Machine learning model
Predict what’s next
- Find patterns and trends in your data
- Anticipate risk, and understand behaviour and demand
- Gain confidence through data-driven predictions
Big Data Analytics
M‑AnalyticsKnow Your Customers Better
A big data analytics platform that collects, processes and analyses your data to find patterns, trends and insights — and gives you a Customer 360 view of every customer.
What is M-Analytics?
M-Analytics is a complete platform to collect, process, analyse data on open-source big data technology, to find patterns, trends & insights.
It optimises with a range of machine-learning algorithms, so your business can make decisions faster and more effectively, whatever the situation.
Benefits
What your company gains.
Big data turns the information you already hold into better and faster decisions.
- 01
Real-time analysis for timely decisions
Predict and analyse the causes of problems in your systems to minimise failures, and use the analysis straight away to decide faster.
- 02
A tool for analysis and business strategy
Understand customer behaviour, preferences and needs, then tune your marketing, optimise customer service and raise satisfaction.
- 03
A foundation for products and growth
Accumulated data becomes analysis you can refer to when developing products and the business.
- 04
Time and cost efficiency
Faster decisions without sacrificing quality make the business more efficient in both time and cost.
01 Real-time analysis for timely decisions
Predict and analyse the causes of problems in your systems to minimise failures, and use the analysis straight away to decide faster.
02 A tool for analysis and business strategy
Understand customer behaviour, preferences and needs, then tune your marketing, optimise customer service and raise satisfaction.
03 A foundation for products and growth
Accumulated data becomes analysis you can refer to when developing products and the business.
04 Time and cost efficiency
Faster decisions without sacrificing quality make the business more efficient in both time and cost.
Solution diagram
From raw data to insight, layer by layer.
Data from inside and outside your organisation flows through acquisition and a three-tier data platform before machine learning and AI turn it into visualisations.
Data sources
- Internal data
- Social media
- Files
M-Analytics
Acquisition
Transport layer
Data management
Tier 1 Data lake
Staging & archiving
Tier 2 Integration layer
Hybrid data model
Tier 3 Access layer
Data martsAnalytics marts
Analytics
Machine learning & AI
Visualisation
Machine learning & AI
Predict what’s next. Then decide what to do.
M-Analytics pairs a machine-learning model that sees what is coming with an AI layer that recommends and automates the response.
M-Analytics AI
Decide what to do next
- Recommend the best action and response
- Automate smart decisions in real time
- Deliver adaptive, personalised outcomes
AI powered by large language models, chosen to fit your environment.
Use cases
What you can do with M-Analytics.
Descriptive statistics
A summary of information from large data sets, as dashboards and reports.
Machine learning
- Fraud detection system
- Sentiment analysis
- Customer journey optimisation
- Recommendation engine
- Dynamic pricing
- Demand forecasting
- Customer 360 view
And more.
Customer 360 view
Understand every customer, completely.
A complete picture of each customer, built from many sources of data. In banking, that means transaction, interaction, demographic, behaviour, credit and risk, and sentiment data.
Customer
Transaction data
Behaviour data
Interaction data
Demographic data
Credit & risk data
Sentiment data
Customer
- Transaction data
- Behaviour data
- Interaction data
- Demographic data
- Credit & risk data
- Sentiment data
How it comes together
Five steps to a Customer 360 view.
- 01
Data collection & integration
Collect data from many sources, such as core banking and CRM, and integrate it into one central system.
- 02
Data enrichment
Use external sources and AI to enrich what you know, and predict customers’ future needs and behaviour.
- 03
Analysis & segmentation
Group customers by behaviour and profile to find specific needs and business opportunities.
- 04
Visualisation & dashboards
Show customer data in real time on dashboards your bank team can reach easily, for faster decisions.
- 05
Actionable insights
Use the data to offer the right solutions, raise satisfaction, and predict opportunities and risks.
Step 1 / 5
Data collection & integration
Collect data from many sources, such as core banking and CRM, and integrate it into one central system.
Step 2 / 5
Data enrichment
Use external sources and AI to enrich what you know, and predict customers’ future needs and behaviour.
Step 3 / 5
Analysis & segmentation
Group customers by behaviour and profile to find specific needs and business opportunities.
Step 4 / 5
Visualisation & dashboards
Show customer data in real time on dashboards your bank team can reach easily, for faster decisions.
Step 5 / 5
Actionable insights
Use the data to offer the right solutions, raise satisfaction, and predict opportunities and risks.
Customer 360 reports
Reports your teams can act on.
Ready-made Customer 360 reports for customer service, account managers, risk and business teams.
Profile summary report
A dashboard that gives an overview of the customer’s relationship with the bank.
Metrics
- Basic customer details
- Account balance and product holdings
- Recent transactions and spending trends
- Risk profile and credit score
Use case
A reference for customer service and account managers.
Transactional activity report
A dashboard that analyses a customer’s transaction history to see patterns and anomalies.
Metrics
- Monthly deposits, withdrawals and transfers
- Spending categories and trends
- High-value transactions
- Recurring payments (e.g. bills, instalments)
Use case
Fraud detection, spending-pattern analysis and personalised offers.
More Customer 360 reports
-
Product usage
How customers use the bank’s products and services.
-
Risk & compliance
Customer compliance with regulations and risk guidelines.
-
Customer profitability
The profitability generated by each customer.
-
Engagement & interaction
Customer interactions across every touchpoint.
-
Financial health
Each customer’s financial health, based on transaction data.
-
Churn risk
The likelihood that a customer will churn.
-
Cross-sell & upsell opportunity
Personalised opportunities to cross-sell and upsell.
-
Customer satisfaction
Measure and monitor customer satisfaction.
-
Peer comparison
A customer’s financial activity compared with their peers.
-
Portfolio summary
Aggregated data for a customer segment, such as MSMEs or students.
Built on open-source big data.
M-Analytics runs on proven open-source tools for storing, moving, processing, learning from and visualising data.
Storage
- Hadoop
- HBase
- Hive
- Cassandra
Hadoop HBase Hive Cassandra
Data ingestion & orchestration
- Apache NiFi
- Apache Sqoop
- Kafka
- Debezium
- Apache Airflow
Apache NiFi Apache Sqoop Kafka Debezium Apache Airflow
Processing
- Apache Spark
- PySpark
- Flink
- Scala
Apache Spark PySpark Flink Scala
Machine learning
- TensorFlow
- scikit-learn
- Jupyter
TensorFlow scikit-learn Jupyter
Visualisation
- Apache Superset
Apache Superset
Frequently Asked Questions
Everything you need to know about M-Analytics.
M-Analytics is a platform to collect, process and analyse data end to end, using open-source big data analytics to find patterns, trends and insights. It optimises with machine-learning algorithms so the business can decide faster and more effectively.
Internal data, social media and files. Data is brought in through an acquisition (transport) layer and managed in three tiers: a data lake for staging and archiving, an integration layer with a hybrid data model, and an access layer with data marts and analytics marts.
The machine-learning model predicts what is next: it finds patterns and trends, anticipates risk and demand, and gives confidence through data-driven predictions. The AI decides what to do next: it recommends the best action, automates smart decisions in real time and personalises the outcome.
A complete picture of each customer built from many data sources — in banking, transaction, interaction, demographic, behaviour, credit and risk, and sentiment data — shown on dashboards your teams can use to act.
Profile summary and transactional activity reports, plus product usage, risk and compliance, customer profitability, engagement and interaction, financial health, churn risk, cross-sell and upsell opportunity, customer satisfaction, peer comparison and portfolio summary reports.