Samsung Card cannot vote on merchant-fee regulation or funding rates, but it fully controls what a service contact costs and how effectively it collects; an agentic frontline that resolves card servicing in-conversation and runs policy-bounded early collections at full-book scale attacks the only cost lines still open in a structurally squeezed P&L.
Samsung Card is one of Korea's largest card issuers, an affiliate of Samsung Group, competing with bank-owned issuers while funding itself in wholesale markets rather than with deposits.
Public factKorean card issuers absorbed another round of government-mandated merchant-fee cuts effective 2025, compressing the payments-revenue side of the industry P&L.
Public factNon-bank issuers like Samsung Card carry funding-cost exposure to rate cycles that bank-owned rivals partially escape via deposits, making cost discipline structurally more urgent.
Public factKorean card companies have publicly shifted toward loan products, installment finance and data businesses as merchant fees shrink — raising the weight of collections and risk operations.
Public factSamsung Card's service volume is dominated by billing, installment-conversion and limit inquiries with strong monthly billing-cycle peaks.
Reasoned inferenceEarly-delinquency outreach likely covers only a fraction of at-risk accounts on current dialer capacity.
Seller hypothesis — validateValidate with the account team before outreach: Actual servicing and collections volumes, coverage rates and cost per contact · Samsung Group technology-affiliate dynamics (Samsung SDS role in any AI vendor decision) · PIPA and FSS constraints on automated collections outreach windows and channels · Incumbent dialer and AICC vendor contracts
Limited internal ability or appetite to build the core platform — strong candidate for the packaged solution
Evidence: Samsung Card is a financial-products company, not a platform engineering organization; while group affiliate Samsung SDS supplies enterprise IT, there is no public card-company program to build conversational-AI infrastructure, and the margin squeeze rewards proven capability bought now over any internal timeline.
Why they won't build the full stack: Every quarter of internal build is a quarter of billing-peak queue costs and under-covered delinquency buckets in a P&L already compressed by regulated fees; the specialized honorific voice and collections-governance stack is exactly what a platform amortizes across issuers and one issuer cannot justify building.
Korea's major card issuers reported net-profit declines for 2025, with the February 2025 merchant-fee cut cited as the primary cause and funding costs as a persistent burden for non-bank issuers.
GTM implication: Open with the industry P&L math: controllable operating costs are the only lever left, and per-contact economics are the largest controllable line.
Korean issuers collectively discontinued hundreds of high-benefit card products amid the profitability squeeze, generating customer-transition service volume.
GTM implication: The product-transition desk is a timely, bounded entry: absorb discontinuation call waves while proving retention economics.
| Workflow | Why it matters here | Value | Complexity | Speed | Channels |
|---|---|---|---|---|---|
Card servicing line Billing, limit and installment servicing resolved in-conversation with transactions executed; peak-day queues absorbed elastically. | Billing and installment inquiries are the structural volume of a card book; every contact costs multiples of what the regulated fee earns on the transactions being asked about. Friction today: Billing-cycle peaks melt queues monthly; IVR deflection frustrates; the app answers questions but cannot negotiate installment conversions conversationally. | VoiceApp chatKakaoTalk | |||
Early-delinquency contact program 100% early-bucket contact attempts within policy windows, offers bounded by the approved matrix, promises written back automatically. | As card-loan books grow, early-bucket effectiveness decides credit costs; full coverage in the first days of delinquency prevents roll-forward. Friction today: Dialer capacity reaches a fraction of early buckets; conversation quality varies; PIPA constrains contact windows and channels. | VoiceKakaoTalkSMS | |||
Retention and product-transition desk Product-transition questions answered consistently at scale with policy-bounded retention offers; churn on discontinuations measurably reduced. | Issuers are discontinuing high-cost card products industry-wide; each discontinuation triggers confused, churn-risk contacts. Friction today: Product-change notices generate call waves; retention offers depend on which agent answers. | KakaoTalkVoiceApp chat |
Voice & channel orchestration, telephony, conversational execution, session/state, routing, integration build. Capability coverage validated during implementation.
API access to these systems is the critical-path dependency.
Identity-bound sessions, policy-bounded actions, 100% audit logging, human approvals at defined points, in-tenant intelligence.
All figures are modelling estimates from the labeled inputs above — nothing here is customer-provided yet. The pilot's first job is replacing these assumptions with Samsung Card's measured baseline. Package price covers implementation only; recurring usage billed separately.
Why this package for Samsung Card: Card servicing plus early collections are two high-volume workflows over shared card-platform integrations — classic Scale scope for an issuer whose economics reward automation faster than almost any Korean financial institution.
Packages cover implementation and integration only. Recurring costs are billed separately: Tilicho Labs platform usage (~$0.15/call-min indicative, usage only), Google Cloud consumption, telephony/carrier charges, managed operations, and support & optimization. No package includes unlimited usage.
“Fee cuts and funding costs are set elsewhere; service and collections economics are set here. An agentic frontline is the largest controllable-cost lever remaining in the card P&L.”
“A governed agent layer above the card platform delivers servicing and collections from one architecture — deployed in quarters, benchmarkable, without diverting engineering from the data-business roadmap.”
“Billing-cycle peaks force staffing for the worst day of the month; elastic capacity absorbs peaks at usage pricing and ends the monthly queue melt.”
“Early-bucket coverage is your credit-cost lever; reaching 100% of new delinquencies with policy-bounded conversations in the first week is what dialer capacity has never allowed.”
“Collections conversations are conduct-risk concentrate; script governance, contact-window enforcement and 100% logging give evidence quality the FSS's consumer-protection focus demands.”
“With merchant fees regulated down and funding costs up, the pilot's cost-per-contact and roll-rate metrics land directly on the two numbers your earnings calls keep explaining.”
Written from this account's own research — the strategic signal, the capability gap, the entry workflow, and the modelled economics — not a mail-merge template. Pick the moment and the persona, edit anything, then copy or open in your mail client.
Customer-safe by construction: drafts are composed only from customer-facing fields. Account tier, build-vs-buy classification, priority score, internal routing, and partner-commercial detail are not inputs to the composer, so they cannot appear in a draft. Money figures are always framed as modelled from the customer's own volumes. Read before sending — you own what goes out.
Korea's leading non-bank card issuer is squeezed from both sides — government-mandated merchant-fee cuts shrinking payment revenue and elevated funding costs raising its cost of money — making per-contact service economics and collections effectiveness the two levers management can actually control.
Margin-defense workshop with operations and collections leadership: baseline cost per contact and early-bucket coverage, then scope a Scale deployment starting with the servicing line at the next billing peak.
Entry: Card servicing line (billing, limit, installment and payment-date inquiries) · Scale package · 10–14 weeks to production across priority workflows. Human fallback throughout; success thresholds agreed before build.
Start the pursuitResearch-based priority-account universe assembled from public information, market scale, communication volume, and solution fit. This is NOT an authoritative list of top Google Cloud customers; existing Google Cloud relationships are noted only where publicly reported. Validate every account with the account team before outreach.