Turkey is one of a handful of large consumer markets where the global marketplaces never took the lead. Amazon operates a Turkish storefront, but domestic demand concentrates on two platforms that were built for local payment habits, local logistics and a local returns culture. For a brand outside the country, that changes the entry question from “how do I add a country to my existing account” into “how do I onboard twice, in a regulatory environment I have not met before”.
This piece walks through what the two platforms actually ask of a non-Turkish company, where the tax and entity requirements sit, how the money moves, and which parts of the model surprise sellers who arrive from the EU, the UK or the US. It is a market map rather than a filing manual, and the legal points below are attributed to the bodies that publish them. For the wider framing, see our guide to understanding global trade for retail and cross-border commerce.
In short
- Two platforms, not one market. Trendyol leads Turkish e-commerce by a wide margin, with Hepsiburada the clear second. Covering the market in practice means running both seller accounts, not picking a winner.
- The entry barrier is administrative, not commercial. A Turkish tax identification number, a route to issuing Turkish electronic invoices and a Turkish IBAN for payouts are the three practical gates, and all three usually involve a local representative or a local entity.
- Currency is a live cost, not a footnote. The Turkish Statistical Institute (TÜİK) reported annual consumer inflation of 31.51% in August 2026, so pricing, payout timing and restocking cadence interact in a way they do not in a low-inflation market.
- Returns are generous by law and by habit. Turkey’s Regulation on Distance Contracts gives consumers a 14-day right of withdrawal, and category norms on these platforms push practical return rates above what many apparel sellers plan for.
- Both owners are now foreign. Alibaba Group holds the majority of Trendyol, and Kaspi.kz has consolidated control of Hepsiburada since January 2025, which is reshaping how both platforms treat cross-border supply.
How the Turkish market splits between platforms
The scale of the opportunity is documented by the Turkish Ministry of Trade (Ticaret Bakanlığı), which publishes an annual e-commerce outlook. Its 2025 edition put total e-commerce volume at TRY 4.57 trillion, up 52.2% year on year, across 5.94 billion transactions. In dollar terms the Ministry reported roughly USD 115 billion, equal to about 6.9% of GDP and 19.3% of total trade. Those are national figures covering travel and services as well as goods, so retail e-commerce is the smaller sub-total the Ministry puts at TRY 2.46 trillion.
Within that volume, concentration is high. Trendyol is generally estimated to hold somewhere in the mid-30s to around 40% of Turkish e-commerce, with Hepsiburada second at roughly 17% to 20%. Those shares come from market research rather than from either company, so treat them as directional. The structural point is more reliable than the decimals: two platforms absorb more than half the market, and neither is a global account you can extend with a checkbox.
Hepsiburada’s own reporting gives a harder number on its side of the split. For full-year 2025 it reported GMV of TRY 257.5 billion, revenue up 13.4% to TRY 84.65 billion, and 87.5 million orders, with an active merchant base of about 102,000 sellers. It also reported a widening net loss, which matters to sellers only in so far as loss-making platforms tend to revisit commission schedules and subsidy programmes. Trendyol does not publish equivalent audited retail detail, because it is not separately listed.
Why the global players did not take the market
The usual explanation is a stack of local advantages that compounded early: cash on delivery and installment card payments, same-day and next-day delivery built on domestic courier networks, and a merchandising style tuned to Turkish apparel manufacturing. Trendyol in particular grew out of fashion, which is also the country’s export strength, so supply and demand were colocated. This is the same pattern visible in several other markets, and the dynamics are covered in our broader look at regional marketplaces that rival Amazon in their home turf.
For a foreign seller, the practical consequence is that category depth and price expectations are set by local manufacturers. A European apparel brand entering Turkey is not competing against other importers. It is competing against factories three hours from the warehouse, which is a different margin conversation than entering a market where everything arrives by container.
A side-by-side view
| Dimension | Trendyol | Hepsiburada |
|---|---|---|
| Founded | 2010, Istanbul | 2000, Istanbul (listed as D-Market Elektronik Hizmetler ve Ticaret A.S.) |
| Controlling shareholder | Alibaba Group, reported at around 70% as of 2024 | Kaspi.kz, majority since January 2025 and above 85% of voting rights per its 2026 filings |
| Listing status | Private, last marked at a USD 16.5 billion valuation in an August 2021 round | Nasdaq-listed (HEPS), so quarterly financials are public |
| Estimated market share | Mid-30s to roughly 40% (third-party estimates) | Roughly 17% to 20% (third-party estimates) |
| Category centre of gravity | Fashion, beauty, home, plus grocery and rapid delivery through Trendyol GO | Electronics, appliances, home, broadening into general merchandise |
| Published seller base | Not separately disclosed | About 102,000 active merchants reported for 2025 |
| Cross-border posture | Operates its own outbound international storefronts across Europe and the Gulf | Historically Turkey-focused, now connected to Kaspi’s regional footprint |
One asymmetry is worth holding onto. Hepsiburada is a public company, so you can read its merchant economics in its filings before you commit. Trendyol is not, so your view of its take rates and support quality comes from the seller agreement and from other sellers.
Entity, tax and representation requirements
This is where most entry plans stall, and it is the part that genuinely differs from adding a European marketplace. Turkey’s framework treats the obligation to issue a compliant invoice to a Turkish consumer as a domestic tax obligation, which is hard to discharge from outside the country. The Turkish Revenue Administration (Gelir İdaresi Başkanlığı) operates the electronic invoice and e-archive systems that marketplace sales flow through, and participation in those systems is tied to a Turkish taxpayer record.
In practice, companies approach this in one of three ways. Some incorporate a Turkish subsidiary or branch. Some appoint a Turkish tax representative and register for a tax identification number without a full entity. Some sell through an importer or a local merchant of record that takes title to the goods and lists as the seller. Each route places the compliance burden, the margin and the customer relationship in a different place.
Value added tax (KDV) is the second piece. The standard Turkish VAT rate has been 20% since mid-2023, with reduced rates applying to defined categories, and the Revenue Administration publishes the current schedule. Rates and category boundaries change by presidential decree, sometimes at short notice, so the published rate on the day you price is the only one worth using. The same caution applies to any import duty position on goods you ship in.
What “representation” actually covers
The phrase gets used loosely, so it helps to separate the functions it bundles. A tax representative handles registration and filings. A legal point of contact receives official correspondence and consumer complaints. A customs broker clears goods. A merchant of record sells to the consumer in its own name. A single local partner may offer several of these, but they are distinct liabilities, and which ones you need depends on which selling model you choose.
There is also a platform-side obligation that sellers sometimes mistake for their own. Turkey’s e-commerce law, Law No. 6563 as amended by Law No. 7416 in 2022, introduced a licensing regime for large intermediary service providers. An intermediary whose net transaction volume in a calendar year exceeds TRY 10 billion and whose transaction count exceeds 100,000, excluding cancellations and returns, must obtain and renew a licence from the Ministry of Trade. That provision took effect on 1 January 2025. It applies to the marketplaces, not to individual sellers, but it is the reason platform terms have been tightening around seller data, advertising and own-brand competition.
Onboarding and document checks for foreign sellers
Both platforms run a seller application that includes a document review. The exact list changes, and each platform publishes its own current version in its seller centre, so the table below describes the categories of document rather than a definitive checklist. Treat it as a preparation map for a conversation with an advisor, not as a filing instruction.
| Requirement | What it typically means for a non-Turkish company | Usual friction point |
|---|---|---|
| Proof of corporate existence | Home-country registry extract, commonly apostilled and translated into Turkish by a sworn translator | Apostille turnaround in the home jurisdiction, not the Turkish side |
| Turkish tax identification number | Obtained through a tax office application, generally with local assistance | Requires a decision on entity versus representative first |
| Electronic invoicing capability | Access to the Revenue Administration e-invoice or e-archive systems, usually via a licensed integrator | Cannot be bolted on after listings go live without order gaps |
| Turkish IBAN for payouts | A local bank account, or a licensed payment intermediary arrangement | Bank onboarding for foreign-owned entities can run longer than the platform review |
| Brand authorisation | Trademark evidence or a distribution letter, depending on whether you own the brand | Grey-market listings in the category can complicate a brand registry claim |
| Local contact and support capability | Turkish-language customer messaging and a response time commitment | Platform service-level metrics apply from the first order |
The sequencing matters more than the list. Tax identification tends to gate bank onboarding, bank onboarding gates payouts, and invoicing capability gates order fulfillment. A seller who runs these in parallel without resolving the entity question first usually discovers the dependency at the worst moment, which is after listings are live and orders are arriving.
Language and content work
Turkish-language listings are not optional in any serious sense. Machine-translated titles survive in categories with no local competition and fail everywhere else, because search on both platforms rewards query-matched Turkish phrasing. Size conversion is a second trap in apparel and footwear, where returns driven by fit are the single largest avoidable cost.
Budget for a local content pass rather than a translation pass. The difference is that a content pass rewrites attributes, category mapping and size guidance for how Turkish buyers search, instead of rendering your existing copy in another language. Sellers who have done the same exercise for other non-English marketplaces will recognise the pattern from our write-up on Allegro for sellers expanding into Central Europe.
Commission, payout cycles and currency risk
Neither platform runs a single flat commission. Both publish category-level schedules in their seller centres and revise them periodically, with additional fees for fulfillment, advertising, campaign participation and in some cases payment installments. Reported ranges circulating in seller communities span roughly the low teens to the mid-20s as a percentage, but a specific number is only meaningful against a specific category on a specific date, so the schedule in your own seller panel is the figure to model from.
What generalises better is the structure of the cost stack. A realistic landed model for Turkey has five layers rather than two, and sellers who model only commission tend to be surprised by the third and fourth.
- Platform commission on the sale value, set by category.
- Fulfillment and shipping, which depends heavily on whether you hold local stock.
- Campaign and advertising participation, which is close to mandatory for visibility in competitive categories.
- Returns handling, including return shipping and the cost of goods that come back unsellable.
- Currency and timing, meaning the gap between the lira price you set and the hard-currency value you eventually repatriate.
Why the payout gap is a pricing input
Payouts on both platforms settle on a cycle rather than instantly, and the cycle lengthens for new sellers and for categories with longer return windows. In a market with low inflation, a two or three week settlement delay is a working-capital question. In Turkey it is also a margin question, because the lira you earn on day one does not buy the same replacement inventory on day twenty-one.
The TÜİK series makes the scale concrete. Annual consumer inflation averaged 58.5% across 2024 and 34.9% across 2025, and stood at 31.51% in August 2026. Even at the lower 2026 rate, a 30-day settlement cycle carries a measurable erosion against replacement cost, before any move in the lira against your reporting currency. Sellers who price once and leave it tend to discover this as a slow gross-margin leak rather than as a visible event.
Three responses show up repeatedly among established cross-border sellers in Turkey. The first is scheduled repricing on a fixed cadence tied to cost inputs rather than to competitor moves. The second is holding a lira working-capital buffer so restocking does not wait on repatriation. The third is accepting a lower nominal margin target in Turkey than in hard-currency markets, and judging the market on volume and brand presence instead. None of these is advice for a particular business, and the right mix depends on your cost base and your treasury position.
Fulfillment options and local warehousing
Both platforms offer a first-party fulfillment service alongside seller-shipped options, and both weight search visibility toward the faster delivery promise. Turkish buyers have been trained on next-day and same-day delivery in the major cities, so a cross-border listing quoting 7–14 days competes at a structural disadvantage in anything other than a genuinely unavailable product.
That pushes most serious entries toward local stock. The models in use break down roughly as follows, and each one shifts the entity and tax picture described earlier.
| Model | Delivery promise | Capital and compliance load | Best fit |
|---|---|---|---|
| Cross-border direct shipping | Slowest, typically a week or more | Lowest capital, but customs and invoicing friction per order | Testing demand, high-value or unique items |
| Platform fulfillment with local stock | Fastest, aligned with platform promise | Inventory committed in-market plus import clearance | Proven SKUs with stable velocity |
| Third-party logistics in Turkey | Fast, close to platform fulfillment | Moderate, with a local partner handling returns | Multi-channel sellers running both platforms |
| Local distributor or merchant of record | Fast, handled entirely locally | Lowest compliance load, lowest margin and least control | Brands prioritising presence over economics |
Importing stock introduces a customs position that is separate from the marketplace relationship. Tariff classification, valuation and any applicable trade remedies determine the landed cost, and those are administered by Turkish customs under its own rules rather than by the platform. The general mechanics of this layer, including why classification decisions tend to dominate landed cost, sit outside the marketplace relationship entirely.
The returns leg is the part people under-budget
Cross-border fulfillment can work for outbound orders and still fail on returns, because a return that has to travel back across a border is often worth less than the freight. Sellers who start cross-border frequently end up operating a hybrid: outbound from abroad, returns absorbed locally through a partner who consolidates, inspects and either restocks or liquidates. Setting that up before the first campaign peak is considerably cheaper than setting it up during one.
Returns culture and buyer expectations
The legal baseline comes from the Regulation on Distance Contracts, published in the Official Gazette on 27 November 2014 under number 29188, pursuant to the Law on Protection of Consumers No. 6502. Under that regulation a consumer may withdraw from a distance contract within 14 days without penalty and without giving a reason, with the period running from delivery for goods. The regulation also addresses return shipping cost, requiring that the consumer be informed of the amount, which may not exceed the delivery cost, and of which party bears it. Refunds are due within 14 days, with the clock starting from defined handover points. The Ministry of Trade publishes the regulation, and amendments have followed, so the current consolidated text is the one that governs.
The legal floor, however, is not what shapes seller economics. Platform policy and category habit sit above it. Both marketplaces compete partly on returns convenience, and in fashion the practical return rate on Turkish marketplaces runs well above what a seller used to wholesale or to a low-return category will have modelled. Multiple-size ordering with the intention of returning most of the order is a normal shopping behaviour, not an abuse pattern.
What that changes in a business case
Three adjustments follow, and they are mostly unwelcome. Contribution margin needs to survive a return rate at the high end of the category range, not the average. Sizing content earns a disproportionate return, because fit is the dominant return reason in apparel. And inventory planning has to account for units cycling out and back rather than selling once, which affects both availability and the condition of what eventually resells.
There is also a service dimension. Both platforms publish seller performance metrics covering response time, cancellation rate and dispute outcomes, and visibility is tied to them. A foreign seller without Turkish-language support coverage during Turkish business hours will accumulate metric damage quietly before it notices a ranking effect.
Trendyol expansion into neighbouring markets
The outbound story is the part of the Trendyol file that gets least attention from inbound sellers, and it is arguably the most strategically interesting. Trendyol entered the European market in October 2020, serving a broad set of countries, and opened its first office outside Turkey in Berlin in 2022. It expanded into Azerbaijan, Saudi Arabia and the United Arab Emirates in 2023, and has pushed further into Central and Southeastern Europe since, with Czechia, Greece, Hungary and Romania named in its 2024 plans. Reported customer counts include around 1.5 million in Germany and about 2 million in Azerbaijan as of late 2024.
For a seller already inside Trendyol’s Turkish ecosystem, that footprint is optionality. A seller account that works in Turkey can in principle serve demand in the Gulf and parts of Europe through the same platform relationship, with Turkish manufacturing and logistics behind it. Whether that is attractive depends on your own channel conflict position, because selling into Germany through a Turkish marketplace is not neutral with respect to your existing German channel.
The ownership angle on both sides
Both platforms are now controlled from outside Turkey, which is new. Alibaba’s majority position in Trendyol connects it to Chinese supply and to Alibaba’s cross-border tooling. On the other side, Kaspi.kz disclosed in filings with the US Securities and Exchange Commission that it completed the purchase of a majority of Hepsiburada on 29 January 2025 for total consideration of approximately USD 1,127 million, of which USD 600 million was paid in cash at closing and about USD 526.9 million was deferred. It subsequently raised its holding above 75%, and a January 2026 agreement for a further 32,885,686 ordinary shares at USD 97 million took it above 85% of voting rights.
Kaspi’s stated strategic logic is to port its Kazakhstan super-app model, which combines payments, marketplace and fintech in one application, into Turkey. If that proceeds, the seller-facing consequences are likely to show up in payments, installment financing and advertising products rather than in the listing flow. We looked at the signals behind that programme separately in our analysis of Kaspi’s super-app push into Turkey, and the same question of how a regional owner reshapes a national marketplace runs through our piece on Coupang for sellers expanding into South Korea.
How to judge whether Turkey is worth the setup cost
The honest framing is that Turkey has a high fixed cost of entry and a genuinely large addressable market behind it. The fixed cost is administrative: entity or representation, tax registration, invoicing integration, banking, local content and a support capability. None of that scales down for a small test, which is why cross-border direct shipping exists as a demand-testing route despite its delivery disadvantage.
A few conditions tend to separate the entries that work from the ones that stall. Products with a real differentiation against local manufacturing, because price-matching Turkish factories on commodity goods rarely ends well for an importer. A category where the legal 14-day withdrawal right and the practical return rate do not destroy contribution margin. A treasury position that can tolerate lira exposure across a settlement cycle. And an operational willingness to run two platform relationships instead of one.
Where several of those are absent, the usual alternative is a local distributor or merchant of record arrangement that trades margin and control for a near-zero compliance load. That is a smaller business than a direct entry, but it is a real one, and it keeps the brand present while the economics are tested. Framing the choice between the broader regional options is the subject of our wider view on global trade for retail and cross-border commerce.
Information, not legal or tax advice
Everything above is general information about how the Turkish marketplace environment works, written for commercial planning. It is not legal, tax or customs advice, and it is not a determination of what any particular company is required to do. Entity structure, tax registration, VAT treatment, invoicing obligations, customs classification and consumer-law compliance all depend on facts specific to your business, your goods and your home jurisdiction.
Before committing to a structure, the sensible step is to take the specifics to a licensed Turkish tax advisor or lawyer, a licensed customs broker for any import position, and the platforms’ own current seller documentation for fees and policies. Rules and figures in this area change, sometimes quickly: VAT rates move by decree, platform commission schedules are revised, and the e-commerce licensing regime under Law No. 6563 as amended continues to be implemented. Every rate, threshold and deadline mentioned here should be verified against the Ministry of Trade, the Revenue Administration or the relevant platform before it is relied on. Figures cited from company filings reflect what those companies reported on the dates stated.
Where this article describes regulatory action or third-party estimates, it reports them as attributed statements. Nothing here is intended to suggest that any named company has acted unlawfully.
FAQ on selling in Turkey
Can I sell on Trendyol as a foreign seller without a Turkish company?
Trendyol operates an international seller route, and companies do sell without a full Turkish subsidiary. In practice that route still generally involves a Turkish tax identification number, a compliant electronic invoicing arrangement and a Turkish IBAN or a licensed payment intermediary, which is why most sellers work with a local representative. Whether your specific structure qualifies is a question for a Turkish advisor and for Trendyol’s current seller terms, both of which change.
Do I need both Trendyol and Hepsiburada, or is one enough?
One platform reaches a substantial share of the market, and Trendyol is the larger of the two by every available estimate. Covering the market properly means both, because the category mix differs: Hepsiburada’s historical strength is electronics and appliances while Trendyol’s is fashion, beauty and home. Many sellers launch on one, prove the operating model, then add the second rather than onboarding in parallel.
What is the VAT rate on marketplace sales in Turkey?
The standard Turkish VAT (KDV) rate has been 20% since mid-2023, with reduced rates for defined categories. Rates and category boundaries are set by decree and have changed several times, so the Revenue Administration’s current published schedule is the only reliable reference. Your advisor should also confirm which party is responsible for collection and remittance under your particular selling model.
How long do payouts take, and can I be paid in euros or dollars?
Both platforms settle on a cycle rather than per order, and the cycle is typically longer for new sellers and for categories with extended return windows. Payouts are made to a Turkish IBAN in lira, so converting and repatriating is a separate treasury step with its own timing and cost. The current cycle length is shown in your seller panel, and it is worth confirming before you build a cash-flow model.
What return rate should I plan for?
The legal baseline is a 14-day right of withdrawal under Turkey’s Regulation on Distance Contracts, but the practical driver is category habit. Fashion and footwear see high return rates on Turkish marketplaces, driven substantially by fit, and multiple-size ordering is normal buyer behaviour. A business case that only survives at the category average return rate is fragile, so most experienced sellers model the high end.
Does the e-commerce licence requirement apply to me as a seller?
No. The licensing obligation introduced by Law No. 7416, which amended Law No. 6563, applies to large intermediary service providers whose annual net transaction volume exceeds TRY 10 billion and whose transaction count exceeds 100,000. It took effect on 1 January 2025 and is administered by the Ministry of Trade. It affects sellers indirectly, because platforms have adjusted their terms on data, advertising and own-brand activity in response.
Is cross-border shipping into Turkey viable, or do I need local stock?
Cross-border shipping works for demand testing and for genuinely unavailable or high-value products. It is structurally disadvantaged in competitive categories, because search visibility on both platforms favours the fast delivery promise that Turkish buyers now expect. Returns are the harder half of the problem, since a cross-border return often costs more in freight than the unit is worth.
How does Turkish inflation affect pricing decisions?
TÜİK reported annual consumer inflation of 31.51% in August 2026, after annual averages of 34.9% in 2025 and 58.5% in 2024. At those levels, the gap between the day you set a lira price and the day you receive and convert the proceeds has a measurable cost against replacement inventory. Sellers commonly respond with scheduled repricing tied to cost inputs and a local working-capital buffer, though the right approach depends on your own treasury position.
Does selling on Trendyol give me access to its European and Gulf markets?
Trendyol operates its own international storefronts, having entered Europe in October 2020 and added Azerbaijan, Saudi Arabia and the UAE in 2023, with further Central and Southeastern European markets following. A Turkish seller relationship can therefore open optionality in those markets, subject to Trendyol’s own programme rules. Whether that is desirable depends on your existing channel arrangements in those countries, because it can create conflict with distributors you already use.
Sources referenced in this article include the Turkish Ministry of Trade for e-commerce volume data and the Regulation on Distance Contracts, the Turkish Statistical Institute for inflation figures, filings by Kaspi.kz with the US Securities and Exchange Commission for the Hepsiburada transaction, Hepsiburada’s own reported results, and background on Trendyol‘s ownership and international expansion. Market share estimates are third-party and directional.