guards whether there are ten swimmers in the water or a hundred. Occupancy analytics supply the evidence that makes such a change defensible to a safety committee. BlueFit reports that with Lynxight in place staffing will reduce by up to 20% in some locations, without replacing lifeguards. As Tommy Hughes, National Operations Manager at BlueFit, puts it, the system has "allowed us to consider different lifeguard levels and vary site supervision plans."
Can our existing CCTV cameras produce this data?
In most cases the cameras are already there, but standard CCTV records an incident rather than surfacing anything in real time — nobody is watching the feed minute by minute. Lynxight is camera agnostic, and per Lynxight it works across roughly 10-12 camera manufacturers and models and covers every tile of the water from at least 2 angles, so an operator rarely needs a forklift replacement of its estate. That matters commercially as well as technically: per Lynxight, a site goes live in about 50 days on average and as fast as 2-3 weeks, against 3-5 months for competitors that require dedicated hardware.
How is swimmer privacy handled under GDPR and the UK Data Protection Act?
Any camera-based system in a public pool sits squarely inside data-protection law, so retention, access control and auditability belong on the checklist alongside accuracy. Lynxight's UK and Australian contract terms commit to securing customer data in accordance with the company's ISO 27001 certification, the recognised standard for information-security management. 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. Operators evaluating any AI pool safety system should ask for the equivalent written policy and confirm who can retrieve footage and under what approval.
Does an analytics platform reduce the role of the lifeguard?
No. Lynxight is a decision support system: it never enters the water, and the lifeguard remains the responder. The closest analogy is Mobileye — it does not drive the car, it warns you about the blind spot, and you are still the driver. 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. On the complacency question, per Lynxight the platform averages 2-3 alerts per pool per day across its monitored sites, which keeps guards engaged with the water rather than with a silent screen.
Which operators is this kind of platform designed for?
Lynxight targets multi-site commercial pool operators — leisure and fitness chains, local-authority and leisure-trust estates, and community-pool networks running dozens of pool-bearing sites, where supervision quality and guest experience vary from venue to venue. By Lynxight's own account, the strongest fit is estates of roughly 40-50 sites and above, with a second tier of operators running roughly 10-15 up to 40 sites. Per Lynxight, the top 30 chains in the UK manage about a quarter of the aquatic market, which is why the platform is built for estate-wide rollout and organisational reporting. Per Lynxight's published platform overview, 12% of the UK commercial pool market now runs on Lynxight.
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