A finance ministry can approve a modern ERP and still find that a supplier invoice waits on three desks, a procurement commitment sits outside the system, and a month-end report arrives too late to guide a decision. The technology may work exactly as specified. The operating model does not.
That is why public sector advisory Africa must begin with how a government plans, commits, spends, approves and accounts for public funds. Public financial management reform in Africa produces stronger results when ERP implementation supports clear controls, named process owners and information that moves between institutions without manual workarounds.
As of 7 October 2026, Kenya, Rwanda and Ghana each offer practical evidence of this principle. Their programmes differ in scope and legal context, but they point to the same judgement: an ERP cannot substitute for financial discipline. It can make that discipline visible, enforceable and auditable.
Start with the public-finance problem, not the software
Before selecting modules or configuring workflows, define the decision or control that currently fails. A delayed payment may reflect a weak supplier-master process. An overspent budget may reflect commitments that officers record after procurement begins. A qualified audit finding may point to incomplete asset records rather than a reporting-tool failure.
We recommend that senior sponsors identify the five to ten moments where money, authority or data changes hands. This is where process redesign earns its place. A programme that starts with screens and user roles often preserves the same paper approvals inside a more expensive system.
Separate the institutions and systems
A ministry of finance or national treasury normally leads core public financial management policy and the government financial management information system. Revenue authorities manage tax administration. Investment agencies promote or facilitate investment. Their mandates can intersect, but they do not automatically operate one shared core financial-management platform.
This distinction matters because project teams often describe e-procurement, tax systems and an IFMIS as though they are one reform. They are not. Integration may be a strategic objective, but leaders should test each interface, data owner and control before claiming interoperability.
Set the control objective in plain language
For each redesigned process, write one control statement that an accounting officer, internal auditor and system administrator can all understand. For example: no officer may raise a purchase order until the system confirms an available budget and an approved commitment.
That statement gives the ERP team a configuration requirement and gives management a basis for testing compliance. It also exposes conflicts early. If procurement staff need to bypass a commitment check to meet an emergency need, the organisation needs a documented exception route, not an unofficial workaround.
Build the process before configuring the ERP
ERP implementation in Africa requires decisions that are organisational before they are technical. We see the strongest programmes move in a defined sequence, with senior finance and operational owners accountable at each point.
1. Map the current process and its exceptions
Follow one transaction from budget allocation to payment and reconciliation. Ask the officer who performs each step what information they receive, what they check, what they re-enter and what causes them to escalate a file.
The step teams most often skip is documenting exceptions. A normal supplier payment may follow a clean workflow, while an emergency purchase, donor-funded activity or correction to an asset register follows an informal route. If the programme ignores those routes, users will create spreadsheets outside the ERP on the first difficult day.
2. Assign process owners with authority to decide
A process map without an owner becomes a workshop output rather than an operating standard. Assign a senior officer to own budget formulation, commitment control, procure-to-pay, cash management, fixed assets and financial reporting.
The owner does not need to configure the system. They must decide what good performance looks like, approve the redesigned process and resolve disputes between finance, procurement, ICT and programme teams.
3. Configure controls around real delegations
System roles should reflect the approved delegation of authority, segregation of duties and escalation rules. Do not give a user the ability to create a supplier, approve a purchase order and release a payment merely because a small team needs flexibility.
This is where audit trails matter. A useful trail records who initiated, changed, approved and released a transaction, with the relevant date and workflow status. It helps internal audit investigate exceptions without asking staff to reconstruct events from email chains.
4. Clean the data before migration
An ERP does not repair duplicated suppliers, inactive bank details or incomplete asset records. It carries them forward at scale.
Start with data ownership. The finance team should approve chart-of-account mappings, procurement should validate supplier records, and asset custodians should confirm descriptions and locations. Record the reconciliation method before loading data, because a post-go-live mismatch can otherwise become an argument about whose spreadsheet is correct.
5. Test the full transaction, not only each module
A purchasing module can pass its own tests while payment fails because the supplier record, budget code or treasury interface does not match. Test complete scenarios from plan-to-budget through procure-to-pay and record-to-report.
Include rejected approvals, insufficient-budget messages, amended invoices and reversals. These are not edge cases. They show whether the system protects the control when normal work becomes complicated.
Take an illustrative county entity with a US$2 million annual procurement budget and three offices. Its project team budgets US$120,000 for implementation, training and data preparation, then configures approvals without first documenting emergency purchases. Within the first quarter, staff begin processing urgent orders outside the platform because no exception workflow exists, weakening the very audit trail the project funded.
The better choice would have been to spend part of the design phase on exception volumes, approval limits and retrospective review rules. The US$120,000 figure is illustrative, not a benchmark for Kenya or any other jurisdiction. Cost depends on existing systems, integrations, users and the condition of master data.
What Kenya, Rwanda and Ghana show in practice
The relevant lesson from country examples is not that one platform or law fits every administration. Each example shows how policy, process and system design must reinforce each other.
Kenya: connect the end-to-end financial cycle
Kenya’s National Treasury identifies the Integrated Financial Management Information System, IFMIS, as an Oracle-based ERP for national and county government. It supports plan-to-budget, procure-to-pay, revenue-to-cash and record-to-report processes, including interfaces with the Central Bank of Kenya.
Those process labels matter because they describe a chain, not separate digital projects. If a county improves purchase-order approval but leaves budget data unreliable, procurement staff still face a weak control environment. If finance records a payment but cannot reconcile the relevant cash movement, management lacks a complete position.
Kenya’s Public Finance Management Act, revised in 2022, places direct responsibility on accounting officers. Failure to comply with financial controls, authorising unlawful expenditure and failing to pay eligible approved bills promptly can amount to improper conduct. Disciplinary action can include revocation of an accounting-officer designation.
The practical implication is clear: accounting officers should not treat IFMIS access as an ICT matter. They need reporting that shows overdue approved bills, commitments against available budget, exceptions to approval flows and unreconciled balances. Kenya’s Treasury also planned asset-and-inventory module operationalisation and a move from cash to accrual accounting, both of which require process and data readiness before technical activation.
Rwanda: use workflow to strengthen traceability
Rwanda’s Ministry of Finance and Economic Planning, MINECOFIN, operates SmartIFMIS. The system covers planning, budgeting, projects, payments, receipts, debt and grants, accounting, assets and inventory. It also includes workflow approvals, audit trails, supplier functions and interfaces with other government systems.
That breadth creates value only when the hand-offs work. A project record should inform budgeting, commitments and reporting through agreed data standards. A supplier function should support payment controls, not become a separate list maintained by each spending unit.
Rwanda’s Public Financial Management Sector Strategic Plan for 2024 to 2029 extends the reform agenda beyond fiduciary controls. It includes fiscal decentralisation, climate finance and domestic revenue mobilisation. This is a useful reminder that an ERP programme needs to support the institution’s policy priorities, rather than only automate historical accounting practices.
Take an illustrative ministry managing a US$15 million climate-finance portfolio across eight districts. It spends US$250,000 on process redesign, data cleansing and targeted system changes, rather than directing the full amount to additional reporting screens. The team first agrees how districts code projects, certify milestones and submit supporting documents, so central finance can compare like with like.
The ministry does not claim a saving before it measures one. Instead, it sets a baseline for late reports, rejected payment requests and unmatched project balances. That is the better discipline where no country-specific audit evidence supports a broad claim about savings or fraud reduction.
Ghana: make commitment control operational
Ghana made prior Ministry of Finance commitment authorisation compulsory for specified central-government contracts from 3 April 2025. The Ministry also reported the migration of 549 MDA and MMDA spending-unit accounts into the Ghana Integrated Financial Management Information System, GIFMIS.
The figure of 549 matters because account migration affects how institutions control cash and spending authority. Yet migration alone does not prove that every commitment follows the intended route. Finance leaders should test whether procurement officers create commitments before contract execution, whether approvals follow the applicable scope, and whether the system produces an exception report that management reviews.
On 30 April 2025, Ghana issued commitment-control and expenditure-management guidelines for principal spending officers, procurement officers and internal auditors. The guidelines give reform leaders a reason to align training and workflows across these roles, rather than limiting ERP training to finance users.
Do not state that Ghana’s prior-authorisation requirement applies to every public contract. The rule concerns its stated central-government scope, and programme teams should confirm the current application before configuration or rollout. We also do not advise claiming a published monetary penalty without checking the relevant amended legislation.
Governance questions leaders should ask before go-live
A go-live date can create urgency, but it should not override control readiness. We advise executive sponsors to ask questions that expose whether the programme can operate after consultants leave.
● Can an accounting officer see commitments, approved bills and payment status without requesting a manual consolidation?
● Does the approved delegation of authority match each ERP role, including temporary delegations and emergency exceptions?
● Who owns supplier data, project codes, asset records and the chart of accounts after migration?
● Can internal audit trace a sample transaction from budget approval to bank-facing payment evidence?
● Which interfaces remain manual, and who reconciles them each month?
If the answer to two or more of these questions is unclear, delay broad rollout for the affected process. A phased deployment may protect service delivery while the programme corrects data, controls or accountabilities. The decision depends on the legal mandate, transaction volumes, existing controls and operational risk.
For enterprise leaders considering ERP implementation in Africa, the same discipline applies. Start with process ownership and control design. Then select and configure technology that supports the institution’s governance model.
Frequently Asked Questions
What is public financial management reform?
Public financial management reform improves how government institutions plan, budget, commit, spend, account for and report public resources. It combines laws, operating processes, people, controls and systems. An ERP contributes when it makes those arrangements consistent and traceable.
Is an IFMIS the same as e-procurement?
No. An IFMIS supports core financial-management processes, while e-procurement focuses on procurement activity. The systems may exchange data, but leaders should define the interface, data owner and control point rather than assume integration exists.
Who should sponsor an ERP reform programme in government?
The senior finance authority should sponsor the financial-control model, with procurement, ICT, internal audit and operational leaders accountable for their parts of the process. An ICT team can lead technical delivery, but it should not decide financial delegations or expenditure controls.
How should leaders measure ERP reform results?
Measure a small set of outcomes before and after rollout, such as the number of late approved bills, unreconciled balances, rejected payment requests, off-system commitments and audit exceptions. The right measures depend on the institution’s mandate and available baseline data.
Public financial management reform deserves the same strategic attention as fiscal policy, procurement governance and institutional accountability. Speak With Our Team through our public sector advisory Africa hub to assess where process redesign and ERP implementation can strengthen your institution’s controls.
