ERP implementation projects fail at an alarming rate globally — studies suggest somewhere between 50% and 75% fail to deliver expected value. In Nigeria and across sub-Saharan Africa, the failure rate is arguably higher, compounded by connectivity challenges, power instability, limited local talent pools for specialist platforms, and organisational dynamics that standard implementation methodologies simply do not account for.
After delivering and rescuing more than 20 ERP programmes across Nigeria, Ghana, and Kenya, we have identified seven mistakes that account for the overwhelming majority of failures. None of them are technology problems.
Mistake 1: Buying Before Designing
The most common path to ERP failure begins with a vendor selection. An executive attends a conference, sees a demonstration, and champions a particular platform. By the time the technology team is involved, the commercial decision has already been made — and the platform may be fundamentally unsuited to the organisation's actual processes.
The fix: Process design must precede platform selection. Document your target operating model first. Then evaluate platforms against your documented requirements — not the other way around.
Mistake 2: Underestimating Data Migration
Data migration is consistently the most underestimated activity in ERP projects. A standard project plan might allocate six weeks to migrate 15 years of transaction history, inventory records, and customer data. The reality is almost always four to six months — with multiple cycles of extraction, transformation, validation, and re-extraction.
The specific African challenge: data quality in legacy systems on the continent is often significantly worse than in equivalent organisations in developed markets. Manual journal corrections, workaround Excel exports stored on individual laptops, and inconsistent field naming across database versions all compound the problem.
The fix: Conduct a data audit in month one of the project. Do not wait until the data migration phase. Budget 20% of your total project cost for data migration and quality remediation.
Mistake 3: Customising the Core
Modern ERP platforms — SAP S/4HANA, Oracle Fusion, Microsoft Dynamics — are designed around industry best practices. Every customisation to the core is a liability: it increases upgrade complexity, extends implementation timelines, and creates maintenance risk. Yet the first instinct of most organisations is to replicate their existing (broken) processes in the new system.
The fix: The question for every gap between the standard platform and your current process should be: "Should we change our process to fit the platform?" In most cases, the answer is yes.
Mistake 4: Ignoring Change Management Until Go-Live
Change management is often treated as a communications exercise in the final weeks before go-live — a few town hall presentations and a series of training sessions. By this point, resistance is already entrenched. Staff who have not been involved in the design process will find every possible reason why the new system does not work for them.
The fix: Embed business process owners in the implementation team from day one. Run co-design workshops. Give staff a voice in how the system is configured. They will defend a system they helped build.
Mistake 5: Underspecifying Infrastructure Requirements
An on-premise ERP deployment in Nigeria faces infrastructure challenges that a US or European implementation simply does not. Power reliability, network bandwidth between sites, and hardware procurement lead times can all extend go-live dates by months if not planned for explicitly.
The fix: Cloud-first deployments (Azure, AWS) eliminate most of these concerns. Where on-premise is mandated (often for regulatory reasons), build explicit infrastructure readiness milestones into the project plan — and validate them, do not assume.
Mistake 6: No Hypercare Plan
The go-live date is not the end of the project — it is the beginning of the most critical phase. The first 90 days after go-live are when most ERP implementations either entrench or collapse. Issues will emerge. Users will revert to old habits under pressure. Workarounds will proliferate if not addressed immediately.
The fix: Plan and resource a dedicated hypercare team for a minimum of 90 days post go-live. This team's only job is to resolve issues, coach users, and prevent the re-emergence of shadow processes.
Mistake 7: Treating Go-Live as Success
The project team disbands. The implementation partner's resources roll off. The steering committee stops meeting. And six months later, the organisation realises that the ERP is being used at 40% of its capability because nobody measured adoption, and nobody is accountable for outcomes.
The fix: Define success as measurable business outcomes, not go-live completion. Retain a small optimisation team for at least one year post-implementation to drive adoption, configure improvements, and deliver the ROI that justified the investment.
ERP implementation is one of the most consequential technology decisions an organisation can make. Getting it right requires treating it as a business transformation programme — not a software installation. The technology is the easy part.