Starting an aircon-servicing business in Singapore, decoded.
It looks like the perfect trade: a hot island where 82% of homes run air-conditioning, demand that never switches off, and a van-and-tools barrier to entry anyone can clear. That low barrier is the trap. The result is a crowded price war where a general service is almost a loss-leader. Here is the picture the “easy money” pitch skips: where the money actually is (the upsell ladder), the rules that really apply (refrigerant and working-at-height, not a mythical “aircon licence”), and the one thing that turns this into a real business — a book of recurring contracts.
81.9%
of households own aircon (2023)
~S$1
min. paid-up capital to start
4×/yr
quarterly contract = the recurring base
No licence
to service — only refrigerant is gated
The MOAT Score: is aircon servicing worth building?
Before the details, our one-number read. The MOAT Score grades a sector's economic quality on the value-investing lens of Graham, Buffett and Munger — four pillars (Margin, Operating moat, Appetite, Treadmill), each out of 25. Aircon servicing scores higher than the one-off home trades for one structural reason: the demand is near-essential and it recurs.
Needs a real edge
Only with a real edge.
How the score is built
The MOAT Score sums four pillars — each scored 0–25 — from the value-investing lens of Graham, Buffett and Munger. No black box: here is the working.
Margin
10/25Does the average operator actually keep money — real net margin and return on the capital tied up?
This sector: asset-light, and the upsell ladder (gas top-up carries an ~85–95%Estimate / modelled, not a measured figureSGAI calc on SG refrigerant wholesale vs retail, 2026R32 gas costs ~S$3–10 per top-up at SG wholesale (~S$8–12/kg, 0.4–0.8kg) and sells for S$50–80 — a wide materials margin. Modelled, not an audited figure. materials margin) keeps real money — but a general service is near loss-leader, and the only audited proxy, ISOTeam, nets ~4–5%SourceISOTeam FY2025 audited (SGX:5WF), 2025.
Buffett 1979 — “a high earnings rate on equity capital… without undue leverage”; 1986 owner earnings.
Operating moat
11/25Pricing power and a durable competitive advantage — can a typical operator raise prices and have customers shrug?
This sector: no aircon licence and no CaseTrust kitemark, so no statutory or brand moat — but a recurring quarterly contract is a genuine, if modest, switching cost that one-off trades structurally lack. Above renovation for exactly this reason.
Buffett, FCIC 2010 — pricing power is “the single most important decision”; 1991 franchise; 2007 moat.
Appetite
19/25Demand durability — steady, recession-resistant repeat demand vs fragile, discretionary or faddish.
This sector: the standout: tropical demand never switches off, aircon is near-essential not discretionary (81.9%SourceSingStat Household Expenditure Survey, 2023 of homes own it), and unlike a one-off renovation the need recurs.
Graham, Security Analysis Ch.2 — inherent stability “derives from the character of the business”.
Treadmillinverted · less is better
13/25Capital intensity and structural drag — rent, churn, fashion, discounting. Scored inverted: less treadmill, more points.
This sector: light capital (a van + tools), and a contract book means revenue isn't re-won from zero each job — but a brutal price war, high paid-search acquisition cost, and a tight foreign-worker manpower regime drag it back.
Buffett 2007 — “the worst sort of business… requires significant capital… Think airlines.”
M + O + A + T, out of 100
The MOAT Score is a transparent SGAI judgement on a sector’s economic quality through a value-investing lens — not a verdict on any individual business, and not a comment on an owner-operated livelihood (a sector can score low on capital returns yet work as a job).
The read: a one-off job is a price war; a contract is a business
The thing that makes aircon servicing attractive — almost no barrier to entry — is exactly what makes the average job unprofitable. You can register a company for S$1SourceACRA minimum paid-up capital, 2026 of paid-up capital, buy a starter tool kit and a van, and you are a competitor. So everyone is. The predictable result is a race to the bottom on the one service customers can easily compare: general servicing, advertised “from S$16–30 a unit” at volume. At those prices, as one operator put it bluntly on a local forum, “how to make money like that?”
The answer is that general servicing is not where the money is — it is the foot in the door. The money is on the upsell ladder: chemical wash, gas top-up, chemical overhaul, and ultimately a new install. A gas top-up sells for S$50–150 on a few dollars of refrigerant. That economics is also why the trade has a trust problem — the same ladder that is legitimate when an aircon genuinely needs it becomes the “you need a top-up every visit” overcharge when it doesn't.
And here is the structural fork that decides everything. A one-off job is a customer you paid to acquire (search clicks for aircon keywords run ~S$4–9Estimate / modelled, not a measured figureSG PPC agency benchmarks (First Page Digital, 2Stallions), 2025Agency-published Singapore home-services CPC ranges; no measured SG CPC for the exact phrase is public. The US 'air conditioning services' vertical ran ~US$9/click (LocaliQ). each) and may never see again. A quarterly servicing contract is the same customer, four guaranteed visits a year, revenue you don't re-win — and four paid chances to sell the ladder honestly. The one-off trades (renovation, handyman) can never have this; aircon can. That recurring base is the closest thing to a moat in the entire home-services world.
Why the money isn't in the service — it's up the ladder
A general service is a near-commodity that operators discount to win the door. Each rung up the upsell ladder carries a far bigger ticket on a small materials cost — which is where a contract visit pays for itself.
- General servicing (per unit)S$35
the price-war rung; near loss-leader at volume
- Gas top-upanchorS$100
wide materials margin — and the #1 overcharge if not from a real leak
- Chemical wash (per unit)S$115
- Chemical overhaul (per unit)S$250
full dismantle; S$120–450 by unit type
Source: SGAI synthesis of SG aircon-company price lists (Billy Aircon, SoCool, Lion City, 338 Aircon), 2024–2026 (indicative)
The map: ~1.2M air-conditioned homes — and the “market-size” numbers everyone quotes are junk
There is no credible published “Singapore aircon servicing market size.” The aggregator reports that rank for it measure air-conditioner equipment sales, not servicing, and they contradict each other by an order of magnitude — from ~US$80MShaky figure — treat with cautionmarket-size aggregators (EMR/TechSci)These size air-conditioner equipment sales, not servicing. EMR/TechSci ~US$79–84M vs Deep Market Insights ~US$870M vs Mordor 'HVAC' US$3.6B — irreconcilable, ~45× spread. One vendor even republished an equipment figure relabelled as 'services'. to US$3.6BShaky figure — treat with cautionMordor 'Singapore HVAC market'Tracks the sale of HVAC product types — equipment, not a servicing line. Disagrees with the air-conditioner aggregators by ~45×.. Ignore them. Size it from the ground up instead: about 1.2 million air-conditioned households (1.46M resident households × 81.9% ownership), each with a few fan-coil units that, on the industry's own advice, want servicing several times a year.
81.9%
of households own aircon
SingStat HES 2023 — up from 76.1% (2012/13)
~1.46M
resident households
SingStat 2024
~24%
of home electricity is aircon
NEA household survey (2017) — dated; among the top end-uses
No clean count
of aircon firms exists
SSIC 43220 bundles plumbers; ~280 NEA-certified servicers
The honest sizing (built up, not looked up)
~1.46M resident households ×81.9%SourceSingStat HES, 2023 ≈ ~1.2M air-conditioned homes, each with roughly 2.5–3 indoor units (a modelling assumption — no official figureEstimate / modelled, not a measured figureno SingStat/NEA data on units per householdUnits-per-household is not published; ~2.5–3 is a trade-norm assumption (a 4-room HDB typically runs a 'System 3'). Treat the resulting market size as a modelled order of magnitude, not a measured number. for units per household) → on the order of 3–3.6M indoor units in service nationally. The inputs are official; the multiplication is ours. That is how you size this sector honestly — and it is why “the market is US$80M” (an equipment figure) is not the number you build a plan on.
Even the firm count is murky: the relevant SSIC code (43220) bundles plumbers and heating installers with aircon, and the cleanest register — NEA's list of servicing companies with certified installers — held roughly 280 firms. The competing population is best read as a few hundred active servicers, many running several near-identical SEO microsites that inflate the apparent count.
The players: a long tail of SMEs — and a graveyard of platforms
Almost no residential aircon brand publishes audited accounts — the financial opacity is itself the finding. This is a long tail of price-competitive private firms, not a sector with reportable scale players. The more instructive lesson is who tried to consolidate it and died: the venture-funded “marketplace for trades” thesis is a graveyard in Singapore.
The survivors own technicians or own an adjacent business
| Player / model | Model | Recurring revenue? | Alive in 2026? | Signal |
|---|---|---|---|---|
| Everyworks GroupDW Aircon + Mr Plumber + Electrician (owned techs) | Multi-trade, company-owned technicians | The closest thing to a roll-up — integrated, not a marketplace | ||
| Billy Aircon / LK BrothersLong-running independent servicers | Servicing + contracts + install, SEO-led | Contract books + 1,000+ reviews; private, no accounts | ||
| Carousell ServicesMarketplace inside a bigger business | ~2.5% take-rate (capped S$30/txn) | Survives because it is subsidised by the wider platform | ||
| Kaodim / Sendhelper / Mr RightVC-funded trade marketplaces | Lead-gen / booking commission | All dead — Kaodim raised ~US$17.6M, ceased 2022 | ||
| ISOTeam (SGX:5WF)Listed building-maintenance / M&E proxy | Project + maintenance contracts | The audited anchor: ~16% gross, ~4–5% net (FY25) |
Everyworks Group
DW Aircon + Mr Plumber + Electrician (owned techs)
- Model
- Multi-trade, company-owned technicians
- Recurring revenue?
- Alive in 2026?
- Signal
- The closest thing to a roll-up — integrated, not a marketplace
Billy Aircon / LK Brothers
Long-running independent servicers
- Model
- Servicing + contracts + install, SEO-led
- Recurring revenue?
- Alive in 2026?
- Signal
- Contract books + 1,000+ reviews; private, no accounts
Carousell Services
Marketplace inside a bigger business
- Model
- ~2.5% take-rate (capped S$30/txn)
- Recurring revenue?
- Alive in 2026?
- Signal
- Survives because it is subsidised by the wider platform
Kaodim / Sendhelper / Mr Right
VC-funded trade marketplaces
- Model
- Lead-gen / booking commission
- Recurring revenue?
- Alive in 2026?
- Signal
- All dead — Kaodim raised ~US$17.6M, ceased 2022
ISOTeam (SGX:5WF)
Listed building-maintenance / M&E proxy
- Model
- Project + maintenance contracts
- Recurring revenue?
- Alive in 2026?
- Signal
- The audited anchor: ~16% gross, ~4–5% net (FY25)
The cautionary tale: the trade-marketplace graveyard
Every funded pure-play that tried to be “the app for home services” in Singapore is gone. Kaodim raised ~US$17.6M and ceased all operations in July 2022. Sendhelper was acquired by PropertyGuru and shut down in February 2025. Mr Right and HomeServices.sg are dead and struck off. The lesson for a founder: the value isn't in matchmaking strangers to one-off jobs — it is in owning the technician, the customer relationship, and the recurring contract.
The customer: the prize isn't the cheapest household — it's the contract
Residential is the volume, and it is where the price war is fiercest. The stickier, higher-value money sits in recurring contracts — condos via their managing agents (MCSTs), and B2B (offices, F&B chains, retail) whose systems run near-24/7 in humidity and who sign annual maintenance deals with priority response. A new entrant who only chases one-off HDB jobs is fighting hardest for the worst economics.
Where the demand sits — and where the contracts are
A rough split of the addressable aircon-servicing demand by segment (indicative). Volume is residential HDB; lifetime value and stickiness climb sharply toward B2B and managed-property contracts.
- HDB households (volume, price-war)55%Modal home; 1–3 split units; most price-sensitive — ~77% of dwellings are HDB
- Condo / private (higher WTP + MCST gateway)22%More units; the managing-agent layer opens bulk common-area contracts
- B2B / commercial (the sticky prize)16%Offices, F&B, retail; near-24/7 use; annual contracts, priority response, highest LTV
- Landed / luxury (multi-split, fewer rivals)7%VRV/VRF; higher ticket, harder to commoditise
Source: SGAI segmentation on SingStat dwelling mix (2023) + sector structure. Shares are indicative of where value concentrates, not a measured revenue split.
Why they buy (and why it's durable)
Two demand modes. Breakdown urgency — “aircon not cold” is an emergency with near-zero price sensitivity at the point of failure. And routine prevention — quarterly servicing to stop the drainage clogs that cause water leaks and mould. The awareness gap (most don't book until it breaks) is precisely what a contract-with-reminders converts into recurring revenue.
What they search at 1am
“aircon servicing price”, “aircon not cold”, “aircon water leaking”, “chemical wash vs general servicing”, “gas top up price”, “best aircon servicing”, “aircon servicing contract”. Every one is dominated by vendor cost-guides — proof of high commercial intent, and an opening for a brand that answers them honestly instead of upselling.
The economics: cheap to start, hard to keep
The startup cost is genuinely low — the van and tools are the wall, not the trade itself. What is hard is the running P&L. Labour is the dominant cost, and the official wage is higher than the job boards suggest: MOM's median for an aircon mechanic is S$3,263/moSourceMOM Resident Occupational Wages (gross median), June 2024 gross, well above the ~S$2,000 some job sites advertise. On top of that sit the manpower constraints, paid-search acquisition costs, and the price war on the one service customers compare.
All-in cost to start an aircon-servicing business
Company registration is ~S$315–2,500; the rest is a basic tool kit (vacuum pump, gauges, recovery machine, flaring tools — low thousands) and a service van. There is no large fit-out or inventory wall — which is exactly why the market is so crowded. No single published 'total startup cost' figure exists; this is a modelled spread.
Source: SGAI estimate on ACRA fees + trade tool/van costs, 2026 (indicative)
Where the aircon-servicing dollar goes
A small servicing firm modelled per S$100 of revenue, with a healthy mix of contracts and upsell work. Technicians are the dominant cost; customer acquisition is the quiet leak that one-off-job operators feel hardest.
- Technician wages + CPF−42%58% left
MOM median aircon mechanic ~S$3,263/mo gross (2024); the swing cost · MOM 2024
- Parts, refrigerant & consumables−16%42% left
Low on a gas top-up, higher on overhaul/install parts · SG wholesale refrigerant pricing
- Vehicle, fuel & tools−11%31% left
The mobile-trade equivalent of rent
- Marketing / customer acquisition−12%19% left
Aircon search clicks ~S$4–9 each; heaviest for one-off jobs, light on contract renewals · SG PPC benchmarks
- Admin, insurance, software, licences−9%10% left
What the owner actually keeps
Verdict: A healthy 10% — there is real margin of safety here.
Illustrative model on SG benchmarks (2024–2026); no audited SG aircon-servicing net margin is published, and the closest listed proxy (ISOTeam) runs ~4–5% net. A book of recurring contracts is what shifts the marketing line down and the net up. Not financial advice.
Aircon servicing & home maintenance
Would a value investor own the average operator here?
A value investor would not want the AVERAGE one-off aircon servicer — it's a no-licence price-taker in a crowded market. But the demand is durable and near-essential, and a disciplined operator who builds a recurring contract book has a real, if modest, edge the one-off trades can't copy.
General servicing is a comparison-shopped commodity; the only pricing lever is a contract relationship and the upsell ladder, not the headline per-unit rate.
Buffett, FCIC 2010 — pricing power is “the single most important decision”
No licence or brand moat — but a quarterly contract (and a B2B/MCST maintenance deal) is a genuine switching cost a one-off renovation or handyman job can never have.
Buffett 2007 — an enduring moat protects returns on capital; the recurring base is the closest thing here
Asset-light, so returns on the little capital tied up can be decent — but only if the contract book and upsell ladder carry it past the price war.
Buffett 1979 — a high earnings rate on capital, unleveraged
A van and tools, not a fit-out. The binding constraint is technicians and trust, not capital.
Buffett 2007 — the worst business needs much capital; this one does not
Tropical, always-on, near-essential demand that recurs — the strongest pillar, and the reason this scores above the one-off home trades.
Graham, Security Analysis Ch.2 — inherent stability is qualitative
A services-classified servicer faces the 35% Dependency Ratio Ceiling and can’t hire Work Permit technicians from South Asia — the tightest labour configuration in the trade.
MOM foreign-worker rules — Services sector DRC 35% + restricted source countries
A recurring contract book (residential quarterly + B2B/MCST), an honest upsell ladder that customers trust, and tight technician scheduling — not the lowest per-unit quote.
Assessment uses the value-investing lens on SG aircon-servicing economics (2024–2026). A lens on economic quality, not a verdict on an owner-operated livelihood — this trade can work very well as a disciplined owner-run business.
Model your own contract book — drag the sliders:
Build a contract book, not a job queue
One-off servicing is a price war. A book of quarterly contracts (4 visits/year) is recurring revenue you don't re-win every time — and the four visits are paid access to sell the upsell ladder. Drag the sliders: watch the recurring share and whether the contract book outruns the technician-and-marketing treadmill.
The contract book
Contract rate; ad-hoc volume jobs bottom out near S$16–30/unit.
The upsell ladder (where money is made)
Chemical wash / gas top-up / overhaul, sold during a contract visit.
Gas top-up ~S$50–150; chemical wash ~S$80–150; overhaul ~S$120–450.
The treadmill
MOM median for an aircon mechanic ≈ S$3,263 gross (2024).
Search clicks for aircon keywords run ~S$4–9 each — the cost of one-off acquisition.
Recurring share of revenue
82%
from the contract book (the moat) vs per-visit upsells
Annual revenue
S$52,704
S$4,392/mo
Annual net
−S$138,096
-262% margin
Visits / technician / working day
0.5
comfortable · to deliver the book at ~26 days/mo
Verdict: Underwater — the technician + marketing treadmill outruns the book. This is the one-off-jobs trap.
Illustrative model on SG benchmarks (2024–2026); no audited SG aircon-servicing margin is published, so this is a starting frame, not financial advice. Recurring gauge is contract revenue ÷ total revenue; capacity gauge: ≤6 visits/tech/day comfortable, 7–9 stretched, >9 over capacity.
How to actually start one (and what the internet gets wrong)
Search results will tell you that you need a “BCA Builder's Licence” and ITE-certified technicians to do residential servicing. That is wrong, and getting it right is a credibility marker. There is no specific aircon-servicing licence. The real regulation attaches to activities, not to the trade.
The stack that actually applies
- ACRA — register the company (from ~S$315; min S$1 capital).
- NEA refrigerant rules — controlled refrigerants (HCFCs like R22; HFCs like R32, R410A) are Hazardous Substances under the EPMA. A Hazardous Substances licence/permit is only needed to import or trade them — most servicers buy from a licensed distributor and avoid this. But everyone handling gas must follow proper handling and spent-refrigerant collection rules (in force since 1 Oct 2022). The real regulated layer.
- Work-at-Heights certificate — servicing high-rise condenser units is work at height under MOM's WSH rules; technicians need a WAH competency cert. The real safety gate.
- BCA Contractors Registration (ME01) — the air-conditioning/refrigeration/ventilation workhead. Needed only to tender for public-sector work or to hire construction Work Permit holders (mandatory for that since 1 Jun 2025). Not required for purely private residential servicing.
- bizSAFE (WSH Council) — a credibility/tender ladder (Level 1 → Star); Level 3 is the rung most often demanded for B2B contracts. Not a licence.
- GST — compulsory once turnover passes S$1M; MELS/MEPS supplier registration with NEA if you also sell/install units.
The real bottleneck
It isn't paperwork — it is technicians. A servicing business classified under the Services sector faces a 35% Dependency Ratio Ceiling and cannot hire Work Permit technicians from South Asia — only Malaysia, China and a few North Asian sources. The aircon trade also has no CaseTrust accreditation scheme (renovation has two), so there is no kitemark to inherit — trust has to be built, which is also the opening (see below).
Where a new aircon business actually wins
The lowest per-unit quote is a race to the bottom. Every winning move below is really one idea: convert commodity one-off jobs into recurring, trusted relationships.
The contract book (the moat)
Quarterly servicing plans turn one-off price-shoppers into recurring revenue you don’t re-win each time — the closest thing to a moat in any home trade. Make the recurring base the business; treat one-off jobs as contract leads.
B2B & MCST contracts
Offices, F&B chains and condos (via managing agents) run systems near-24/7 in humidity and sign annual deals with priority response — far stickier, higher LTV, and far less price-shopped than HDB jobs.
Trust, productised
No CaseTrust scheme exists here and the trade has a scam reputation (the gas-top-up overcharge). Fixed transparent pricing, "top-up only on a proven leak", written diagnosis and before/after photos directly attack that — a kitemark gap the industry doesn’t fill.
Niches with fewer rivals
Landed/luxury multi-split, VRV/VRF systems, and commercial-kitchen (F&B) work are higher-ticket and far harder to commoditise than HDB split-unit servicing.
The energy-efficiency angle
Aircon is among the largest home electricity loads and the single biggest in many commercial buildings; honest, well-serviced units cut bills and carbon — a credible pitch to facility managers under tightening NEA efficiency rules.
The honest software edge
Where AI/software genuinely earns its place: online booking, automated next-quarterly-service reminders (which directly attack the awareness gap), technician route and schedule optimisation, and renewal management. Retention and efficiency — not a moonshot. It sharpens a good contract book; it won’t rescue a price-war model.
Questions founders ask
How much does aircon servicing cost in Singapore?
Indicatively (2024–2026): general servicing roughly S$25–50 per unit, chemical wash S$80–150, chemical overhaul S$120–450, and a gas top-up S$50–150 depending on the refrigerant (R32 is cheaper; the older R22 costs far more). Multi-unit and annual-contract rates fall lower — advertised "from S$16–30/unit" figures are volume-discount floors, not a realistic single-job price. These are aircon-company price-list figures, not an official tariff.
How often should I service my aircon in Singapore?
The industry standard cited in Singapore is about every three months (quarterly), because the humidity clogs drainage pipes and breeds mould fast — that is what causes water leaks and weak cooling. But be clear: quarterly servicing is a sensible climate-driven recommendation that also sells contracts, not a legal or universal manufacturer mandate. Light-use homes can often stretch the interval.
Do I really need a gas top-up every service?
No. Refrigerant runs in a sealed loop — it does not get "used up." If the gas is low, you have a leak, and topping up without fixing the leak just lets it escape again. A routine "top-up every visit" is the single most-cited overcharge in the trade. A trustworthy servicer top-ups only after locating and fixing a leak.
What licence do I need to start an aircon-servicing business in Singapore?
There is no specific "aircon-servicing licence" — you register a company with ACRA like any business. The real regulated layers attach to specific activities: a NEA Hazardous Substances licence/permit if you import or trade controlled refrigerant (most servicers simply buy from a licensed distributor and avoid this); BCA Contractors Registration (workhead ME01) only for public-sector work or to hire construction Work Permit holders; a Work-at-Heights certificate for staff servicing high-rise condenser units; and GST registration once turnover passes S$1M. The internet claim that you must hold a "BCA Builder’s Licence" to do residential servicing is wrong.
Is an aircon-servicing business profitable in Singapore?
One-off jobs are a brutal price war where general servicing is close to a loss-leader. The money is in two places: the upsell ladder (chemical wash, gas top-up on a genuine leak, overhaul, replacement) and — the real edge — a book of recurring quarterly servicing contracts you do not have to re-win each time. The closest audited proxy, listed maintenance contractor ISOTeam, runs about 16% gross and ~4–5% net. Demand is durable and near-essential, but margins reward operators who build recurring contracts and sell honestly, not the lowest bidder.
Want a report like this for your company or sector?
Tell us what you're trying to figure out. We'll build you a Deep-Context report — the same depth, focused on your business.
About this report. Built with SGAI's Deep-Context Engine — human-directed, AI-accelerated. Figures draw on SingStat (Household Expenditure Survey 2023), MOM (occupational wages + foreign-worker rules), NEA (refrigerant / Hazardous Substances controls, Kigali Amendment, MELS/MEPS), BCA (Contractors Registration), SGX filings (ISOTeam), and aircon-company price lists (2024–2026). There is no credible published “Singapore aircon servicing market size” — the aggregator figures measure equipment and contradict each other by ~45×, so we size it bottom-up and say so. Prices are indicative company-list figures, not an official tariff; no audited SG aircon-servicing margin is published. The MOAT Score is a transparent SGAI judgement on economic quality, not a verdict on an owner-operated livelihood. Verify fees and regulatory steps with each agency before acting.
We're not a grant office — we build AI and intelligence for Singapore SMEs. Talk to us.