Escape of Water and Accidental Water Damage Claims in Indian Commercial Buildings 2026: Data Centres, High-Rises and the Slow-Leak Exclusion

There is no single peril called 'escape of water' in the standard Indian property contract. What global wordings bundle under that phrase is split across two named perils in the Standard Fire and Special Perils Policy (SFSP) structure that IRDAI carried into its 2021 standard products. Peril 9 is bursting and/or overflowing of water tanks, apparatus and pipes. Peril 11 is leakage from automatic sprinkler installations. Both respond to water that suddenly gets out of a system it was meant to stay inside.

Since the April 2021 standardisation, the container depends on the size of the risk. A commercial building insured up to Rs 5 crore sits under Bharat Sookshma Udyam Suraksha; between Rs 5 crore and Rs 50 crore it is Bharat Laghu Udyam Suraksha; above Rs 50 crore, insurers use their own filed fire products priced off IIB burning-cost data. All three keep the burst-pipe and sprinkler-leakage perils, but the sub-limit and add-on architecture differs, and that is where data-centre and premium-office losses get decided.

For a broker placing a server hall or a fitted-out floor in a Grade-A tower, the practical point is that water damage is not a standalone cover. It is a consequence peril tucked inside a fire policy, and the fire policy was written around structural loss, not around chilled-water pipes running above raised floors. The fire policy will pay for the servers that a burst pipe soaks, but only after the claim clears the sudden-and-accidental test, the maintenance exclusion, and whatever inner limit sits on the water peril. Understanding that stack before the loss, not after, is what separates a paid claim from a disputed one.

The single most important distinction in a water-damage file is temporal. The named perils respond to a discrete, identifiable event: a tank ruptures, a joint fails, a valve lets go, a sprinkler head discharges. They do not respond to water that has been finding its way in slowly for weeks. That is the difference between a covered burst and an uncovered seepage.

The doctrine doing the work here is proximate cause. Under Section 55 principles of English-derived Indian insurance law and settled market practice, the insurer asks what the dominant, effective cause of the damage was. If a chilled-water line in a data centre fails suddenly and floods a server rack, the proximate cause is the pipe failure, a covered peril. If the same rack corrodes over months because a fitting has been weeping and nobody acted, the proximate cause is gradual seepage plus neglect, and the exclusion bites.

Brokers should test this at placement. Ask the underwriter, in writing, how the wording treats a sudden failure that follows a known slow leak, and whether any gradual-water extension is available. Some filed products offer a limited seepage and gradual leakage buy-back with a modest inner limit and a higher deductible. For a Rs 200 crore data-centre fit-out, even a small carve-back changes the negotiating posture when the surveyor's forensic timeline is ambiguous. The wording decides who carries the doubt.

Every Indian fire and property wording carries a general exclusion for loss caused by or attributable to wear and tear, gradual deterioration, corrosion, rust, and lack of maintenance. This is not a minor clause. It is the provision most frequently invoked to reduce or decline water-ingress claims on commercial buildings, and it is doing more work each year as buildings age and mechanical systems run harder.

The exclusion operates in two ways. First, it removes the deteriorated component itself: if a corroded pipe bursts, the cost of the pipe is not covered even where the resulting water damage to servers is. Second, and more dangerously, it lets the insurer attack the whole claim by arguing that lack of maintenance was the real proximate cause of the failure. A rusted-through tank base, a sprinkler system that missed its scheduled flow test, a pump gland that had been leaking for a quarter: each gives the surveyor a hook.

What defeats the exclusion is a documented maintenance regime. For data centres and high-rises this means:

• Annual sprinkler flow and pressure tests with dated certificates, aligned to the fire NOC conditions.
• Planned preventive maintenance logs for chilled-water plant, pumps, tanks and pressurised lines.
• Leak-detection systems under raised floors and in plant rooms, with alarm records showing prompt response.
• A defect-rectification trail proving that flagged issues were closed, not carried forward.

Without this, the insurer's position is simple: you cannot prove the failure was sudden rather than the end of neglect, so the exclusion applies. With it, the burden effectively shifts. The surveyor must then explain why a well-maintained system's failure was anything other than a covered accident. Maintenance evidence is not paperwork for auditors; it is claims evidence, and it should be filed as such.

A data centre can carry a nine-figure sum insured on the building and contents, yet recover a fraction of a water loss because the water peril sits under an inner sub-limit that nobody negotiated up. This is the quiet trap in 2025-2026 placements.

On larger filed fire products, insurers routinely apply an inner limit to the burst-and-overflow peril and, separately, to sprinkler leakage. The limit may be expressed as a rupee cap or as a percentage of the sum insured, and it is often set with a warehouse or factory in mind, not a room full of servers where a single accidental discharge above a hot aisle can destroy tens of crores of equipment in minutes. When the inner limit is Rs 5 crore and the electronic loss is Rs 40 crore, the sub-limit, not the sum insured, defines the recovery.

The interaction with electronic equipment insurance matters here. EEI is written on an all-risks basis and can respond to accidental water damage to the equipment itself, often on better terms than the fire policy's water peril, but it too carries sub-limits, a data-restoration cap, and its own deductible. Brokers should map, line by line, whether a soaked server rack is recovered under the fire policy's inner water limit or under the EEI all-risks section, and make sure the two do not leave a gap where each insurer points at the other. Aligning the deductibles and confirming which policy is primary for water damage to electronics is basic placement hygiene that is skipped far too often.

Water-damage claims are won or lost on the reconstruction of what happened and when. Unlike a fire, where the burn pattern and the fire-brigade report anchor the event, water leaves an ambiguous trail that both sides can read differently. The insured says sudden burst; the insurer says slow seepage. Whoever has the better evidence of sequence controls the outcome.

The evidence that holds up, in rough order of weight:

1. The failed component, preserved. Do not discard the burst pipe, split tank or failed valve. A metallurgical or forensic plumbing report on the actual part is the strongest proof of a sudden mechanical failure versus long-term corrosion.
2. Time-stamped detection data. Leak-detection alarms, BMS logs, and CCTV showing dry floors minutes before and flooded floors after establish suddenness better than any statement.
3. The first-response record. The maintenance callout log, the plumber's job card, and internal incident notes made on the day, before any claim strategy formed.
4. The maintenance history, as above, rebutting the neglect argument.

Appoint the licensed surveyor fast and give them the site before clean-up destroys the evidence. IRDAI's framework requires surveyor appointment without undue delay on larger losses, and the surveyor's material damage assessment will drive the settlement. A common and costly mistake is emergency remediation that strips out wet fit-out and dries the slab before the cause is documented, leaving the insured unable to prove the loss was accidental. Mitigation is a policy duty, but it must be paired with a preservation protocol: photograph, sample, and log before you demolish. Where the cause is genuinely contested, an independent forensic engineer instructed early, rather than a rebuttal filed late, is what shifts a disputed file toward settlement.

Given how the perils, exclusions and sub-limits stack, the placement work for a water-exposed commercial risk is specific. Generic fire cover on a data centre or a premium office floor will underperform on a water loss unless it is deliberately built out.

The checklist that matters:

• Negotiate the water sub-limits up to a figure that reflects a realistic single-event electronic loss, not a factory default. Price the higher limit; do not accept the standard cap by inertia.
• Confirm the gradual-seepage position and buy back a limited seepage extension where the plant risk justifies it, accepting the higher deductible as the cost of removing the ambiguity.
• Align fire-policy water cover with EEI so accidental water damage to servers has one clearly primary policy, matched deductibles, and an adequate data-restoration limit.
• Cover the fit-out on reinstatement value rather than indemnity, so soaked interiors, cabling and finishes are replaced new-for-old rather than depreciated.
• Check the average clause: under-insurance on a fit-out that appreciates with each upgrade will see any water claim proportionately reduced.
• Endorse in a leak-detection and maintenance warranty on your own terms, so compliance protects the claim rather than a breach voiding it.

For tenant fit-outs in GCC campuses and Grade-A towers, clarify in the lease and the policy who insures the base-building mechanical services and who insures the tenant's installations, because a burst in the landlord's riser that floods a tenant's server room raises subrogation and dual-insurance questions that are far easier to resolve on paper than in a loss.

Sarvada exists for exactly this comparison work. Rather than reading each insurer's fire and EEI wording line by line to find the water sub-limit, the seepage treatment and the maintenance warranty, brokers can search insurer policy wordings across the market in one place and see how each carrier handles escape-of-water on high-value electronic risks. If your team is placing data-centre, high-rise or fit-out property in 2026, Request Access to compare the water-damage terms that actually decide these claims.