Customers

Market results and customer value

Over five years we have delivered strong results in precise risk prediction, lending scale, profitability and product service: institutions we serve in China have lent more than RMB 2 trillion and earned more than RMB 50 billion in net profit, with no deviation in the default-probability predictions of the products we support.

3+ products served per clientEvery client still served today80+ client specialists trained

Results across our China portfolio

Scale, breadth and depth

2 trillion+cumulative lending (China)RMB

50 billion+net profit for institutions servedRMB

Zerodeviation in default-probability predictionsAcross every product served

20+retail products consolidatedLimits from RMB 3,000 to RMB 10M

30+scenario categoriesTax · invoices · utilities · logistics · power and more

150+models co-developed and deployed

4,000+strategies optimized in production

3+products served per clientAnd every client is still served today

80+client specialists trainedRisk · operations · precision marketing · product design

Representative customers

Measured before and after

CASE 01

ICBC · Rebuilding the small-business credit portfolio

60+ small-business products consolidated into 7, with annual lending rising from RMB 240B+ to RMB 850B.

+RMB 610Bannual lending uplift

60+ → 7products; 80% less complexity

+17 ptsapproval rate 15% → 32%

−168 bpsaverage M1 delinquency

← Swipe to see the full comparison →

ICBC · Rebuilding the small-business credit portfolio
Metric	Before	With us	Improvement
Products	60+ small-business products	Consolidated into 7	80% less product complexity
Lending & balances	Cumulative RMB 700B+; RMB 240B+ per year	Cumulative RMB 1.7T+; RMB 850B per year	Annual lending up RMB 610B+
Period	Jan 2018 – Aug 2020	Aug 2020 – Aug 2022	—
Product terms	Average ticket RMB 800k; 3–12 months, up to 60 (fully guaranteed); interest-only; due diligence on every loan	Average ticket RMB 1.6M; 3–12 or 24 months, up to 60 (50% guaranteed); interest-only; instant approval and disbursement below RMB 800k, due diligence above	Ticket up RMB 800k; average rate down 110 bps; approval time down 70%; personalized due diligence, 60% faster
Performance	Approval 15%, drawdown 50%	Approval 32%, drawdown 65%	Approval +17 pts, drawdown +15 pts
Delinquency	M1 rate (masked)	M1 rate (masked)	Average M1 delinquency down 168 bps
Customer base	Settlement-account customers 100%	Settlement accounts 70%; 20+ scenario segments 30%	20+ new scenario segments
Our services	None (in-house ML model and strategy set)	Product design, whitelist lending, demand-based whitelist lending, anti-fraud, pre-approved credit, admission-rule optimization, personal master credit model, enterprise credit-bureau model, cash-flow model, drawdown model, limits and pricing, digital operations, in-loan renewal, post-loan early warning, collections, 20+ scenario risk models, limit and credit enhancement	—

CASE 02

Bank of Ningbo · A self-operated consumer loan, from zero to one

From a referral-driven loan to a product that is 80% self-operated, with annual lending up from RMB 2B to RMB 100B.

RMB 2B → 100Bannual lending

−800 bpsaverage rate 16% → 8%

+11 ptsapproval rate 10% → 21%

−570 bpsaverage M1 delinquency

← Swipe to see the full comparison →

Bank of Ningbo · A self-operated consumer loan, from zero to one
Metric	Before	With us	Improvement
Product	Zhijie Dai (direct loan)	Ninglaihua	—
Lending & balances	Cumulative RMB 4B+; RMB 2B per year; balance RMB 3.2B	Cumulative RMB 360B+; RMB 100B per year	Annual lending up RMB 98B
Period	Jan 2018 – Dec 2019	May 2020 – Sep 2024	—
Product terms	Average ticket RMB 57k; average rate 16%; 3–12 months; equal installments	Average ticket RMB 100k; average rate 8%; 3–12 or 24 months; equal installments	Ticket up RMB 43k; average rate down 800 bps
Performance	Approval 10%, drawdown 20%	Approval 21%, drawdown 35%	Approval +11 pts, drawdown +15 pts
Delinquency	M1 rate (masked)	M1 rate (masked)	Average M1 delinquency down 570 bps
Customer base	Referral traffic 100%	Referral 20%, self-operated 80%; white-collar workers, sole proprietors and micro-business owners nationwide	Self-operated share up 80%
Our services	None (in-house rule set)	Anti-fraud, pre-approved credit, admission-rule optimization, credit model, drawdown model, limits and pricing, digital operations, in-loan coaching	End-to-end rules + models + due diligence, predicting bad loans with high precision

What followed: on the strength of this product, the bank acquired Huarong Consumer Finance, with RMB 50 billion in outstanding loans.

CASE 03

Bank of Sanxiang · Deepening construction-machinery lending

Building on Gongcheng Yidai, the bank added Gongcheng Yidai Plus and Gongcheng Qidai, extending from corporate clients to individuals as well.

+RMB 2.8Bannual lending uplift

−90%approval time

−400 bpsaverage rate 15% → 11%

−600 bpsaverage M1 delinquency

← Swipe to see the full comparison →

Bank of Sanxiang · Deepening construction-machinery lending
Metric	Before	With us	Improvement
Product	Gongcheng Yidai	Gongcheng Yidai Plus + Gongcheng Qidai	—
Lending & balances	Cumulative RMB 1.1B+; RMB 0.4B+ per year	Cumulative RMB 4B+; RMB 3.2B+ per year; balance RMB 2.6B+	Annual lending up RMB 2.8B
Period	Jan 2021 – Sep 2023	Oct 2023 – Dec 2024	—
Product terms	Average ticket RMB 600k; average rate 15%; 6 or 12 months; interest-only; due diligence on every loan, plus guarantee	Average ticket RMB 900k; average rate 11%; 6 or 12 months; interest-only; instant approval and disbursement below RMB 500k, due diligence above, plus guarantee	Ticket up RMB 300k; average rate down 400 bps; approval time down 90%; personalized due diligence, 60% faster
Performance	Approval 45%, drawdown 35%	Approval 56%, drawdown 45%	Approval +11 pts, drawdown +10 pts
Delinquency	M1 rate (masked)	M1 rate (masked); first-cycle NPL (masked)	Average M1 delinquency down 600 bps
Customer base	SANY Group corporate customers	SANY Group individual and corporate customers	One new customer segment
Our services	None (in-house rules and strategy set)	Product design, whitelist lending, demand-based whitelist lending, anti-fraud, pre-approved credit, admission-rule optimization, personal master credit model, enterprise credit-bureau model, cash-flow model, drawdown model, limits and pricing, digital operations, in-loan renewal, post-loan early warning, collections, limit and credit enhancement	—

Note: absolute values of M1 delinquency and similar indicators are masked; the tables show relative improvement only.

Building client talent

We deliver products and models — and build our clients’ own capabilities

Over five years our “external brain, master and apprentice” model has trained 80+ client specialists in risk, operations, precision marketing and product design, helping clients build teams that run on their own.

We deliver products and models — and build our clients’ own capabilities
Institution	Outcomes
ICBC	20+ risk specialists trained across anti-fraud and pre-, in- and post-loan; the team independently manages risk for over RMB 1 trillion in annual lending; three risk staff received exceptional promotions to deputy division head, division head and deputy general manager for their work on this project cluster.
Bank of Ningbo	12 risk modelers trained; products designed by 2 product designers have lent RMB 30 billion; 2 digital-operations specialists drove more than RMB 70 billion in new lending for self-operated products.
Bank of Sanxiang	6 risk specialists trained across anti-fraud, pre-loan rules, master scorecards, limits and pricing, and in-loan roles; they build models and optimize rules independently, supporting risk-strategy iteration for six of the bank’s products.
Pizhou Rural Commercial Bank	5 specialists trained in marketing feature design, risk-model variable engineering and deployment, process design and digital marketing operations; they independently run digital marketing, variable development, model and strategy monitoring, and risk-strategy analysis.

Clients

The full spectrum of financial institutions, in China and abroad

China

ICBC
China Construction Bank
Bank of Ningbo
Bank of Sanxiang
Pizhou Rural Commercial Bank
China UnionPay
TCL Financial Holdings

International

ING
Wells Fargo
JPMorgan
Citibank
Barclays

Our scientists and financial experts have supported and advised 50+ leading financial institutions worldwide, delivered 100+ solutions and solved 4,000+ retail credit growth challenges.

Get started

Start with one business question. Let SAM deliver a model you can sign off.

Whether you are in banking, FMCG and marketing, pharmaceuticals or another industry, try SAM online first — or talk to our modeling experts about your use case.

Talk to us Apply to try online