Systems Development
Promilezi builds and maintains the systems that finance and public sector teams work in every day. Our developers sit in the same practice as our auditors and accountants, which means the people designing your reporting have spent the year auditing reporting like it.
We build applications around the way your teams already work. Each engagement opens with a discovery phase: we map the current process and agree what the system has to produce before anyone writes code. Controls are identified at that stage too, not retrofitted afterwards. You then see working software in stages, which leaves room to change direction while changing direction is still cheap.
- Requirements gathering and process mapping
- Web and mobile application development
- Role based access and audit trails built in from the start
- User acceptance testing and handover documentation
Most finance teams lose month end to assembling numbers instead of interpreting them. We automate the collection and reconciliation of that data so the pack builds itself and your team reviews it. Dashboards follow your existing reporting formats, including annual financial statement and statutory templates, so nobody has to translate between what the system shows and what the auditors ask for.
- Automated management and statutory reporting packs
- Interactive dashboards for finance and operations
- Exception and variance reporting
- Scheduled distribution to management and stakeholders
Finance, payroll, billing and operational platforms usually hold overlapping versions of the same data. We build the interfaces between them so a figure is captured once and agrees everywhere it appears. Replacing a legacy system is a bigger job. There we reconcile source to target in full, keep a documented trail of what moved, and rehearse the cutover before the real one.
- Interface design between finance and operational systems
- Legacy data extraction, cleansing and validation
- Opening balance reconciliation and sign off
- Migration testing, dry runs and rollback planning
We start by watching how work actually moves through your organisation, which is rarely how the procedure manual describes it. Steps that survive only out of habit come out, and what remains gets automated. Approvals, document circulation and compliance checklists are the usual first targets, because that is where delays and control weaknesses collect.
- Business process review and redesign
- Electronic approval and authorisation workflows
- Document management with version control
- Segregation of duties enforced by the system
Not everything belongs in the cloud, and moving the wrong workload costs more than leaving it where it is. We assess what should move, then schedule the migration around your reporting calendar so it never lands in the middle of year end. Backup, access control and disaster recovery are configured to a standard your auditors will accept, and documented so you can demonstrate it.
- Cloud readiness assessment and cost modelling
- Phased migration planning and execution
- Backup, retention and disaster recovery configuration
- Access control and security hardening
Most systems fail slowly, through patches nobody applied and documentation nobody updated. Our support arrangements put response times in writing and security updates on a schedule. We also keep the documentation current, so that knowing how the system works never depends on one person being available.
- Service level agreements with defined response times
- Security patching and version upgrades
- Enhancement requests under change control
- User training and refresher sessions
We build AI into systems where it earns its place: reading documents, flagging anomalies, forecasting figures that used to be estimated by hand. Every model we deploy is documented with how it was trained and what it was tested against, because a system influencing a financial decision has to be explainable to whoever audits it later. Governance and assurance work sits on our AI Services page.
- Document extraction and classification
- Anomaly and exception detection
- Forecasting and predictive models
- Model documentation for audit and review
