ERP implementation is a business change programme that happens to involve software. Organisations that treat it as an IT project — delegated to a technical team, measured by configuration completed — reliably struggle at go-live, regardless of which product they selected. This guide describes the delivery shape that works and the specific pressures that Kuwaiti organisations encounter.
The phases
A typical mid-market implementation
- 1
Discovery and process design
Document current processes, agree the target process for each area, and record every point where the system will differ from how you work today. This document is the project's constitution — every later dispute resolves against it.
- 2
Configuration
Chart of accounts, item masters, customer and supplier records, approval matrices, user roles and reporting structures. Most of the enduring quality of an implementation is decided here.
- 3
Data migration
Extract, cleanse, map and load. Run it at least twice before the real cutover, and reconcile every load against source totals.
- 4
Integration
Build and test each interface, including failure behaviour. What happens when the third-party system is unavailable is part of the specification, not an afterthought.
- 5
User acceptance testing
Business users, not consultants, execute real scenarios against migrated data. If your own people cannot complete their daily work in the test environment, you are not ready.
- 6
Training
Role-based, hands-on, close enough to go-live that it is still remembered. Train on your configured system with your data, never on a generic demonstration environment.
- 7
Cutover and go-live
A rehearsed sequence with defined stop points and a documented rollback position.
- 8
Stabilisation
Two to eight weeks of elevated support while exceptions surface. Budget for it explicitly — it is not free vendor goodwill.
The decisions that determine the outcome
A handful of decisions, usually taken early and often taken casually, account for most of the variance between successful and troubled implementations.
High-leverage implementation decisions
| Decision | Why it matters |
|---|---|
| Who owns the project internally | A named business owner with authority to decide. Projects sponsored by 'the management team' collectively have no owner. |
| Fit-to-standard position | Which processes adapt to the system, and which the system must adapt to. This single decision drives cost, timeline and upgrade profile. |
| Migration scope | Full history or opening balances. Full history is defensible for some businesses and pure risk for most. |
| Phasing | Big-bang or phased. Phasing costs more in total and reduces risk substantially. For multi-branch groups it is usually the right trade. |
| Chart of accounts design | Changing it after go-live is painful and sometimes impractical. Design it for the reporting you want, not the reporting you have. |
| Approval matrix | Configure the real one, including thresholds and escalation. Simplifying it for the project guarantees rework afterwards. |
What is specific to Kuwait
- Family-owned group structures often carry approval hierarchies that reflect ownership rather than org charts. Configure what actually happens, not what the chart says.
- Arabic and English operate side by side in most organisations — warehouse and production staff frequently need Arabic while finance works in English. Test both interfaces with the people who will use them.
- KWD's three decimal places create rounding issues in systems designed around two-decimal currencies. Test discounting and reporting, not just data entry.
- Ramadan and summer periods materially reduce available business-user time. Plan UAT and training around them rather than through them.
- Multi-branch operations are the norm rather than the exception, so branch-level reporting and stock visibility should be tested early, not late.
- Regional expansion into other GCC markets is a common three-year plan. Ask how the configuration you are agreeing now will accommodate a new entity.
Internal resourcing
The most common cause of ERP schedule failure in the mid-market is not vendor capability. It is that the organisation committed internal resource it could not actually release. Business users who are meant to test and validate are still doing their day jobs, decisions queue, and the timeline slips while the vendor bills for standby time.
Go-live readiness
Do not go live until every one of these is true
- Business users have completed end-to-end scenarios against migrated data without consultant intervention
- Migrated balances reconcile to source, and the reconciliation is signed off by finance
- Every integration has been tested including its failure behaviour
- The approval matrix has been tested with real approvers, including at least one escalation
- Reports required for the first month-end exist and have been validated
- A rollback position is documented and the decision point for invoking it is agreed
- Support arrangements for the stabilisation period are confirmed in writing
- Users have been trained on the configured system, not a demonstration environment
If you are planning an implementation and want an independent view of scope and readiness, request an ERP assessment.
Book ERP Assessment