Your ERP is ten years old. Users are complaining, IT is patching holes, and a vendor just pitched you a migration project worth $400,000. Before you sign anything, one question needs answering: is your ERP genuinely inadequate, or is it simply under-exploited?
The difference between the two can mean months of project work and hundreds of thousands of dollars. Yet according to integrators’ field experience, a significant share of migration projects discover during the scoping phase that the existing ERP was under-configured rather than unfit for purpose. The migration was the most expensive solution to a problem that could have been solved another way.
This guide offers a structured diagnostic method in eight steps, designed for a CIO or IT director at a mid-market company (100 to 1,000 employees) whose ERP has been in production for five years or more. The output is a 40-point scoring grid that will objectify your decision without depending on the bias of your current vendor or integrator.
Why Audit Before Deciding: Avoiding the Frustration-Driven Migration
The frustration-driven migration is the most expensive and most common scenario. A department has been complaining for two years, IT has accumulated support tickets, and leadership calls for change without a rigorous diagnostic ever having been conducted.
What an audit reveals that users do not say spontaneously:
- Most Excel workarounds signal a training or configuration issue, not an ERP limitation.
- Slowdowns are often caused by ageing infrastructure or an unoptimised database, not the software itself.
- Purchased modules that sit unused represent untapped potential before any replacement is considered.
The three-year rule. An ERP less than three years old always warrants an in-depth audit before any replacement project. The return on investment of an ERP implementation is rarely measurable before the third financial year.
The risk of outsourced diagnostics. Having this audit conducted by your current integrator or a competing vendor creates a structural bias. The audit presented here is designed to be run internally, with three to five person-days distributed between IT and business unit leaders.
Step 1: Map the Processes Your ERP Actually Covers
The objective is not to list what your ERP is capable of doing, but what it actually does in your organisation.
What the ERP is supposed to do vs. what it really does
Build a four-quadrant matrix for each functional domain (finance, procurement, sales, inventory, production, HR):
- Active and satisfactory functions: what the ERP covers well — to be preserved in any future system.
- Active but circumvented functions: the ERP officially covers the process, but users rely on spreadsheets or email instead.
- Configured but unused functions: purchased modules, configured, never adopted.
- Missing functions: genuine business needs not covered by any module.
Identifying workarounds (spreadsheets, email, shadow IT)
Every parallel spreadsheet is a diagnostic signal. During business interviews, ask systematically: “For this process, what is the last step you handle outside the ERP?” The answers map the real gaps.
Step 1 score (out of 5): 5 if coverage is complete and active across all key processes / 0 if more than half of critical processes are circumvented.
Step 2: Measure Actual Module Utilisation
An ERP costs the same whether it is used at 40% or 90% of capacity. Utilisation rates reveal the real value you are extracting from it.
Purchased but unused modules (dead cost). List every module or option active in your licence agreement. For each, estimate actual utilisation: the number of monthly transactions or active logins divided by the theoretical expected volume. A module running below 20% utilisation after two years of deployment is a dead cost.
Active users vs. licences paid. Compare the number of named or concurrent licences against the number of users who logged in at least once in the past 90 days. A licence utilisation rate below 70% justifies renegotiating your contract before any migration decision.
Step 2 score (out of 5): 5 if module utilisation exceeds 75% and more than 80% of licences are active / 0 if more than one-third of modules are unused.
Step 3: Assess Technical Debt (Version, Customisations, Integrations)
Version level and vendor end-of-support date
Check the installed version and compare it to the latest available release. An ERP two major versions behind has accumulated debt that makes an upgrade as expensive as a full migration. Consult your vendor’s roadmap for the end-of-support date of your current version: if it is less than 18 months away, that is a strong warning signal.
Number and criticality of custom developments
Ask your integrator or internal team for the count of active custom developments (excluding standard reports). A volume exceeding 20% of the standard scope signals drift that complicates every version upgrade and adds weight to any future migration project.
Existing integrations (ERP to WMS, CRM, BI, e-commerce)
Map all active interfaces. For each, record: the exchange format (REST API, flat file, EDI, manual copy-paste), the frequency, the volume, and the observed error rate. Manual interfaces are priority candidates for automation in any future system.
Step 3 score (out of 5): 5 if the version is current, custom developments are under 10% of scope, and all interfaces are automated / 0 if the version is at end-of-support, custom developments exceed 30%, and several critical interfaces remain manual.
Step 4: Qualify Master Data Quality
Data migration is the leading cause of schedule overruns in ERP projects, according to the 2024 Panorama Consulting Group ERP Report. A data diagnostic conducted before the migration decision prevents discovering this problem mid-project, when it is too late to adjust the budget.
Method: completeness score across 5 key entities
| Entity | Indicator | Acceptable threshold |
|---|---|---|
| Customers | Duplicate rate | Below 3% |
| Suppliers | Mandatory field completeness | Above 90% |
| Items / Products | Data current (less than 12 months old) | Above 85% |
| Chart of accounts | Consistency with closed fiscal years | 0 discrepancies |
| Cost centres | Active vs. mis-archived | Below 5% anomalies |
A duplicate rate above 10% in customer records means you would be migrating chaos into a brand-new system. This diagnostic takes half a day with an Excel export and a few simple queries.
Step 4 score (out of 5): 5 if all entities are within thresholds / 0 if two or more entities show critical anomalies.
Step 5: Collect Business Dissatisfaction by Department
An unpopular ERP is not necessarily an inadequate one. This step distinguishes frustrations caused by poor configuration or insufficient training from genuine functional blockers.
Method: 5 structured interviews of 45 minutes each
Target five representative stakeholders: CFO (or finance director), head of logistics or supply chain, sales director, HR director, and head of production (if applicable). Ask each the same three questions:
- For your five most critical processes, how would you rate the current ERP coverage (1 to 5)?
- How many hours per week do you estimate spending outside the ERP to compensate for its gaps?
- What changes would lead you to recommend this ERP to a peer at another company?
Impact-by-frequency matrix. Compile the friction points in a matrix: vertical axis (business impact), horizontal axis (frequency of occurrence). Friction points in the top-right quadrant (high impact, high frequency) are the only ones that justify a migration. Everything else falls under training or configuration.
Step 5 score (out of 5): 5 if all identified friction points are in the “low impact” or “low frequency” quadrants / 0 if several critical friction points are identified as independent of configuration.
Step 6: Calculate the Real Cost of Your Current ERP (Full TCO)
Before comparing the cost of a new system, you need to know the real cost of what you have. ERP total cost of ownership includes line items that are systematically under-estimated in commercial comparisons.
| Cost item | Frequency of omission in audits |
|---|---|
| Licences + annual maintenance | Rarely forgotten |
| On-premise infrastructure (servers, backups, OS) | Sometimes |
| IT headcount dedicated to ERP maintenance | Often |
| Annual cost of custom development maintenance (TMA) | Often |
| Cost of workarounds (hours lost to re-entry) | Very often |
| Cost of regulatory non-compliance (potential penalties) | Almost always |
Estimate the full annual TCO of your current ERP. This figure will serve as the baseline for any ROI calculation on a migration.
Step 6 score (out of 5): 5 if the total TCO is under control and below industry benchmarks / 0 if TCO significantly exceeds expected costs and hidden items represent more than 40% of the total.
Step 7: Benchmark Technical Performance
Perceived performance and measured performance frequently diverge. Degraded response times can come from infrastructure, an unoptimised database, or the network — not from the software itself.
Response times on critical queries. Measure the display time for the ten most-used screens in your organisation (order entry, inventory lookup, invoice generation, financial dashboard). Beyond three seconds, productivity drops measurably. If slowdowns are concentrated in a single module or a single complex query, this is typically an indexing or SQL query problem — not an ERP problem.
Measured availability vs. contracted SLA. Retrieve uptime logs for the past 12 months. Compare them against the SLA in your vendor or hosting contract. A recurring gap signals a support or infrastructure issue.
Critical incidents over the past 12 months. Count the number of incidents that blocked operations for more than one hour. Analyse their distribution: if they cluster around one or two modules, that is a signal of localised technical debt. If they are diffuse, it points to more serious systemic instability.
Step 7 score (out of 5): 5 if response times are below 2 seconds, availability meets the SLA, and critical incidents are fewer than 3 over the year / 0 if several of these indicators are persistently out of threshold.
Step 8: Decide — Upgrade, Replace, or Extend?
This is the synthesis step. Based on the seven preceding steps and the scoring grid below, three scenarios are available to you.
Criteria for staying with the current vendor (upgrade)
- The current version is fewer than two major releases behind the latest available.
- Business friction points identified in Step 5 relate to configuration or training.
- TCO remains within norms and technical debt is localised.
- The vendor has a credible roadmap for the next five years.
Criteria for switching to a new ERP (replace)
- The version is at end-of-support or within 18 months of EOL (End of Life).
- More than 30% of the scope is covered by custom developments, making any version upgrade near-impossible.
- Business friction points identified in Step 5 are structural (processes not natively supported).
- Current TCO exceeds the amortised cost of a new system over five years.
The “replace and extend” option
A third path involves retaining the current ERP for the accounting and financial core while adding best-of-breed solutions for the most critical processes (dedicated WMS, specialist CRM, BI tooling, e-commerce platform). This approach is particularly relevant when the ERP is solid on finance but weak on logistics or commerce. It avoids a full migration while resolving high-impact friction points. For more on this scenario, see our article on legacy ERP decommissioning.
The 40-Point Composite Scoring Grid
Compile the scores from the eight steps to obtain your overall diagnostic.
| Step | Dimension assessed | Score (out of 5) |
|---|---|---|
| 1 | Actual process coverage | __ / 5 |
| 2 | Module utilisation rate | __ / 5 |
| 3 | Technical debt (version, custom code, integrations) | __ / 5 |
| 4 | Master data quality | __ / 5 |
| 5 | Business dissatisfaction by department | __ / 5 |
| 6 | Full TCO (visible + hidden costs) | __ / 5 |
| 7 | Measured technical performance | __ / 5 |
| 8 | Decision clarity (criteria alignment) | __ / 5 |
| Total | __ / 40 |
Score interpretation
- Score above 28/40: your ERP is still viable. A targeted optimisation project (training, configuration, data clean-up, version upgrade) will be faster and less costly than a migration.
- Score between 16 and 28/40: amber zone. Some dimensions are failing but the system remains operational. Analyse the steps scoring below 3: if two or more concern technical debt or compliance, migration becomes worth serious consideration over the next 12 to 18 months.
- Score below 16/40: urgent migration. The ERP is accumulating technical debt, high hidden costs, and structural business friction. Every month of delay increases the cost of the future migration and the operational risk.
When an upgrade is sufficient vs. when you must migrate
An upgrade (major update within the same vendor) is sufficient when Steps 1, 2, and 5 score well and problems are concentrated in Steps 3 and 7 (technical debt and performance). Conversely, migration to a new ERP is required when Steps 1 and 5 reveal structural functional gaps that an upgrade will not resolve.
This diagnostic should be conducted before any external consultation, before requesting a budget from your executive committee, and above all before inviting a vendor in for a demonstration. Once a salesperson is in the room, confirmation bias does the rest.
For the next stage of the process, two complementary resources:
- If your diagnostic points toward migration, our guide on SAP S/4HANA: Greenfield vs. Brownfield vs. Selective Data Transition will help you structure the key architectural decision.
- To select the right partner to run your project, read our ERP integrator scoring framework.
Download our ERP evaluation grid to benchmark three vendors side by side across 30 criteria and 100 points — a direct complement to the diagnostic grid presented here.
