iCents Group Holdings Berhad (0366.KL) — Deep Dive Research Report

Prepared 9 June 2026. Sector: Industrials (Industrial Engineering / Construction). Listing: ACE Market, Bursa Malaysia. Reporting currency: Malaysian Ringgit (RM). Financial year ends 30 June.

Two structural caveats stated up front. iCents listed on 17 July 2025, so (1) it has no earnings conference calls — Malaysian ACE-market small-caps do not hold them — and only three quarterly report filings exist as a listed entity (Q1, Q2 and Q3 FY2026). Where the brief asks for "five concall transcripts," I use the available listed-entity quarterly filings plus the IPO prospectus and management's on-record statements, and I flag the shortfall explicitly in Sections 7 and 9. (2) No full financial year has closed since listing, which constrains Section 10. Neither gap is hidden; both are worked around with the primary disclosures that do exist.

1. What the company does

iCents builds the rooms that other factories cannot afford to get wrong. When a semiconductor packager, a vaccine plant, a pharmaceutical line or a hyperscale data centre needs a space where the air is filtered to a controlled particle count, the temperature and humidity are held inside a tight band, and the walls, ceilings and floors do not shed dust, iCents designs that controlled environment, builds it, installs the mechanical and air-handling systems, tests it against the required cleanliness class, and hands it over working. In industry shorthand this is cleanroom EPCC — engineering, procurement, construction, and testing & commissioning.

The company sits at the intersection of two things that are usually separate: it is both a contractor (it wins and executes projects on a customer's site) and a manufacturer (through its factory it makes the physical building blocks of a cleanroom — wall panels, ceiling systems, doors, windows). Most cleanroom contractors buy their panels from third parties; iСents makes a meaningful share of its own. That vertical piece is small in revenue terms (management has said manufacturing of cleanroom fixtures is roughly 10-15% of group revenue, per the IPO coverage) but strategically it lets the group control quality, lead time and margin on the components that go into its own projects, and increasingly to sell those components as standalone product exports.

The group is organised as an investment holding company over three operating subsidiaries:

• VC Engineering Sdn Bhd — the contracting and project-delivery arm. It holds a CIDB Grade G7 licence, the highest Malaysian construction-contractor grade, which carries no upper limit on tender value. This is the entity that signs the large project sub-contracts.
• Maytech Cleanroom Manufacturing Sdn Bhd — the manufacturing arm, making sandwich wall panels (aluminium honeycomb, rockwool and EPU cores), ceiling systems, doors and windows, and acting as authorised distributor for Dawn Modular raised-access floor systems. Its new factory in Mantin, Negeri Sembilan was launched in late 2025.
• iCents Engineering Pte Ltd — the Singapore-incorporated vehicle for regional sales, marketing and engineering support.

The managing director is Ir. Ts. Vincent Ong Mum Fei, an engineer by training who, together with executive directors Foo Siang Leng and Tan Wei Ying (Foo and Tan are husband and wife) and shareholder Lim Teng Hong, controls the company. The group's reputation rests on a decade-plus track record of delivering cleanroom projects for multinational customers; the public holding company itself was assembled for the July 2025 IPO, which raised RM27 million at RM0.24 per share (The Star).

"We are ready to embark into a new phase of our growth journey and strengthen our market presence." — Ir. Ts. Vincent Ong Mum Fei, MD, on listing (The Exchange Asia)

The plain-language version: iCents is a Malaysian cleanroom and controlled-environment contractor that also makes its own panels, is using IPO money to add factory capacity and push into Indonesia, Singapore and Thailand, and is increasingly chasing the fastest-growing slice of its market — cooling and clean-build work for AI data centres.

A concrete walk-through. A multinational construction firm wins the main contract to build a data centre in Indonesia. It needs the specialised internal systems built to a high specification. It awards iCents' subsidiary Maytech a roughly RM14.1 million package (IDR 59.5 billion) to design, supply and deliver the data-centre systems, for completion by June 2026 (MAXIT). iCents engineers the components to the client's spec, manufactures the panels and systems at Mantin, ships them across the border, and installs/commissions them. The customer never sees the factory; it sees a finished, tested, spec-compliant facility delivered to a deadline. That is the business, repeated across semiconductor, pharma and data-centre sites.

2. Business segments

iCents reports along two operating lines. The cut is not by industry served (semiconductor vs pharma) but by what kind of work the group is doing — building a controlled environment, versus the broader engineering/facility scope that surrounds it.

Segment A — Cleanroom services

What it does. This is the core franchise: end-to-end EPCC of cleanrooms. iCents designs the controlled space, procures and manufactures the panels and systems, constructs it, then tests and commissions it to the required ISO cleanliness class. It serves the most demanding end-markets — semiconductor and electronics, pharmaceutical and life-science manufacturing, biotech, laboratories, hospitals and food-and-beverage. Embedded inside this segment is the Maytech manufacturing activity (panels, ceilings, doors, raised floors), which management has sized at roughly 10-15% of total group revenue.

The core capability. A cleanroom is unforgiving: get the airflow, pressure cascade, panel sealing or particle count wrong and the customer's product yield collapses. The hard part is not pouring concrete; it is the integration of architecture, mechanical air handling and contamination control to a measurable, certifiable standard, under a fixed-price contract and a tight schedule. The decade-plus reference list of completed multinational projects is the asset — it is what lets iCents pass a prospective customer's qualification screen. The in-house panel factory is the second capability: controlling the supply of the building blocks lets the group protect lead times and margin that a pure assembler cannot.

Why it is a distinct segment. It carries the higher technical content and the certification burden, and it is where the manufacturing vertical lives. It is the part of the business that a generic main contractor cannot do.

Competitive position. Within Malaysia, iCents competes against private cleanroom specialists (Cleanroom Industries, EEPS Engineering, Conwall) and, at the larger and more technical end, against listed Kelington Group. iCents wins on price-competitive, integrated panel-plus-build delivery for mid-sized projects; it loses to Kelington on the very large, ultra-high-purity-intensive semiconductor fab scopes that require gas/chemical delivery engineering iCents does not provide. Management itself puts the group's Malaysian cleanroom market share at under 5% (cleanroomtechnology.com).

Role in the group. This is the franchise and the brand. It anchored 57.4% of the unbilled order book at the June 2025 IPO (RM53.53m of RM93.21m).

Segment B — Other facility services

What it does. The broader engineering scope that sits around a clean build: hook-up of machinery and equipment, supply and installation of heavy-duty ceiling systems, general construction works, mechanical and electrical / air-conditioning and mechanical ventilation (ACMV) systems, and maintenance. This is the VC Engineering G7-contractor scope. It is where the large ACMV data-centre sub-contracts land — for example, the RM34.54 million ACMV package won in May 2026 (The Star).

The core capability. Project management at scale and the G7 licence. The barrier here is lower than in cleanrooms (more contractors can do MEP/ACMV), so the edge is execution reliability, the ability to bid unlimited-value tenders, and the cross-sell from being already on site doing the clean build.

Why it is distinct. Different (lower) technical content, different margin profile, broader competitor pool, and revenue that is lumpier and project-driven. It exists partly to capture the wider wallet on a project where iCents is already engaged on the cleanroom, and partly to ride the data-centre ACMV wave independently.

Competitive position. Crowded — many Malaysian MEP/ACMV contractors compete here. iCents differentiates through the bundle (clean build + facility systems from one party) and its growing data-centre reference base.

Role in the group. The volume and growth engine of late. In Q1 FY2026 it was actually the larger of the two lines (facility services contributed roughly 58% of revenue versus 42% cleanroom, per the Q1 results commentary), driven by data-centre work. It held 42.6% of the IPO-date order book.

3. Products and business detail

The product and service catalogue.

• Cleanroom EPCC — full design-build-test of controlled environments to ISO cleanliness classes, the flagship offering.
• Sandwich wall panels — manufactured by Maytech with aluminium-honeycomb, rockwool and EPU (rigid polyurethane) cores. These are the structural skin of a cleanroom: flat, sealed, non-shedding, fire-rated (rockwool core) where required. The choice of core trades off rigidity, fire performance and cost.
• Cleanroom doors, windows and heavy-duty ceiling systems — the complementary fixtures, made in-house.
• Raised-access flooring — distributed as authorised agent for Dawn Modular Floor systems (resale rather than manufacture).
• ACMV / mechanical & electrical systems — supply, installation, testing, commissioning and maintenance of air-conditioning and mechanical ventilation, the core of the data-centre sub-contracts.
• Prefabricated modular cooling / thermal management for data centres — the newest line, being developed via the June 2026 partnership with China's WXYD (see Sections 5 and 7).
• Equipment hook-up and facility/maintenance services — the recurring, lower-intensity tail.

Process knowledge and certifications that matter. The cleanroom work requires the ability to design to and certify against ISO 14644 cleanliness classes (and GMP standards for pharma). The contracting arm's CIDB G7 registration is the gating credential for large Malaysian public/industrial tenders — it removes the contract-value ceiling that lower grades face. The hard, slow-to-build asset is the qualified reference list: multinational customers will not hand a cleanroom to a contractor that cannot point to comparable completed projects, so the track record is itself a barrier.

Manufacturing and the Mantin expansion. Maytech manufactures at a facility in Mantin, Negeri Sembilan, formally launched in late 2025 (a key milestone flagged in the Q2 FY2026 results), with the group citing 42,754+ sq ft of facility space. The Mantin plant was explicitly built to strengthen cleanroom and data-centre-related product manufacturing capacity — a deliberate tilt of the factory toward the data-centre demand wave. New machinery funded by the IPO (17.3% of proceeds, ~RM4.675m) feeds this line.

Geographies and export markets. The home base is Malaysia (offices in Subang Jaya, Nilai and Johor; the Mantin factory; and a stated regional project-delivery focus in Kuching, Sarawak). The international push, financed by the IPO, runs through:

• Singapore — sales, marketing and client-support hub (iCents Engineering Pte Ltd).
• Indonesia (Jakarta) — sales and technical-service office; already producing revenue via the RM14.1m data-centre system contract (March 2026).
• Thailand — cleanroom product exports and engineering/installation, an early regional beachhead.

Milestones that changed the business.

• 17 July 2025 — ACE Market IPO; RM27m raised, public tranche oversubscribed 2.3x.
• Late 2025 — Mantin manufacturing facility launched.
• March 2026 — first cross-border data-centre contract (Indonesia, RM14.1m).
• May 2026 — RM34.54m domestic data-centre ACMV sub-contract (to Feb 2027).
• June 2026 — VC Engineering / WXYD partnership on prefabricated modular data-centre cooling for Southeast Asia.

4. Customers

Who buys. Two buyer types. First, end-users in regulated, contamination-sensitive industries — semiconductor and electronics manufacturers, pharmaceutical and life-science producers, biotech firms, F&B and palm-oil processors, hospitals and laboratories — who need a cleanroom for their own production. Second, and increasingly the source of the largest tickets, main contractors and construction firms building data centres, who sub-contract the specialised internal systems (clean build, ACMV, cooling) to iCents. The Indonesia win, for example, came from "a multinational construction company" building a data centre, not the data-centre operator directly; the May 2026 RM34.54m ACMV package came from "a local construction company."

Who decides, and on what. For end-user cleanrooms, the buyer is typically the customer's facilities/engineering function and project management office, choosing on demonstrated cleanliness-class compliance, on-time delivery against a fixed schedule, price, and reference projects. Sales cycles are project-length (months from tender to award, then months to delivery). For data-centre sub-contracts, the main contractor's procurement and project teams decide, weighting reliability, the G7 credential, capacity to meet aggressive build timelines, and price.

Why they choose iCents. The integrated proposition — design, in-house panel manufacture, build, and test under one roof — compresses lead time and single-points accountability, which matters when a delay cascades into the customer's production ramp. The G7 licence lets it bid the big jobs. The multinational reference list passes qualification screens. Management's own framing is that demand from electrical & electronics and pharmaceutical industries is "expected to drive demand for cleanroom facilities" (The Star, July 2025).

Switching costs. Moderate and project-bound rather than installed-base lock-in. A cleanroom is a one-off capital build; the customer is not "locked in" the way a software customer is. But within a build, switching mid-project is costly (re-qualification, schedule loss), and a contractor with a clean delivery record on a customer's first facility is strongly favoured for the next phase or sister site. The lock-in is reputational and repeat-business-driven, not contractual.

Concentration. Disclosure is thin (no named individual customer concentration figures are public post-IPO), but the order book is project-lumpy by nature: a handful of large contracts (the RM34.54m and RM14.1m wins) can dominate a given period's revenue. This is normal for a contractor and is a revenue-volatility risk more than a single-customer-dependency risk, on current disclosure.

Contract structure. Predominantly project-based, fixed-scope sub-contracts and EPCC awards with defined completion dates (e.g. the May-2026 ACMV job runs to 15 February 2027). Revenue is recognised over the build and is therefore lumpy and order-book-led, with a maintenance tail. The unbilled order book — RM93.21m at IPO (RM53.53m cleanroom / RM39.68m facility services) — is the forward-revenue visibility metric to watch, replenished by new wins.

5. Competitive landscape

The Malaysian cleanroom and controlled-environment market is fragmented and competitive, with no single dominant player at iCents' size and a long tail of private specialists. iCents itself states a sub-5% domestic cleanroom market share. The structure has three tiers relevant to iCents:

1. The large listed integrator — Kelington Group Berhad (KGB). The clear listed comparable, but more an adjacent giant than a like-for-like rival. Kelington's core is ultra-high-purity (UHP) gas and chemical delivery for semiconductor fabs, plus large cleanroom and general-contracting construction. It plays at the top, most technical and largest-ticket end of the market that iCents largely does not reach, and it carries a ~RM1.4bn order book and an industrial-gas business iCents has no equivalent to. Where their scopes overlap (cleanroom construction), Kelington wins the largest, most UHP-intensive fab work; iCents competes below it on mid-sized, panel-integrated builds.

2. Private cleanroom specialists. Cleanroom Industries Sdn Bhd (est. 1999, capable to Class 1), EEPS Engineering (semiconductor-focused custom cleanrooms), Conwall, Vicfil and others. These are iCents' true head-to-head competitors on mid-market cleanroom EPCC. iCents' edge over most is the in-house panel manufacturing plus the G7 contracting licence under one roof; its exposure is that these are nimble, established, relationship-driven private firms.

3. The crowded ACMV/MEP and data-centre sub-contractor pool in the "other facility services" segment, where many local engineering contractors compete and differentiation is thinner.

A separate, upstream competitive axis is panel supply: global sandwich-panel makers (e.g. Ireland-listed Kingspan and others) and regional panel manufacturers compete with Maytech on the components — but iCents' panel output is largely consumed internally, so this is more an input-cost/quality question than a market-share battle.

Barriers to entry. Real but not impregnable. The qualifying reference list, ISO/GMP-class delivery know-how, the G7 licence, and (for iCents specifically) the capital tied up in a panel factory all raise the bar above a generic contractor. But they do not stop a well-capitalised, well-connected new entrant, and they do not protect pricing in the commoditised ACMV tier. There is no patent moat and no recurring-revenue lock-in. The honest read: iCents has a credible mid-market position and a useful vertical-integration angle, not a wide moat.

Structural shifts. The dominant shift is the AI data-centre buildout across Southeast Asia, which is enlarging the addressable market for both clean builds and cooling/ACMV and pulling contractors like iCents toward data-centre work (its three largest 2026 wins are all data-centre-related). This is a tailwind that also invites more competition into the data-centre cooling niche.

Market caps are peer-size references only (currencies and as-of dates as shown); they move and are not applied to iCents. Sources: Kelington i3investor, Pentamaster stockanalysis.

6. Industry

Demand drivers. iCents' fortunes track capital spending in industries that require contamination control: semiconductor and electronics manufacturing, pharmaceutical/biotech/life sciences, and — the newest and fastest — hyperscale and AI data centres. The underlying engines are Malaysia's growing E&E sector, rising regional pharmaceutical manufacturing, and a Southeast Asia data-centre construction wave fed by cloud and AI demand. Demand is capex-cycle driven: when chipmakers and data-centre operators commit to new fabs and facilities, cleanroom and clean-build contractors get work; when capex pauses, order books thin.

Size and growth. The Asia-Pacific cleanroom technology market generated about USD 2.15bn in 2024 and is projected to reach ~USD 3.20bn by 2030, a ~6.8% CAGR (Grand View Research). The semiconductor/electronics sub-segment is the fastest-growing slice (projected ~8.5% CAGR through 2032 in one estimate), pulled by China's 300mm fab pipeline, Taiwan's advanced-node program and South Korea's capacity expansions. Southeast Asia specifically benefits from electronics-assembly migration and the data-centre buildout — the part of the map where iCents operates.

Position in the supply chain. iCents is a downstream integrator/contractor: it does not make chips or run data centres; it builds the controlled spaces and installs the systems that those operations need. It sits between upstream panel/equipment suppliers (some of which it has internalised via Maytech) and the end-user/main-contractor customer.

Import substitution. A relevant angle for the manufacturing arm: Maytech making sandwich panels and data-centre systems domestically lets Malaysian and regional projects source locally rather than import, and lets iCents export panels into Thailand and Indonesia — i.e. iCents is on the substituting/exporting side of the panel trade, not the import-dependent side.

Regulation. The market is shaped by cleanliness-class standards (ISO 14644), GMP requirements for pharma, and, in Malaysia, the CIDB contractor-grade regime. These are qualification gates rather than restrictive licences, but they meaningfully narrow the field of who can bid serious work.

Cyclicality. Project-based and capex-linked, so cyclical. A semiconductor downturn or a data-centre capex pause would hit order intake. The current phase is a tailwind: AI-driven data-centre construction and resilient regional E&E/pharma capex.

Net industry tailwinds/headwinds. Tailwinds: data-centre buildout, regional E&E and pharma capex, electronics supply-chain diversification into Southeast Asia. Headwinds: capex cyclicality, competitive fragmentation compressing margins (especially in ACMV), and input-cost (steel/aluminium) exposure on panels.

7. Growth triggers

Source note: iCents holds no earnings concalls. The triggers below are drawn from its three listed quarterly report filings (Q1-Q3 FY2026), its IPO prospectus/listing commentary, and dated material-contract and management announcements. Each is cited to its dated source.

• Mantin manufacturing facility ramping data-centre-related product capacity. The new Maytech plant was launched in late 2025 and explicitly built to strengthen cleanroom and data-centre product manufacturing. (Q2 FY2026 results announcement, Feb 2026 / I3investor)

• Regional expansion into Singapore, Indonesia and Thailand now producing revenue. Offices established via iCents Engineering; the Indonesia office is already converting to contracts. (IPO commentary, Jun-Jul 2025, cleanroomtechnology.com; reiterated in Q3 FY2026, The Star)

  Foo Siang Leng, on the Indonesia win, said it "marks another step in expanding our presence beyond the domestic market." (MAXIT)

• RM34.54m domestic data-centre ACMV sub-contract (to Feb 2027). Supply, install, test, commission and maintain the air-conditioning and mechanical ventilation system; adds near-term order-book visibility. (13 May 2026, The Star)

• RM14.1m Indonesia data-centre system contract (completion June 2026). First cross-border data-centre revenue, via Maytech. (March 2026, MAXIT)

• WXYD (Wuxi Yiding Zhizao) partnership — prefabricated modular data-centre cooling for Southeast Asia. A new product line: VC Engineering pairs its project execution with WXYD's thermal-management technology to develop, market and deliver prefab modular cooling and intelligent control systems for AI/cloud/hyperscale data centres regionally. This is the clearest forward strategic bet. (3 June 2026, FinancialContent / AccessWire)

  "This collaboration represents a strategic step in strengthening our capabilities within the data centre ecosystem." — Vincent Ong Mum Fei, MD.

• IPO proceeds funding two-to-three years of capex, expansion and product development. Management has earmarked the unspent IPO funds for machinery, expansion, product development and marketing over the next 2-3 years. (Q1 FY2026 results, Nov 2025, moomoo)

• ~10% revenue-growth ambition for 2026. Management publicly framed a ~10% growth objective at listing. (BERNAMA, Jul 2025, id=2446161)

8. Key risks

• Project/order-book lumpiness and capex cyclicality. iCents' revenue is contract-driven, so a slowdown in semiconductor or data-centre capital spending directly thins order intake, and the timing of large wins (RM34.54m, RM14.1m) swings reported periods. Mechanism: one or two delayed or cancelled awards can produce a weak quarter. Calibration: high-probability moderate drag, structural to the contracting model. The order book (RM93.21m at IPO) is the buffer to watch — if it shrinks faster than it is replenished, growth stalls.

• Thin, low-barrier competition in the facility-services tier. The ACMV/MEP segment that has driven recent volume is crowded and price-competitive, with no real moat. Mechanism: margin compression as more contractors chase data-centre ACMV work. Calibration: high-probability moderate margin risk; this is the less-defensible half of the business and currently the larger half.

• Execution risk on the regional and product-line expansion. Simultaneously ramping the Mantin factory, opening Singapore/Indonesia offices, executing cross-border contracts, and launching a brand-new prefab-cooling product line via WXYD is a lot for a recently-IPO'd small company. Mechanism: cost overruns, cross-border execution slips, or a new cooling product that does not gain traction would burn the IPO proceeds without the expected return. Management's own contract language repeatedly hedges on delivery "in accordance with agreed specifications, timelines and quality standards" — an acknowledgement that execution is the binding constraint.

• Single-project / single-counterparty execution exposure (data-centre cooling). The WXYD cooling push depends on a Chinese technology partner; the value and durability of that partnership are undisclosed, and the product is unproven in the field for iCents. Mechanism: if the partnership underdelivers or the technology disappoints early customers, the most-hyped growth leg disappoints. Calibration: low-to-moderate probability, but it is the leg the market is most likely to be pricing optimism into.

• Input-cost exposure on manufacturing. Maytech's panels consume aluminium, steel and PU; fixed-price project contracts mean cost spikes between bid and build compress margin. Calibration: moderate, manageable, but real for a vertically-integrated contractor.

• Key-person and governance concentration. The business is closely held and run by its founder-managers (Ong, Foo, Tan). Mechanism: loss of, or distraction in, the senior team would hit relationships and execution; closely-held control also limits external governance checks. Calibration: low-probability, high-impact.

• Newly-listed, short public track record. Less than a year of public reporting, distorted prior-year comparatives from the pre-IPO restructuring (some FY2025 quarterly comparatives at the listed-entity level show as zero revenue), and an untested promise-keeping record. Calibration: an information risk rather than an operating one, but it raises the cost of trusting guidance (see Section 9).

9. Walk the talk

Concall dates used: none exist. iCents holds no earnings calls. The credibility assessment below is built from the five datable management touchpoints available: the IPO prospectus/listing statements (June-July 2025), and the three listed quarterly filings — Q1 FY2026 (results released ~Nov 2025), Q2 FY2026 (released ~Feb 2026, I3investor), and Q3 FY2026 (released 23 May 2026) — plus the dated contract announcements through June 2026. The most recent of these (Q3 FY2026, 23 May 2026, and the WXYD announcement, 3 June 2026) is well within 90 days of today. With under a year of public history, this is necessarily a short-track-record assessment, and I say so plainly.

At listing (June-July 2025), management set out a concrete agenda: use the RM27m to fund working capital, new machinery, regional expansion into Singapore/Indonesia/Sarawak, and a new manufacturing facility, while publicly framing a roughly 10% revenue-growth ambition for 2026 and pointing to a RM93.21m unbilled order book as the visibility base.

Against that, the early scorecard is encouraging on delivery of the operational promises. The Mantin manufacturing facility, promised at IPO for around December 2025, was launched on schedule and flagged in the Q2 FY2026 results — a kept, datable commitment. The regional-expansion promise has moved from plan to revenue: the Indonesia office translated into the RM14.1m data-centre contract (March 2026), and Thailand exports began, so the "expand beyond the domestic market" language has been backed by signed contracts rather than left as aspiration. The IPO-proceeds plan has been transparently tracked: Q1 FY2026 disclosed RM14.34m of proceeds already applied to working capital and listing expenses, with the remainder earmarked for machinery and expansion over 2-3 years — i.e. management is reporting against its own stated use-of-proceeds rather than going quiet on it.

On the headline growth ambition, the ~10% figure looks conservative against outcome: nine-month FY2026 revenue grew low-double-digits year-on-year and Q3 FY2026 revenue rose sharply versus the prior-year quarter, helped by the data-centre wins. So management has, if anything, under-promised and over-delivered on top-line direction in its first listed year. The recurring caveat in its contract statements — that wins will "contribute positively … provided there are no unforeseen disruptions" — is appropriately hedged rather than promotional.

The honest limits on this assessment: the prior-year comparatives are distorted by the pre-IPO restructuring (some FY2025 quarterly comparatives show zero revenue at the listed-entity level), which flatters the optics of "growth," and there is simply not enough public history yet to judge whether management is durably accurate or merely benefiting from a favourable first year. The WXYD cooling partnership and the broader regional product push are promises whose outcomes are still entirely in the future.

Assessment: on the evidence available, this is management that has so far done the concrete things it said it would — built the factory on time, turned regional plans into signed contracts, and reported transparently against its IPO use-of-proceeds — while keeping its public revenue ambition modest enough to beat. That is a credible start. But it is a start: under a year listed, distorted comparatives, and the most exciting promises (data-centre cooling, regional scale-up) not yet proven. Treat the early track record as a positive signal, not yet as a settled reputation.

10. Shareholder friendliness index

Dividends. iCents listed on 17 July 2025 and has not declared any dividend since listing, nor has it publicly adopted a formal dividend policy (prospectus coverage). The conventional "last three financial years" of dividends-per-share cannot be assessed for a public investor: the three years prior to listing (FY2023-FY2025) were spent as a private group, and any pre-IPO distributions are not part of the public record. No DPS figures are publicly verifiable, so I do not estimate them. The posture is explicit and unsurprising for a growth-stage ACE-market company: capital is being retained and recycled into the Mantin factory, regional expansion and product development funded by the IPO, not returned to shareholders.

Buybacks and dilution. MoatMap records zero buybacks in the trailing ~90 days (since 11 March 2026). Searching the wider record back to listing, there is also no announced or executed share-buyback programme — a company under a year old typically has not yet sought (or used) a treasury-share mandate. On the dilution side, the share count was enlarged to 500.0 million shares at the July 2025 IPO (112.5m new shares issued plus a 30m private placement and an offer-for-sale of existing shares by the founders), and there is no evidence of further issuance or option-driven dilution since. So over the only window that exists as a public company — roughly eleven months — the count is flat at ~500m, with no buyback to retire shares and no new dilution beyond the listing event. The founders (Ong, Foo, Tan, Lim) sold part of their existing holdings into the IPO offer-for-sale, which is a normal listing mechanic rather than a post-listing distribution signal.

Verdict: Hoards capital — by design, growth-stage. iCents currently returns no capital and pays no dividend, deliberately retaining the IPO proceeds and cash flow to fund factory capacity, regional expansion and the data-centre product push; this is consistent with a recently-listed expansion story, but income-oriented or capital-return-focused holders should expect nothing on that front in the near term.

11. Insider activities

Source: per the venue rule for Malaysia (Bursa's disclosure portal is gated and not reliably web-searchable for insider dealing), the MoatMap cross-market disclosure database is the canonical source for recent insider transactions and is used as the sole source here. Data current as of 2026-06-09 12:15 UTC.

Recent transactions (last 12 months).

(Bursa Changes in Director's/Principal Officer's Interest, dates as shown, via MoatMap.)

Buys. There are no insider purchases recorded in the trailing 12 months. No director, officer or substantial shareholder has bought on the open market since listing. There is therefore no bullish cluster-buying or conviction-buy signal to report.

Sells — the why. All three transactions are sales by one person, the Financial Controller, Lee Hui Jing, over a four-day window (25-28 May 2026), totalling 35,000 shares for roughly RM16,725 combined. These are trivially small — a fraction of a percent of shares outstanding, each individually a few thousand ringgit, and the entire run amounts to less than a typical mid-level salary. No reason is disclosed in the filings, and none is reliably inferable; the size and timing are consistent with routine personal liquidity rather than any view on the business (the FC is not a board member or substantial shareholder, and the amounts are immaterial). I record this as reason not disclosed, with the strong caveat that the magnitude makes it economically meaningless as a signal. The founder-directors (Ong, Foo, Tan) — whose holdings are the ones that would carry signal — show no open-market transactions in the window beyond the IPO-stage offer-for-sale already discussed.

Net assessment. Insiders are technically net sellers over the last 12 months, but only because of three immaterial open-market disposals by the Financial Controller; there has been no buying and no substantial-shareholder or board-level dealing. The activity is concentrated in one non-board officer and is too small to read as a view on the company. Signal: neutral. There is neither a bullish insider-conviction signal (no buys) nor a genuine red flag (the sells are housekeeping-scale). Watch for any open-market activity by the founder-directors — that, if it appears, would be the meaningful signal.

12. Scenarios

Bull case. The Southeast Asian data-centre buildout keeps accelerating, and iCents proves it can ride it as more than a low-margin ACMV sub-contractor. The Mantin factory runs at high utilisation, churning out data-centre panels and systems for both domestic projects and exports into Indonesia and Thailand. The WXYD partnership turns into a real, differentiated product: prefabricated modular cooling systems that iCents can sell across the region at better margins than commodity ACMV work, winning repeat orders from main contractors who value a one-stop clean-build-plus-cooling partner. The Singapore and Jakarta offices convert into a steady regional pipeline rather than one-off wins, the order book climbs well above its IPO-date level, and the founder-managers keep delivering projects on spec and on time, building the reference list that lets iCents climb into larger tickets. Two to three years out, iCents is a credible mid-market regional cleanroom-and-data-centre specialist with a growing manufacturing and product business, no longer dependent on lumpy domestic contracting.

Base case. iCents delivers roughly what it has guided. Domestic and regional cleanroom and data-centre contracting grows at a low-double-digit pace, in line with the ~10% ambition and the industry's mid-single-digit-plus market growth. The Mantin factory ramps steadily and the Indonesia/Thailand exports add incremental revenue, but the WXYD cooling line takes time to prove and contributes modestly at first. The order book is replenished by a normal flow of mid-sized wins, with the occasional larger data-centre sub-contract creating lumpy strong quarters. Margins stay under pressure in the crowded facility-services tier but are supported by the in-house panel vertical. Management continues to report transparently against its use-of-proceeds, pays little or no dividend while reinvesting, and gradually lengthens its public track record. A solid, unspectacular execution story tracking its end-markets.

Bear case. The data-centre and semiconductor capex cycle cools, and iCents' lumpy, project-driven order book thins faster than it is replenished. The facility-services segment — now the larger half of revenue and the least defensible — gets squeezed as more contractors pile into data-centre ACMV, compressing already-thin margins. The simultaneous strain of ramping a new factory, opening overseas offices, executing cross-border contracts and launching an unproven cooling product overwhelms a small management team: a project overruns, a cross-border job slips, or the WXYD cooling line fails to gain traction, and IPO proceeds are spent without the promised return. Input-cost spikes on aluminium and steel erode fixed-price contract margins. With no dividend, a short track record, distorted comparatives and a closely-held board, sentiment toward the stock deflates as the early growth optics normalise and the company is revealed as a competitive, cyclical small-cap contractor rather than a structural data-centre winner.

Sources: iCents Group corporate site; KLSE Screener 0366; The Star — IPO proceeds and cleanroom-demand growth and RM34mil DC contract; The Exchange Asia — IPO and Indonesia DC contract; cleanroomtechnology.com; BERNAMA — 10% growth; moomoo — Q1 FY2026; I3investor — Q2 FY2026; MAXIT — Indonesia contract; FinancialContent — WXYD partnership; Kelington Group i3investor; Pentamaster market cap; Grand View Research — APAC cleanroom market; MoatMap cross-market disclosure database (insider data).