The ROI Challenge#
ERP ROI modelling is notoriously difficult. Benefits are often intangible, time-to-value is uncertain, and the baseline for comparison is poorly defined. Yet organisations must make investment decisions with the best available information.
The Framework Components#
1. Define the Baseline#
Before calculating benefits, you must understand current state costs:
Direct costs: Existing system licences, maintenance, support staff.
Indirect costs: Workarounds, manual processes, error correction.
Opportunity costs: What can't you do because of system limitations?
2. Identify Benefit Categories#
Cost reduction: Staff reduction, process efficiency, error reduction.
Revenue enablement: Faster order processing, better customer experience, new capabilities.
Risk reduction: Compliance, audit, disaster recovery.
Strategic enablement: M&A integration, geographic expansion, business model change.
3. Quantify Benefits#
For each benefit, estimate: - Annual value at full realisation - Time to full realisation - Probability of achieving the benefit
4. Model Investment Costs#
Use the TCO framework from our cost modelling articles.
5. Calculate ROI#
ROI = (Benefits - Costs) / Costs × 100
Payback Period = Investment / Annual Benefits
Benefit Quantification Challenges#
Staff Reduction#
The most commonly claimed benefit is also the most commonly overstated.
Reality: ERP implementations rarely result in immediate staff reduction. Work changes but often increases during transition.
Better approach: Model productivity improvement rather than headcount reduction.
Process Efficiency#
Claimed: Processes will be X% faster.
Reality: Process efficiency gains are often offset by process standardisation that removes useful flexibility.
Better approach: Model specific process improvements with conservative estimates.
Inventory Reduction#
Claimed: Better visibility will reduce inventory by X%.
Reality: Inventory levels are determined by many factors beyond ERP.
Better approach: Model specific inventory optimisation initiatives separately from ERP.
Time-to-Value Assumptions#
Most ROI models assume benefits begin at go-live. This is rarely accurate.
Realistic benefit realisation curve: - Go-live to Month 6: Minimal benefits, possible productivity loss - Month 6 to Month 12: Benefits begin emerging - Month 12 to Month 24: Majority of benefits realised - Month 24+: Full benefits achieved
The Assumptions That Fail#
"We will implement on time": Most implementations run 50-100% over timeline.
"We will implement on budget": Most implementations run 50-100% over budget.
"Users will adopt the system": User adoption is the most common cause of failed ROI.
"We won't need much customisation": Customisation scope typically expands during implementation.
Sensitivity Analysis#
Good ROI models include sensitivity analysis. What happens to ROI if: - Implementation takes 6 months longer? - Costs are 50% higher? - Benefits are 30% lower? - Benefit realisation is delayed by 6 months?
NZ/AU Specific Factors#
Currency: Factor in NZD/AUD exchange rate exposure for offshore vendors.
Market size: Smaller market may mean higher implementation costs.
Local support: Factor in cost of local implementation partners.
Conclusion: Model Honestly#
ERP ROI modelling is not about justifying a decision that's already been made. It's about understanding the investment and returns realistically. Conservative models that prove too pessimistic are far better than optimistic models that fail to materialise.