|
Why Us — Comparison of Alafranga with Platforms, Generalist Agencies, and Single-Language Specialists Sourcing six conference-grade Arabic-English interpreters in Istanbul is not straightforward. Arabic-English is one of our strongest pairs in the city, so we identified, vetted and assigned six interpreters with real background in migration and human rights, where the terminology around asylum, legal frameworks and procedure has to be right the first time |
Three parallel rooms meant three interpreting booths, two interpreters per booth working in rotation so quality held through full-day sessions. Each room ran on its own audio channel, with a soundproof booth, console feed and delegate receivers per workshop. A technician handled setup, sound checks before each session and live support across all three rooms.
Ahead of the event we sent agendas, speaker notes and reference material to every interpreter so they could prepare terminology in advance. All six were contracted and paid up front. One of our coordinators stayed on site for the full programme, moving between rooms to keep timing, handovers and equipment aligned.
| Platform / Marketplace | Generalist Multilingual Agency | Single-Language Specialist | Alafranga Language Solutions | |
|---|---|---|---|---|
| Business models | Automated matching. Client uploads file, algorithm assigns translator | Full-service agency covering all languages and domains with generalist vendor pool | One language pair, typically one or two translators operating independently | Founder-led specialist agency: Turkish at the core, 40+ languages expanded through real client demand |
| Who makes decisions | Algorithm | Project manager, often no domain background | The translator | Volkan Güvenç, founder and director, active in every client relationship since 2002 |
| Accountability | Platform terms of service | Account manager may change | The individual translator | Named founder with 23 years of institutional knowledge and direct accountability |
| Turkish capability | Translator pool of variable quality, no structural guarantee of Turkish specialisation | Turkish is one language among many, typically handled by a generalist vendor, not a Turkish specialist | Strong if the specialist is genuinely qualified, no surrounding infrastructure | Turkish founding language pair active since 2002, specialist vendor network, TSE terminology knowledge, regulatory framework coverage |
| Language depth vs breadth | Breadth only, no structural depth in any language | Breadth, all languages, variable depth | Depth in one pair, no capacity for multilingual programmes | Depth in Turkish + structured capacity in 40+ languages, every language added through real client demand, not catalogue expansion |
| Vendor selection | Automated or semi-automated lowest bidder or availability-based | Vendor pool, selection criteria vary, often opaque | The specialist themselves or a small fixed team | 5,100+ vendor records, ~2,000 active specialists meeting quality and pricing criteria selected by language pair, domain, CAT tool proficiency, and performance history |
| Subject-matter expertise | Not guaranteed, translator self-declares specialisation | Variable, dependent on which vendor is assigned | Strong if the specialist is genuinely domain-qualified | Domain-matched selection — engineering, legal, medical, energy, automotive, industrial automation — verified through real project performance, not self-declaration |
| ISO 17100:2015 certification | Rarely, platform model is incompatible with ISO 17100 workflow requirements | Sometimes, certification level varies | Rarely, individual specialists cannot certify under ISO 17100 | ✅ ISO 17100:2015 certified, audited by Bureau Veritas |
| Independent review | Not structural, depends on translator | Sometimes, often self-review or light post-editing | Rarely, self-review is standard | ✅ Structural, every HumanExpert project includes a second independent specialist working from source |
| ATC membership | Not applicable | Sometimes | Not applicable | ✅ ATC and EUATC member since 2007 |
| AI-assisted workflow | Generic MT, no TM or glossary governance, no controlled prompting | Variable, often uncontrolled MT with light post-editing | Variable, individual translators using MT tools independently | ✅ SmartEdit, AI draft via API inside MemoQ or SDL Trados, TM and glossary applied at prompt level, human review before every delivery |
| Translation memory | Sometimes, often platform-owned, not client-specific | Sometimes, may or may not be maintained across projects | Sometimes, maintained by the individual | ✅ Client-specific TM built and maintained across all projects, client owns their TM assets |
| Glossary governance | Rarely structural | Sometimes, quality varies | Sometimes | ✅ Client-specific glossary applied at prompt level for SmartEdit and locked in CAT tool for all workflows |
| Shared TM for team projects | Not applicable, single-translator model | Variable, depends on infrastructure | Not applicable | ✅ Real-time shared TM, all translators on the same project connected to the same memory simultaneously |
| CAT tools | Platform-native, not always compatible with client tools | MemoQ or SDL Trados, usually | Varies by individual | ✅ MemoQ and SDL Trados natively, client-specific configuration |
| Project management platform | Platform dashboard | Internal PM software, varies | Email and file exchange | ✅ ]Project Open[, open-source, hosted in Spain, active since 2009, covering client and translator portals, full project history queryable since 2009 |
| LSP / white-label capability | Not applicable | Sometimes, depends on agency policy | Not applicable | ✅ White-label Turkish translation for LSPs, inside client tools, under client brand, NDA standard, CAT discount matrix compatible |
| Data privacy | Platform terms govern data handling, data may be used for model training | Variable, depends on agency policy and tools | Variable | ✅ Client content processed inside CAT tool via controlled API, never used to train public models, NDA standard |
| GDPR compliance | Platform-level, not always granular | Variable | Variable | ✅ GDPR-compliant workflows, data handling documented per project |
| Regulatory documentation | Not suitable, no structural quality guarantee | Variable, depends on vendor assigned | Suitable if the specialist has regulatory experience | ✅ CE technical files, KVKK compliance, BDDK/SPK/EPDK submissions, TSE standards terminology — active regulatory documentation experience since 2002 |
| TSE standards terminology | Not available, no structural mechanism for standards reference | Rarely, not a stated capability | Only if the individual specialist maintains active standards access | ✅ Active TSE and TS EN standards reference — evirici vs invertör, fesih vs iptal, signal word conventions, Machinery Directive documentation |
| Sworn translation | Available through platform, quality not structurally guaranteed | Available, quality depends on vendor | Available if the specialist is a registered sworn translator | ✅ Available, fully human HumanExpert workflow, no AI involvement |
| Subtitle localization | Sometimes — generalist subtitlers | Sometimes — variable media specialisation | Rarely | ✅ Alafranga Media — Disney+ and Fox TV credits, dedicated PM and specialist team, separate operation at subtitlelocalization.com |
| DTP and file engineering | Rarely, translation only | Sometimes, outsourced | Rarely | ✅ In-house DTP, InDesign, Illustrator, QuarkXPress, multilingual layout, bidirectional text, CJK character handling |
| Multilingual programme management | Not suitable, single-project model | Available, but consistency across languages variable | Not suitable, single language pair | ✅ Multilingual programmes, one point of contact, shared glossaries, consistent QA across 40+ languages simultaneously |
| Client references — regulated industries | Not typically verifiable | Variable | Variable | ExxonMobil, Rolls-Royce, Siemens Gamesa, Daikin, Zeiss, Mitutoyo, Fluke, Schell GmbH, Oman Oil Company, Solplanet/AISWEI |
| Client references — legal | Not typically verifiable | Variable | Variable | Squire Patton Boggs, Ocean Works, European legal firms — Turkish jurisdiction, KVKK compliance, arbitration documentation |
| Client references — technology | Not typically verifiable | Variable | Variable | OpenAI, Getac, LetsGrow.com, Neuro.net, Playata, HiRez |
| Continuity across projects | Not guaranteed, different translator each time | Variable, depends on vendor availability | Strong if the specialist is available | ✅ Structural, same translators retained across client programmes, TM and glossary continuity built into every project |
| Communication | Platform ticketing system | Account manager | Direct with the translator | Direct with founder or senior PM — no account manager layer |
| Transparency | Platform dashboard, limited visibility into process | Variable | High, direct with translator | ✅ Full, translator and reviewer identities available on request, AI engine and prompt version disclosed for SmartEdit projects |
| Founded | Varies, most platforms post-2010 | Varies | Varies | 2002, Istanbul. Alafranga Europe Ltd registered in London. |
| Headquarters | Varies, often US or platform-native | Varies | Varies | 71–75 Shelton Street, London WC2H 9JQ, registered in England and Wales |
| Target market | Volume buyers, price-sensitive, low complexity | General corporate, all industries, all languages | Clients with a specific single-language need | European industrial, legal, and technology companies with Turkish language requirements and LSPs needing a specialist Turkish back-office partner |