Carousell

1.1 Identity and operating model

Carousell began in Singapore in 2012 as a mobile-first peer-to-peer classifieds product built by Quek Siu Rui, Marcus Tan, and Lucas Ngoo. The original workflow reduced listing friction by allowing a seller to photograph an item, publish it, and communicate with a buyer in-app; the early product left payment and fulfilment largely to the parties. Current company material describes a broader multi-category classifieds and recommerce group spanning Singapore, Hong Kong, Indonesia, Malaysia, the Philippines, and Taiwan, while also naming brands and verticals in Vietnam, Malaysia, Indonesia, and Singapore. The product surface now includes seller tools, paid business accounts, integrated delivery in selected markets, direct-sale programmes, certified listings, and escrow-protected Buyer Protection in selected markets. This is a meaningful change in economic model: the company is no longer only a listing-and-chat utility, but public evidence still does not disclose a consolidated mix of marketplace fees, seller subscriptions, managed recommerce, or financial services.[CO001, CO002, CO003, CO004]

1.2 Founders, funding, and leadership

The founding team has remained central to the public narrative: TechCrunch identified Quek as CEO and described all three founders as National University of Singapore graduates, while the Singapore Management University case identifies Quek, Tan, and Ngoo as the three friends who started the company. The disclosed financing path includes an US$800,000 seed round in 2013, US$6 million Series A in 2014, US$35 million Series B in 2016, US$85 million Series C in 2018, US$80 million in 2020, and US$100 million from STIC Investments in 2021. Independent coverage reported the 2021 round at a US$1.1 billion valuation; a separate 2020 case study reported US$550 million, showing that valuation is event-specific rather than a current mark. Public sources identify major backers including Rakuten Ventures, Sequoia/Peak XV, Golden Gate Ventures, 500 Global, Naver, OLX Group, Telenor Group, EDBI, Mirae Asset, DBS, and STIC. The evidence does not establish the current board, voting control, liquidation preferences, debt, secondary transactions, or whether the founding trio retains operational and economic control.[CO006, CO008, CO009, CO011, CO012]

1.3 Scale, milestones, and diligence gaps

Carousell has disclosed scale through several different lenses rather than one consistent current KPI set. In 2016, the company claimed 35 million listings, 70 new listings per minute, 17 minutes of average in-app time, five countries, and more than 90 staff; the 2020 academic case described seven markets and a US$550 million valuation; current company material says the group has a leading presence in seven markets and serves tens of millions of users, while the Vietnam Chợ Tốt brand reports more than 55 million monthly visitors and REFASH reports more than 7 million clothing pieces processed. These figures are not interchangeable: listings, visitors, users, app engagement, and processed garments have different denominators and periods. The current public app listing confirms a live consumer product with 4.7/5 from 3.1K ratings, but app-store ratings are not active-user or transaction measures. An IPO tracker updated April 2, 2026 reports no IPO announced, while the company’s public evidence does not disclose a current clearing valuation. The investment case therefore has a credible operating history and regional footprint, but requires refreshed evidence on headcount, revenue mix, GMV, cash, debt, governance, cohort economics, and trust-and-safety losses before a high-confidence current valuation conclusion.[CO014, CO015, CO016, CO017, CO018]

1.4 Adverse evidence and control considerations

The principal adverse evidence is structural rather than a disclosed enforcement event. The 2020 academic case describes online fraud as a continuing challenge and asks how Carousell could provide safety and assurance while monetising without undermining community building. Sendbird’s customer case says that leaving conversations outside the app fragmented the experience and reduced the company’s visibility into transaction failures and unacceptable behaviour; Carousell subsequently adopted in-app chat and moderation controls. The current app listing shows user-generated content, messaging, tracking and linked financial or contact data, alongside regional limitations on delivery, Buyer Protection, AI listing, and direct-sale features. These facts support a diligence focus on fraud loss, dispute reserves, privacy controls, moderation effectiveness, and geographic consistency of product protections. They do not establish a current legal breach or a quantified incident rate; no such public metric is retained in this chapter’s assigned sources.[CO020, CO021]

1.5 Exhibits

2.1 Market boundary and adoption logic

The relevant market is not all digital retail. Carousell's own fact sheet describes two linked models: multi-category classifieds that let users buy and sell secondhand directly, and an omnichannel recommerce business that adds consigning, trusted partners, and physical retail. The category boundary therefore includes peer-to-peer listings, professional listings, recommerce services, and selected high-value verticals such as cars, luxury, mobile phones, fashion, home and living, property, and services. It excludes first-party general merchandise retail, food delivery, and payment value that does not pass through Carousell's marketplace. The buyer value proposition is affordability, discovery, and convenience; the seller value proposition is monetizing unused goods or acquiring demand. In practice, the substitute set includes new-item e-commerce, offline classifieds, social selling, specialist vertical marketplaces, and simply retaining or giving away an item. This boundary matters because a regional e-commerce headline is an upper context, not Carousell's own serviceable market.[CM001, CM002, CM006, CM008, CM010]

2.2 Multiple sizing lenses and what they do not prove

Two third-party lenses quantify the surrounding market but use different definitions and methodologies. Statista's current topic page reports an e-commerce market value of US$181 billion for Southeast Asia and separately lists a US$263 billion internet-economy figure; its page also notes that its statistics have different update cycles and may not be complete. UnivDatos reports US$201.92 billion for Southeast Asia e-commerce in 2024 and a forecast CAGR of 21.13% for 2025–2033, segmented by type, payment method, and country. These figures are not interchangeable: one is a current topic-page statistic with access limitations, while the other is a paid-market-study estimate whose methodology combines secondary research, expert interviews, top-down and bottom-up approaches, and triangulation. A narrower recommerce lens is supported by the Carousell Recommerce Index summary reported by eTailAsia: 72% of 3,029 respondents across eight markets had purchased secondhand before. That is a survey incidence metric, not a GMV or revenue estimate. The evidence therefore supports a large and growing context plus meaningful secondhand adoption, but not a precise Carousell TAM, SAM, or SOM. The missing bridge is category-level GMV, take rate, active transacting users, and country-level reach.[CM003, CM004, CM007, CM014, CM016, CM017, CM018]

2.3 Buyers, budget owners, and adoption constraints

The demand system has several distinct budget owners. Individuals are both users and payers for low- and mid-value secondhand goods; professional sellers, agents, merchants, and dealers are the economic buyers for visibility, subscriptions, or transaction tools; and category partners or advertisers fund high-intent commercial placements. The Singapore Management University case records Carousell's freemium model, paid visibility services for used-goods sellers, and subscriptions for professional sellers such as real-estate agents. Growth drivers include mobile penetration, local payment rails, affordability, sustainability preferences, and the shift from new-item purchase to resale. eTailAsia's eight-market survey provides evidence of secondhand familiarity, while the Statista page highlights social commerce and major platform competition as structural forces. Constraints are equally material: users may leave the app for messaging, trust and fraud concerns can slow conversion, localization raises operating complexity, and high-value categories need verification, delivery, or financing. Sendbird's customer case says off-platform conversations reduced Carousell's visibility into failed transactions and unacceptable behavior; Carousell's official fact sheet describes buyer protection, tracked delivery, verification, and fraud prevention as trust enablers. Adoption timing and valuation therefore depend on monetizing intent without degrading liquidity or trust.[CM006, CM007, CM011, CM012, CM015]

2.4 Exhibits

3.1 Competitive landscape and buyer alternatives

The relevant competitive set is broader than named resale apps. Carousell's own product surface spans fashion, luxury, furniture, electronics, cars, collectibles, home services, integrated delivery, escrow-protected buyer protection, and direct selling to Carousell in selected markets. That creates direct overlap with peer-to-peer classifieds and recommerce platforms, but also exposes the company to category specialists and generalist commerce platforms with stronger logistics or checkout. The 2020 SMU case describes Facebook Marketplace as a new entrant contesting the classified-platform business, while KrASIA identifies Carsome and Carro as strong used-car players. The practical buyer choice is therefore category-specific: a seller can list on Carousell, cross-list to another marketplace, use a specialist dealer, or keep the item offline. Carousell's regional presence and local teams help it compete where language, trust, and local transaction norms matter, but do not remove the need to win each vertical.[CP001, CP002, CP003, CP004]

3.2 Capability, distribution, and trust comparison

Carousell's differentiation is a bundle rather than a single feature. Its app listing flow, buyer-seller chat, ratings, localized market operations, certified categories, delivery, and buyer protection address the friction between discovery and completed trade. The Google customer story reports that Gemini reduced median listing time to less than two minutes and increased items listed per seller by 20%; these are company/partner-reported operating outcomes, not independent benchmarks. Sendbird's customer case shows why retaining communication in-app matters: off-platform messaging fragmented the experience and reduced visibility into failed transactions and unsafe behavior. This supports a credible trust-and-conversion thesis, but also reveals dependency on third-party infrastructure and the ongoing cost of moderation. Competitors can match individual features; the harder question is whether they can match local liquidity and trusted completion rates simultaneously.[CP005, CP006, CP007, CP008]

3.3 Pricing, packaging, and switching costs

The observable monetization model is freemium and attention-based: the SMU case describes paid visibility services for used-goods sellers and subscriptions for professional sellers, while the App Store listing shows Carousell Coins and a paid three-day bump. This supports low-friction entry, but it also means sellers can multi-home and compare reach before paying. Carousell's direct-sale, certified, delivery, and protection services can deepen monetization beyond listing promotion, especially in higher-value categories, yet public evidence does not disclose realized take rates, seller retention, or category-level contribution margins. Switching costs therefore sit mainly in accumulated reputation, listing inventory, buyer conversations, and local liquidity rather than contractual lock-in. A seller with a differentiated audience or inventory can still cross-list, and a buyer can move to Facebook Marketplace, Shopee, Lazada, or a specialist auto platform when price, trust, or fulfillment is better.[CP009, CP010, CP011, CP012]

3.4 Moat durability and displacement risk

The moat case is conditional. Carousell has operated across seven markets, built a decade of buyer-seller behavior data, and is adding AI to listing, fraud detection, translation, and discovery. Its FY25 announcement says recommerce revenue grew 40% year on year, represented 45% of revenue, and that 29 physical stores contributed to more than 20% GMV uplift in selected businesses. Those claims indicate strategic movement from classifieds toward trusted, transaction-led recommerce, but they are not proof that the model is insulated from competition. KrASIA reports that trust remained the biggest barrier in automotive commerce and that competition was expected to intensify. The most material displacement risks are a specialist winning a high-value vertical, a horizontal platform subsidizing distribution, or fraud and moderation costs rising faster than monetization. FY26 execution should be judged on repeat usage, completed transactions, take rate, and category contribution rather than listing volume alone.[CP013, CP014, CP015, CP016]

3.5 Exhibits

4.1 Revenue model and mix

Carousell's disclosed model is broader than a single classifieds fee. Company statements reported by current news sources describe multiple revenue streams and a growing transaction business, while a lower-reputation business-model explainer identifies paid listing visibility, seller subscriptions, transaction fees, and advertising. Those mechanisms should not be treated as audited revenue lines or current realized pricing. The strongest quantitative evidence is FY25 revenue of US$141 million, up 18% from US$119.3 million in FY24. Recommerce grew about 40% and reached 45% of group revenue, showing that transaction-led activity has become material. Public evidence still does not separate subscription, advertising, transaction, or classifieds revenue, so recurring revenue, take rate, concentration, and recognition policy remain open.[CI001, CI003, CI006, CI007]

4.2 Monetization and service-delivery economics

The public record supports several monetization points but almost no pricing depth. Historical academic evidence describes a freemium model with premium visibility services and professional-seller subscriptions. Current reporting adds payments, shipping, inspection, trade-in, and physical stores as trust and convenience infrastructure. Carousell operated 29 stores by FY25 year-end and attributed more than 20% GMV uplift across luxury, mobile phones, and fashion businesses to those stores, but the statement supplies neither a baseline nor store-level contribution. A supplier case says buying chat technology reduced engineering cost and accelerated transaction velocity from hours to minutes. Both are useful operating proxies, yet neither establishes CAC, payback, gross margin, refund cost, fraud loss, or incremental profit.[CI004, CI005, CI009, CI012, CI013]

4.3 Profitability and capital adequacy

Carousell reported positive FY25 EBITDA, but the amount was not disclosed. Management told The Business Times that it intended to preserve EBITDA positivity, maintain cost discipline, and keep investing selectively in AI and recommerce; it also characterized the balance sheet as strong without publishing a cash or debt figure. Positive accounting EBITDA is not cash generation, and historical financing cannot be treated as cash still available. A 2021 US$100 million financing at a reported US$1.1 billion valuation demonstrates past access to capital, while contemporaneous coverage noted that the announced use of proceeds lacked specifics. The assigned evidence provides no current cash, burn, runway, debt facilities, covenants, working-capital profile, or next-round trigger.[CI002, CI010, CI011, CI014, CI015]

4.4 Financial verdict and diligence threshold

The operating trajectory has improved: revenue grew, recommerce became a large mix component, and the company crossed its stated EBITDA threshold. The evidence is still insufficient to infer durable free cash flow. An academic case frames fraud, new competition, and possible tension between paid services and community participation as unresolved business-model risks. Those risks matter financially because higher trust-and-safety expense or seller resistance could absorb the apparent operating leverage. The underwriting stance should therefore remain conditional. Before assigning value to the profitability inflection, diligence should reconcile revenue by stream, gross and contribution margins, EBITDA adjustments, store cohorts, CAC and retention, cash conversion, and downside runway. Unsupported quantities should remain explicit gaps rather than scenario inputs presented as facts.[CI001, CI002, CI003, CI016]

4.5 Exhibits

5.1 Product and operating model

Carousell delivers a peer-to-peer marketplace in customer workflow terms: a seller creates a listing, a buyer discovers it, the parties communicate, and the transaction can move through in-platform payment and delivery. The product is no longer only an open listings surface. Tech in Asia describes a group of nine brands in seven markets spanning online and omnichannel resale, while the company’s programs add managed buying, consignment, certification, and physical drop-off or retail touchpoints. The result is a portfolio operating model: the general marketplace supplies liquidity and discovery, while vertical brands and managed services address categories where trust, condition, or convenience matter more. This is a plausible product wedge, but public disclosures do not establish the relative contribution, repeat usage, or profitability of each workflow.[CE001, CE002, CE003]

5.2 Workflow, trust, and transaction controls

The transaction workflow is built around reducing two marketplace frictions: uncertainty and coordination cost. The platform supports chat, reviews, verified badges, secure payment, buyer protection, delivery tracking, and dispute handling; selected verticals add inspection, authentication, warranties, or diagnostic checks. Sendbird’s customer case describes the earlier problem clearly: buyer-seller communication moved to WhatsApp or iMessage, fragmenting the experience and reducing Carousell’s visibility into trust-and-safety issues. Carousell’s own safety surface says in-platform records can support investigation and dispute resolution. These controls improve observability and conversion potential, but they do not eliminate off-platform behavior or fraud. Singapore’s MHA reported that scam cases on Carousell rose from 470 to 513 year over year for April-June while total reported scams fell from 1,125 to 720, showing why control effectiveness needs to be tracked by denominator and scam type.[CE004, CE005, CE006, CE007, CE008]

5.3 Architecture, data, and differentiation

The public record supports an applied marketplace technology stack rather than a fully disclosed proprietary infrastructure architecture. It includes mobile listing and discovery, localized content and support, real-time chat, moderation filters, machine-learning reply suggestions, secure payments, logistics integrations, verification signals, and category-specific inspection or pricing tools. A case study says Carousell used Sendbird rather than building chat internally and implemented presence, delivery receipts, read receipts, typing indicators, profanity filters, regular-expression moderation, and shadow banning. KrASIA reports that the auto product was developing AI-assisted detailed listings, partnerships for certified vehicle condition, and use of historical data to improve price fairness. These are concrete capabilities, but vendor-managed components and partner dependencies mean diligence should test portability, data governance, model performance, service-level history, and whether data advantages translate into defensible liquidity rather than merely feature breadth.[CE009, CE010, CE011, CE012]

5.4 Maturity, roadmap, and technical risks

Product maturity is mixed by layer. Core marketplace functionality is established across multiple markets, while certified programs, managed resale, physical retail, AI-assisted trust, and specialist acquisitions are still evolving. CNBC reported in July 2026 that the CEO said AI was being used to prevent bad actors and proactively detect and remove bad listings. That is a current management statement, not an independently measured safety outcome. MHA’s continuing monitoring and the company’s planned stronger authentication options indicate that account takeover and identity abuse remain live operating risks. The product roadmap therefore has a dual mandate: expand convenience and vertical depth while improving safety controls without imposing excessive seller friction. The public evidence does not disclose uptime for the whole marketplace, incident severity, false-positive rates, model recall, security audit scope, or category-level unit economics; these are material implementation and underwriting gaps.[CE013, CE014, CE015]

5.5 Exhibits

6.1 Customer segments and value exchange

Carousell's customer base is a two-sided local marketplace rather than a single enterprise buyer. Individual sellers use mobile listing, chat, ratings, and local delivery to turn unused goods into cash; buyers seek lower-priced, unique, or nearby inventory; and merchants use storefront and visibility tools to reach demand without building a separate commerce stack. The same product supports fashion, electronics, furniture, luxury, autos, and home services, so the economic value of a customer varies materially by category. Public evidence identifies the roles and use cases, but does not disclose segment-level revenue, active-buyer frequency, or seller conversion denominators.[CU001, CU003, CU004, CU005, CU006]

6.2 Adoption surfaces and operating scale

Adoption is visible through several different lenses that should not be treated as interchangeable. Historical reporting cited more than 250 million listings and tens of millions of users, while a current partner case describes tens of millions of monthly active users across seven markets. The App Store listing provides a current product surface and a 4.7/5 rating from 3.1K ratings, but ratings are not retention or transaction completion. Google reports that its AI listing workflow reduced median listing time to under two minutes and contributed to a 20% increase in items listed; this is a company case-study result, not an independently audited cohort measurement.[CU002, CU004, CU005, CU013, CU014]

6.3 Named customer and partner proof

Named proof is strongest where a customer or merchant describes a concrete use case rather than where Carousell displays a logo. The Straits Times profile gives a named used-car merchant, a stated fleet size, and a reported improvement in inquiry conversion after the circuit breaker. Carousell's SME and community programmes also provide examples of merchant onboarding and demand generation, although these are campaign-level outcomes rather than recurring-revenue proof. The LuxLexicon acquisition supplies a named specialist capability in luxury resale and a category-growth signal, but the purchase price and listing growth were reported by sources connected to the transaction and do not establish retention.[CU007, CU008, CU011, CU015]

6.4 Durability, expansion, and unresolved concentration

Durability depends on whether Carousell can convert episodic decluttering into repeat marketplace activity and paid merchant usage. Localization, in-app chat, escrow or buyer protection in selected markets, certified categories, and market-specific logistics reduce friction, while auto, luxury, and recommerce categories create expansion paths. However, public sources do not provide NRR, GRR, churn, repeat-purchase cohorts, customer concentration, CAC payback, or category-level contribution margins. Trust and safety also remains a structural customer risk: the platform must detect fraud and off-platform behavior without making casual sellers or buyers abandon the marketplace. These are diligence gaps, not evidence that retention is weak.[CU009, CU010, CU012, CU014, CU016]

6.5 Exhibits

7.1 Ranked risk assessment

Carousell has moved from a lightweight classifieds product toward managed recommerce, which raises the operating bar. The strongest evidence supports a high residual risk around trust and transaction integrity: a 2020 academic case explicitly framed online fraud as an increasing problem, while a supplier case study says off-platform messaging weakened review of failed transactions and unacceptable behaviour. Carousell responded with in-app moderation and a security-focused chat implementation, but those controls are described by the technology vendor rather than independently measured. Competitive risk is also material. Current regional evidence describes Shopee and TikTok Shop taking most tracked platform GMV while Lazada contracts, against an estimated US$159 billion regional e-commerce market in 2024. Independent reporting says price competition, promotions, warehousing and logistics are pressuring profitability across the sector. These dynamics can raise acquisition costs or force Carousell to spend more to preserve liquidity and trust. Positive adjusted EBITDA is encouraging, but it is one reported operating milestone, not evidence of durable free cash flow or a disclosed reserve for fraud, disputes, and compliance.[CR001, CR002, CR003, CR004, CR005, CR006, CR013]

7.2 Regulatory, legal, operational and security exposures

The public pool does not document a current enforcement action, license failure, material outage, breach, or litigation outcome for Carousell; that absence should remain an evidence gap rather than a clean bill of health. Marketplace operations span multiple countries, languages, currencies, logistics environments and product categories. A localization source describes local teams and country-specific experiences, while another source records partnerships with delivery providers and government-backed digitalisation programmes. This footprint increases the number of jurisdictions and counterparties in which privacy, consumer protection, payments, advertising, counterfeit, safety and employment obligations may differ. The platform's own transaction design also matters: a vendor case says users previously left the app for WhatsApp or iMessage, limiting data governance and trust-and-safety review, and reports moderation tools for profanity, PII and bad actors. The control is plausible, but the public evidence does not quantify false negatives, appeals, response times, or whether the same controls cover every market and category. The diligence priority is a market-by-market control inventory tied to incident and complaint data.[CR007, CR008, CR009, CR010, CR011]

7.3 Partner, dependency and financial-model risks

Carousell's regional model depends on local market liquidity, category specialists, delivery and payments partners, and the ability to monetize both individual sellers and professional merchants. Historical reporting describes a hybrid footprint that includes general classifieds and autos, with partnerships and acquisitions used to deepen market coverage. That creates execution leverage but also dependency risk: a logistics or payments partner failure, a weak local team, or a category-specific trust event could impair conversion in one market without being visible in group-level revenue. Automotive is especially exposed to certification and assurance requirements; a sector source identifies trust as the biggest barrier and says competition is likely to intensify. Financial disclosure is incomplete. The latest available article reports US$141 million revenue, 18% year-on-year growth and first positive EBITDA, but neither it nor the assigned evidence discloses EBITDA margin, cash, debt, working capital, fraud loss, or trust-and-safety spend. Investment judgment should therefore use explicit monitoring triggers rather than treating adjusted EBITDA as proof that the model is self-funding.[CR012, CR013, CR014, CR015, CR016]

7.4 Monitoring indicators and thesis-break triggers

The risk posture is monitorable if management supplies operating data that public sources do not provide. The first trigger is trust: rising scam complaints, counterfeit or unsafe-listing takedowns, dispute losses, or slower moderation responses would indicate that transaction growth is outrunning control capacity. The second is competitive economics: sustained share loss, higher promotional intensity, or seller churn as Shopee and TikTok Shop scale would pressure monetization and require a clearer category strategy. The third is profitability quality: a reversal from positive EBITDA, rising trust-and-safety or logistics costs, or a cash balance that cannot fund a downside period would break the operating-improvement thesis. The fourth is dependency resilience: a material partner outage, regulatory restriction, or failure of an acquired/local platform to meet service or compliance standards would test whether Carousell has substitutes. These are diligence gates, not forecasts; current public evidence is insufficient to set numeric thresholds for most of them.[CR003, CR005, CR008, CR014, CR016]

7.5 Exhibits

8.1 Investment case and anti-thesis

The investable thesis is local marketplace density paired with paid visibility, professional-seller subscriptions, and higher-value vertical services rather than a generic software multiple. Singapore Management University documents the evolution from a mobile-first classifieds community into paid premium services and subscriptions for professional sellers, while Carousell's own survey evidence points to broad secondhand adoption across eight Asian markets. The anti-thesis is substantial: users can multi-home, fraud and moderation are costly, and the available public record does not disclose the current operating data needed to test whether liquidity converts into durable margin. Supplier-reported chat performance is useful product evidence, but it is not independent proof of company-wide transaction economics.[CV001, CV002, CV003, CV004, CV011]

8.2 Price discipline and comparable signals

The retained evidence consistently places Carousell's $1.1B valuation in the September 2021 financing; it does not establish a current primary or secondary clearing price. Consequently, neither an entry multiple nor a numeric Carousell valuation range is supportable from public evidence. Comparator dispersion reinforces the need for business-model-specific underwriting: Sacra estimates Back Market's 2024 revenue at $415M and an approximately 13x revenue multiple, whereas the retrieved October 2024 ThredUp page reports a 0.28 price-to-sales ratio. These are contextual observations, not independent benchmarks for Carousell, because category mix, geography, growth, profitability, and transaction structure differ.[CV005, CV006, CV007, CV008]

8.3 Scenarios and decision

The base case is continued private compounding with a track recommendation: engage in diligence, but do not convert the 2021 mark into a present buy price. The bull case requires recurring professional-seller or vertical revenue, durable marketplace liquidity, and contribution-margin evidence sufficient to support a premium marketplace multiple. The bear case is a valuation reset if multi-homing, competition, fraud, or moderation costs prevent operating leverage. Because current Carousell revenue, EBITDA, cash flow, dilution, and price are not established by retained public evidence, every scenario leaves the valuation range unquantified and uses operating milestones rather than false precision.[CV009, CV010, CV011, CV012]

8.4 Exit readiness and final diligence asks

Vinted's April 2026 secondary transaction shows the profile that can attract institutional liquidity: an €8B equity valuation alongside reported 2025 GMV of €10.8B, revenue of €1.1B, and net profit of €62M. That transaction is a comparator signal, not a transferable Carousell valuation, and Vinted raised no new primary capital in it. Carousell could pursue a future listing, strategic sale, or continued private compounding, but a premium exit requires audited reporting, segment economics, governance, and a credible margin narrative. The final diligence gate is a current security price and terms, audited financials, cohort retention, paid-seller conversion, fraud-loss trends, vertical contribution margins, debt, and a fully diluted preference waterfall.[CV013, CV014, CV015, CV012]

8.5 Exhibits