Finance & Performance Management
MarsadA Financial Analytics Workspace
An illustrative analytics platform that gives the finance office one trusted number, from the executive dashboard down to the original journal line.
01
The challenge
Many finance teams build reporting from manual exports across several systems: two accounting platforms for different entities, bank statements and budget spreadsheets. Reconciliation eats days every month, and the board pack arrives after events have moved on.
The deeper problem is definitions. “Revenue” in the sales report differs from revenue in the finance report, and gross margin is calculated two ways. When a board member asks why margin dipped in Q3, there is no path from the number back to the transactions that produced it.
Without data lineage or quality controls, every figure becomes a debate instead of a starting point for decisions.
02
Proposed solution
Marsad assembles a modern data stack: ingestion from accounting and banking systems through connectors and change data capture, bank statements in MT940 and camt.053, then modeling with dbt in a fast columnar warehouse, with data-quality tests on every run.
In the middle sits a semantic layer that defines each metric once (revenue, margin, working capital) and serves dashboards, reports and exports alike. From any number you can drill progressively: entity, cost center, account, all the way to the original journal line.
An intelligence layer adds explainable cash-flow forecasts, expense anomaly detection and suggested variance commentary that an analyst reviews before it is published. Row-level security ensures each user sees only their own entities.
03
Design rationale
The balance we aim for is dense but calm. Dashboards use tabular figures for easy comparison and a visual hierarchy that leads with each role’s most important question. Western digits are used in both Arabic and English to avoid any ambiguity in a financial context.
We made a deliberate call on Arabic charts: the page layout mirrors fully to RTL, but the time axis on line charts keeps running left to right, in line with common financial convention, and it is configurable. Color uses a color-blind-safe palette, so profit and loss never depend on red and green alone.
Analysts tend to prefer dark mode for long sessions, while board packs export in a print-ready light layout in both languages. Annotations can be pinned to any data point and stay attached to that figure across periods.
04
Example architecture
An illustrative layering of the system, from the screens people use down to the infrastructure. A starting point we would validate during discovery, not a final design.
Layer 01
Sources
- ERP & accounting systems
- Core banking & bank statements (MT940 / camt.053)
- Budget & planning spreadsheets
Layer 02
Ingestion & orchestration
- Airbyte connectors
- Debezium change data capture
- Apache Airflow scheduling
Layer 03
Storage & modeling
- ClickHouse analytical warehouse
- Apache Iceberg tables on object storage
- dbt models with tests
Layer 04
Semantics & governance
- Cube semantic layer & metric definitions
- Row-level security per entity
- OpenLineage lineage
- Great Expectations data quality
Layer 05
Intelligence
- Explainable cash-flow forecasting
- Expense anomaly detection
- Analyst-reviewed variance commentary
Layer 06
Experience
- React dashboards with Apache ECharts
- Drill-down to the journal line
- Bilingual board-pack generator
05
Relevant technologies
Tools we would consider for this concept. They indicate engineering ecosystems, not partnerships or certifications.
- dbt
- Apache Airflow
- Airbyte
- Debezium
- ClickHouse
- Apache Iceberg
- Cube
- Python
- Great Expectations
- OpenLineage
- React
- Apache ECharts
- Kubernetes
06
Key features
Single-definition metrics
A semantic layer that guarantees “revenue” means the same thing in every dashboard and report.
Drill-down to the journal line
From the board-level figure to entity, cost center, account and journal entry in a few clicks.
Variance analysis
Actuals against budget and forecast, with suggested commentary that analysts review.
Explainable cash forecasting
Cash-flow forecasts that show their drivers, not just a number.
Data quality and lineage
Automated tests and clear lineage for every metric, from source to dashboard.
Board packs
Scheduled bilingual reports in a print-ready layout.
07
Potential benefits
These are design goals the solution is built toward, not measured results.
Aims to shorten the monthly reporting cycle
A shared language of numbers between finance and other departments
Decisions grounded in figures that can be traced to source
Early detection of variances and unusual spend
Less reliance on manual spreadsheets and the errors that come with them
Related services & industry
Industry
Secure digital banking, payments and analytics platforms engineered for trust and scale.
Services involved
Reliable pipelines, governed warehouses, and executive dashboards that turn scattered data into decisions.
LLMs, agents, RAG and computer vision grounded in your data, with clear controls on quality, security and cost.
Bilingual corporate sites, online stores and web platforms engineered for speed, security and scale.
Secure, scalable cloud foundations, automated delivery pipelines and full observability on AWS, Azure and Google Cloud.
Research-led product design, from user journeys to production-ready design systems for complex enterprise interfaces.
This is an illustrative solution created by AMA GROUP to explain a possible approach. It is not a client engagement: the interface uses sample data, and no client, logo or result is implied.
Have a similar challenge?
Share the problem you are facing. We will come back with a clear approach, scope and next steps.