---
title: "Encar data: reading the Korean used car market in English"
description: "Encar is South Korea's dominant used-vehicle platform and a major export source. What its listings contain, why Korean market data behaves differently, and how to consume it without reading Korean."
slug: encar-korean-used-car-data
canonical: https://thecarapi.com/blog/encar-korean-used-car-data
category: Sources
author: TheCarApi Engineering
published: 2026-04-29
updated: 2026-08-07
tags: [Encar, Korea, used cars, vehicle export]
---

# Encar data: reading the Korean used car market in English

Korean supply is the most underused dataset in European vehicle sourcing, mostly because of a language barrier that is a solved problem.

Encar is the dominant used-vehicle marketplace in South Korea, and South Korea is a significant net exporter of used cars. For European buyers, exporters and marketplace operators, that combination makes it one of the more interesting supply pools available — and one that most tooling ignores, largely because the listings are in Korean.

This covers what is actually in the data, the ways the Korean market differs from European wholesale in ways that affect your code, and how to work with it in English.

## Why the Korean market behaves differently

Four structural properties, each of which has a consequence for anyone modelling the data.

### Inspection culture

Korean used-vehicle sales operate with a strong emphasis on documented inspection and disclosed accident history. Listings routinely carry structured condition information at a level of detail that is unusual by European wholesale standards. For a data consumer this is a genuine advantage: the condition signal is richer, and it is more consistently present.

### Odometer reliability

Mileage disclosure norms in the Korean market are widely regarded as strong. That does not make individual readings guaranteed, and normal diligence still applies — but as a population, Korean mileage data has less of the systematic distortion that makes European odometer figures a modelling hazard.

### Specification divergence

This one will bite you. Korean-domestic-market trims frequently do not correspond to European trim names for the same nameplate. A Korean-market Sonata or K5 may have equipment combinations, engine options and trim labels that simply have no European equivalent. Any model catalog built purely from European listings will fail to match a meaningful share of Korean stock.

### Third-country status

For an EU buyer, Korea is outside the customs union. Import duty and import VAT apply, on top of shipping that takes weeks rather than days. Landed cost is not a small adjustment to the listed price — it is a different calculation entirely, and it is why fee modelling has to be per-source rather than one formula. See [the import calculator endpoints](/docs/calculator).

> **Do not reuse EU landed-cost logic for Korean stock** — Applying an EU-internal cost model — no duty, VAT already accounted for — to a Korean vehicle understates the final bill substantially. If your product shows a landed price, the source has to drive which model runs.

## The language problem, and what solving it means

Encar listings are written in Korean. Machine-translating them at read time is the obvious approach and the wrong one, for reasons that are worth spelling out:

- **Vehicle terminology translates badly out of context.** Trim names, equipment codes and damage terms are domain jargon, and general-purpose translation mangles them in ways that look plausible.
- **It is not idempotent.** Translate on every read and the same field yields slightly different English on different days, so your filters and your cached values disagree.
- **It is expensive at volume** and adds latency to every request.
- **It cannot be indexed.** You cannot search or facet on text that only exists after a translation call.

Translation belongs at ingest, once, with the result stored as a first-class field and the Korean original retained. That gives you something searchable, stable and auditable:

```json
{
  "auction_id": 38112900,
  "site_name": "encar",
  "car_name_en": "BMW 320d xDrive Touring M Sport",
  "clean_make": "BMW",
  "clean_model": "320d",
  "registration_year": 2020,
  "mileage": 45000,
  "fuel_group": "Diesel",
  "gearbox_group": "Automatic",
  "public_price_eur": 21500,
  "car_country_extended": "KR",
  "car_identification": { "...": "original Korean payload, unmodified" }
}
```

_`car_name_en` is resolved at ingest and indexed. The Korean source payload stays in `car_identification` so a disputed translation can be checked against what the listing actually said._

## Querying Korean stock

Encar is the site slug `encar`, and it takes the same parameters as every other source:

```bash
# Korean stock, petrol automatics from 2019, under 60,000 km
curl -s "https://api.thecarapi.com/api/search?site=encar&fuel=Petrol\
&gearbox=Automatic&year_from=2019&kilometers_to=60000&sort=price_low&limit=24" \
  -H "X-API-Key: $API_KEY"

# The Korean domestic marques carry the most volume
curl -s "https://api.thecarapi.com/api/search?site=encar&brand=hyundai&limit=24" \
  -H "X-API-Key: $API_KEY"
```

> **Two useful filter habits for Korean stock** — Drive brand and model values from [the catalog endpoints](/docs/catalog) rather than hardcoding them — Korean-market nameplates include models never sold in Europe. And filter on `country` rather than assuming the source implies it, because normalized country is what your users actually mean.

## What Korean data is good for

Three uses where it earns its place rather than just adding rows:

1. **Arbitrage analysis.** The same model, the same year, priced in two disconnected markets. Comparing Korean listing prices against European retail asking prices from [the classifieds surface](/market-data) is the clearest form of this — but the comparison is only meaningful once landed cost is included.
2. **Supply for models that are scarce in Europe.** Certain hybrids and Korean-market specifications are far more available domestically than they are through European wholesale.
3. **Condition-rich training data.** The disclosure norms make Korean listings unusually good input for anything modelling the relationship between documented condition and price.

## Practical cautions

- **Currency and FX timing.** Prices originate in Korean won. Know whether the euro figure you are reading was converted at ingest or at read time, because on a volatile pair those differ.
- **Specification matching.** Budget for the fact that a European model catalog will not cleanly match all Korean trims. Fall back to the normalized model with the raw variant preserved rather than dropping unmatched rows.
- **Shipping duration is part of the price.** A car that arrives in two months carries two months of market movement and depreciation risk that a German car collected next week does not.
- **Right-hand drive is not an issue here** — Korea drives on the right and produces left-hand-drive vehicles, unlike Japanese domestic supply.

For the wider picture of how Encar fits alongside the European sources, [the sources page](/sources) has per-platform detail, and [the auction data guide](/blog/car-auction-api-guide) covers the cross-source fundamentals.

## Frequently asked questions

### Is there an Encar API for developers?

Encar does not publish a general-purpose, self-serve English-language inventory API for third-party developers. Access to its data programmatically generally runs through a commercial arrangement or a data provider that maintains the integration. Verify current options directly with Encar if it is central to your product.

### How do you handle Korean-language listings?

Translation belongs at ingest rather than at read time. Resolving an English vehicle name once and storing it as an indexed field makes it searchable, stable across reads, and cheap — whereas translating on every request is slow, non-idempotent, and produces text you cannot filter or facet on. The Korean original should be retained so translations can be audited.

### Is Korean used car mileage trustworthy?

Korean market disclosure norms around odometer readings and inspection are widely regarded as strong, and as a population the data shows less systematic distortion than some European sources. That is a statement about the population, not a guarantee about any individual vehicle — normal verification still applies.

### Do Korean cars cost more to import into the EU than German ones?

Yes, and materially. Korea is outside the EU customs union, so import duty and import VAT apply on top of sea freight that takes weeks. An EU-internal purchase from Germany incurs neither. Any landed-cost calculation has to branch on the source country — applying EU-internal logic to Korean stock will understate the final figure substantially.

### Are Korean cars left-hand drive?

Yes. South Korea drives on the right and its domestic vehicles are left-hand drive, so the steering-side objection that applies to Japanese domestic stock does not apply here. Specification differences from European-market trims are the real matching challenge, not drive side.