transcosmos review

Each line is a datapoint with its grade. Grades are the ones the anchors act on: a sub-criterion scored 5 requires at least one E1 or E2 source, and evidence that is entirely E3 caps the sub-criterion at 3.

12 claims on this sheet, by grade: E1 4 E2 3 E3 5 E5 3 Counts sum above 12 because 3 claims cite more than one grade.

• 71,683 consolidated employees at 31 Mar 2026 (44,247 Japan, 27,436 overseas), 187 sites across 36 countries and regions, and 43,050 contact-centre workstations, on consolidated net sales of ¥393.9bn and operating profit of ¥16.6bn FY2026 Business Report, Tokyo Stock Exchange Prime 9715, posted 24 Jun 2026 E1
• ISO/IEC 27001:2022+Amd 1:2024 held under certificate JQA-IM0223, issued by the Japan Quality Assurance Organization, registered 18 Mar 2005 and expiring 17 Mar 2029, with scope explicitly naming "Contact Center Services; Business Process Outsourcing Services". ISO/IEC 27017:2015 also held (JQA-IC0092, same expiry). This is one of only two providers in the field publishing a certificate number, issuing body, expiry and scope trans-cosmos.co.jp security page E1E2
• Four overseas delivery entities are published against ISO/IEC 27001:2013 — an edition withdrawn after the October 2025 transition deadline: Thailand (IND18.0460/U), Taiwan (ISMS171), Malaysia (IS 713284) and Indonesia (IS 724974). Overseas delivery is 27,436 of 71,683 staff, so the currency of the certificates covering that share is unresolved on the published page trans-cosmos.co.jp security page, checked Aug 2026 E3
• Everest Group Leader, CXM Services in APAC PEAK Matrix Assessment 2024 — a fourth consecutive year, from 29 providers assessed. No Everest, ISG or Gartner placement dated 2025 or 2026 was located, and the company awards page carries no 2026 entries trans-cosmos.co.jp news, Sept 2024 The absence stated in this claim is established by our own search of the named platforms and is graded E5 accordingly; the positive finding keeps its own grade. E1E5
• COPC CX Standard for Outsource Service Providers certified by COPC Inc.: two Japan contact centres totalling roughly 800 agents including a 450-agent fully home-based centre (Nov 2022), and South Korea certified six consecutive years to Dec 2024 with 55 COPC CSP Implementation Leaders developed trans-cosmos.co.jp news, Dec 2024 E1
• PCI DSS v4.0 "Level 1 certification" is claimed for the South Korea BPO entity, announced 27 Jun 2024 as the first for a BPO in South Korea across 512 testing procedures. The release names no QSA, publishes no Attestation of Compliance, states no service scope, and no card-network registry listing was located; the claim covers one country entity only PRWeb, 27 Jun 2024 E3
• No GDPR processor posture is published at all: no public DPA, no subprocessor list, no data-residency commitment and no named DPO. The Japanese privacy policy cites only the Act on the Protection of Personal Information and JISQ15001, and the transcosmos Europe notice is a website-visitor notice rather than a client-facing processor agreement trans-cosmos.co.jp privacy, checked Aug 2026 E5
• Glassdoor 3.3/5 across 293 reviews, 58% recommend, 65% CEO approval, 39% positive business outlook; recurring recent themes include micromanagement, difficulty taking leave and inflexible management style. The employer footprint is fragmented across at least three Glassdoor entities Glassdoor, checked Aug 2026 E2
• Production AI outcomes published for South Korea contact centres, all with anonymised clients: human-agent connection rate fell from 32% to 13%; processing time from five minutes to one with response rate rising 50% to 85%; QA evaluation sample expanded roughly 620-fold; and self-service resolution rose from about 19% to 52% after replacing a repair-service voice bot with a voice AI agent. No client is named and no human-in-the-loop sampling or escalation-trigger design is described PR Newswire, 18 May 2026 E3
• Only one named-client case combines client identity, a date and a quantified outcome: Sumitomo Mitsui Trust TA Solutions (Jan 2021) — 9,200 annual labour hours saved, roughly 10%, and three days per person cut from new-hire training, quoted by the client’s president. The 14 named clients on the Japanese case index (including SMBC, Mizuho Bank, ANA, McDonald's Japan, Panasonic, NTT Docomo and Microsoft Japan) and the nine dated English cases carry no quantified outcomes trans-cosmos.co.jp customer case E3
• Independent vendor corroboration of platform and continuity: 8x8 documents transcosmos migrating four European contact centres to 8x8 Contact Center and 8x8 Work, later extended to Poland, Hungary and the Philippines, supporting roughly 40 BPO clients with IVR, live dashboards, wallboards and historical reporting, quoting the COO and Head of IT Operations by name on remote-work continuity through COVID-19 8x8 customer story E2
• No pricing model, minimum engagement size or initial contract term is published on any transcosmos property; the Feb 2026 brochure names only "pay-as-you-go" BPaaS/SaaS. Footprint figures also differ across its own materials — 187 sites and 43,050 workstations in the FY2026 report against 186/53,580 and 188/54,140 in two brochures. The English information-security page returns 404 trans-cosmos.co.jp, checked Aug 2026 E5E3

Computed from this provider’s anchor scores against the 29-provider field it is scored in, from the 12 public claims above: the sub-criteria where it sits furthest above the field median, and furthest below. Selection is by distance from the field, not by raw score — a 5 where most of the field scores 5 says nothing. These are scores for what a provider publishes, so a low score can mean a weak record or a quiet one, and this block cannot tell them apart.

Sub-criterion anchor scores in criterion order, with the weighted criterion total computed from them. Multiply each anchor by its sub-criterion weight over five and sum: that is the whole arithmetic, and it is why this total can be checked without taking anyone's word.

• Largest remaining evidence gap: no GDPR processor posture published at all

Every line above is meant to be checkable. If a datapoint is wrong, out of date, or missing a certificate identifier we could not find, the corrections process is open: submit the evidence and the score is re-derived through the anchors. Facts are correctable; scores follow facts.