Micro-loan - KES 50,000
Creditworthiness and behavioral history are the primary signals.
Intelligence Layer · BISE
Reading the stories
Africa has been writing
all along.
Caspre builds financial infrastructure for the Africa that actually exists. BISE turns everyday behavioral signals into fair, explainable, sovereign scores for institutions and individuals.
M-Pesa
Bank
Telco
SACCO
ERP
Income
Behavior
Climate
Credit
CA
Sovereign720
BISE Score
Explore
02 · Core Engine
Six perspectives.
One intelligence layer.
BISE doesn't rely on a single score. It evaluates financial behaviour from multiple perspectives, then combines the signals based on the decision being made.
06 CCSContextual question
What does it mean for this decision?
Contextual Composite Score — dynamically combines all signals based on the product being evaluated.
- PCS30%
- PBS28%
- RAS18%
- RMS12%
- PredCS12%
742
Confident
Contextual output
Current decision
- PCS30%
- PBS28%
- RAS18%
- RMS12%
- PredCS12%
The Right Score For The Right Decision.
The same person can be evaluated differently depending on the financial product. BISE dynamically adjusts the importance of each score to match the decision.
03 · Data Domains
Nine domains.
One intelligence layer.
BISE turns raw, heterogeneous data into sovereign intelligence. Every variable passes through signal extraction, normalisation and contextual weighting before it reaches a score.
Telco, Bank & Mobile Money
Bill payment consistency, wallet patterns, inflow/outflow ratios, and Chama activity.
Wallet activity index
82.4+6%
Health & Wellness
Contribution discipline, preventive care utilization, and adherence indicators.
Coverage & adherence
- NHIF / SHAActive
- Preventive careRegular
- Treatment adherence91%
Social & Digital
Network quality, marketplace reputation, and participation stability.
Trust graph
Education & Skills
Credential verification, continuous learning, and fee payment history.
Growth signals
- CredentialsVerified
- Learning rate+18%
- Digital literacyHigh
Weather & Agro-Climate
Rainfall deviation, NDVI crop health, and seasonal income alignment.
NDVI crop health
0.71+4%
Macro-Economic & Sector
GDP, inflation, sector conditions, and commodity normalization context.
County economic tier
Tier 2+2%
Consumption Patterns
Utilities, essentials vs discretionary spend, and household consistency.
Spend discipline
- UtilitiesOn time
- Rent historyConsistent
- Essential ratio0.64
Geo-Location & Movement
Hashed and aggregated mobility and location context, never raw tracking.
Location context
- Region●●●●
- Cell density●●●●
- Dwell pattern●●●●
Hashed and aggregated — never stored raw
SME & Business Performance
Revenue momentum, tax signals, supplier reliability, and growth.
Revenue trend (POS)
$18.2K+11%
03 · Value Pillars
Built for two audiences at once.
The institutions that need to make better lending decisions, and the individuals whose financial lives have been illegible for too long.
scroll to explore
Score what legacy models can't see.
40% more approvals · 30–40% lower defaults
Serve the 60–85% of African workers in the informal sector who have no formal credit footprint. BISE gives banks, MFIs, and digital lenders a behavioral data edge that adapts to every product type.
For Financial Institutions
- Score and serve the 60–85% of African workers in the informal sector who have no formal credit footprint.
- Behavioral data outperforms thin formal records. Tax-compliant SMEs show 30–40% lower default rates when scored through BISE.
- The Contextual Composite Score adapts weights to your exact product: personal loans, agri loans, SME working capital, insurance, and mortgages.
- REST API with sub-second response time, 99.9% uptime, batch portfolio refresh, webhook alerts on score changes, and mobile SDKs.
- Multi-jurisdictional compliance built in: Kenya DPA, Uganda DPPA, Tanzania PDPA, GDPR, EAC framework, and central bank oversight (CBK, BoU, BoT).
04 · The Differentiator
Built for Emerging Markets,
SMEs & Individuals.
Traditional credit models were built for economies where banking is universal, employment is formal, and data trails are clean. Emerging markets are none of these things. BISE is built for emerging markets, SMEs, and individuals as they actually are.
Built From First Principles
Six structural realities define every design decision in BISE. No legacy model accounts for all six. Most account for none.
Sovereign Intelligence
Behavioral scoring built for underserved SMEs & individuals
Six Structural Realities
No legacy model accounts for all six. Most account for none.
1
Financial Data Is Key Data: Mobile Money, Bank and More
East Africa leads the world in mobile money. The richest behavioral data trail in the region lives inside telco wallets like M-Pesa, Airtel Money, MTN MoMo, T-Kash, Equitel, Wave and Orange Money, alongside bank accounts. We score it all natively.
2
The Informal Economy Is the Majority
Sixty to eighty-five percent of African workers are informal. WhatsApp Business activity, marketplace reputation, supplier payment patterns and Chama participation are first-class signals in BISE, not afterthoughts.
3
Income Is Seasonal
Over 60% of East Africans depend on agriculture. Income follows planting, growing and harvest. Our Seasonal Income Buffer prevents agricultural workers from being penalised simply for being farmers.
4
Community Finance Is Disciplined Finance
Chama, ROSCA and SACCO participation represent some of the most disciplined financial behavior on the continent, with millions contributing weekly at near-perfect reliability. Legacy models cannot see this. BISE explicitly scores it.
5
Climate Risk Is Systemic, Not Individual
Erratic rainfall, droughts, floods and locust invasions are forces no individual controls. CHIRPS rainfall data and NDVI satellite imagery trigger Climate Shock Shields that suspend negative score movement for 90–180 days in affected regions.
6
Africa Is Young and Digital-First
Africa's median age is 19.7 years. Digital natives conduct commerce on WhatsApp, build businesses on TikTok, pay with M-Pesa and accumulate financial histories entirely in digital space. BISE captures and rewards this from the earliest possible signal.
System Differentiators
What sets us apart.
The BISE engine implements sovereign design principles that legacy credit models were never built to understand.
Native, Not Retrofitted
BISE is not a Western model with African data poured in. It is built from first principles using behavioral signals native to African economies.
Active Layer
Behavioral, Not Just Transactional
We measure habits, trajectories, recovery speed, network quality and life-stage transitions, not just what someone paid, but how, when and in what context.
Active Layer
Contextual, Not Flat
Six score derivatives, dynamically reweighted by product type. The score for a mortgage is not the score for a microloan, and that distinction matters.
Active Layer
Personalised Over Time
After six months of data, BISE applies individual-specific feature weights, learning which signals are most predictive for THIS person, not just the population average.
Active Layer
Fair by Architecture
Climate Shields, Macro Adjusters, Seasonal Buffers, First-Time Entrant Boosts and Post-Shock Recovery Credits are coded into the system. Fairness enforced by code, audited annually.
Active Layer
Explainable by Default
Every score ships with confidence intervals, top three positive drivers, three concrete improvement pathways, and all contextual adjustments applied. A score the individual can't act on is a wall.
Active Layer
05 · The Origin
About Caspre
Caspre is a behavioral intelligence company building the financial infrastructure for the Africa that actually exists, not the Africa imagined by legacy systems built for somewhere else.
We were founded on a simple observation: every African has a behavioral story. Every bill payment, every Chama contribution, every harvest sale, every clinic visit, every certificate earned, every seed bought before the rain. These are the chapters of an economic life. Traditional credit systems cannot read them. We built one that can.
Our Mission
To make 800 million underserved Africans visible to formal finance: fairly, transparently, and on terms that reflect who they actually are.
Our Vision
To be the definitive behavioral intelligence layer across Africa: a continental digital public infrastructure that every person can carry across borders.
What We Believe
- Africa is not a data desert. The data exists. It just doesn't live in the places legacy models look.
- A score is a ladder, not a wall. Every score we produce comes with a pathway to improve it, in plain language.
- Systemic shocks are not personal failures. Droughts, pandemics and elections are not character flaws. Our architecture protects individuals from forces beyond their control.
- Data belongs to the individual. We build with consent at the centre, portability by default, and revocation always available.
Assurance
Independently assessed.
Security certifications, registry listings and AI governance commitments.
SOC 2Service organization control report
ISO/IEC 27001Information security management
GDPREU data protection
Cloud Security AllianceCorporate member · Solution provider
CSA STAR Level OneSelf-assessment · STAR registry
CSA STAR for AI Level OneSelf-assessment · AI security
AI Trustworthy PledgeCloud Security Alliance · 2025
07 · Architecture
Sovereign Processing Architecture.
1
Layer 1 · Data Ingestion and Source Connectivity
OAuth 2.0 connections, rate-limited APIs, real-time streaming and daily batch updates. Privacy transformation happens at ingestion. Raw PII is never stored beyond retention windows.
Continental coverage
Kenya · Tanzania · Uganda · Rwanda · Ethiopia · Nigeria · Ghana
Strategic Reach
We are where Africa's economy actually lives.
The opportunity is generational. 800 million underserved Africans. A median age of 19.7 years. Mobile money systems that already process more transactions than PayPal globally. The infrastructure for inclusion exists. BISE is the sovereign intelligence layer that reads it.
- Underserved Africans targeted for access
- 800M+
- Live markets across East & West Africa
- 8
- Scored individuals targeted by Year 2
- 200M
- Africa's median age, digital-first, mobile-first
- 19.7