ITSM migration over budget situations follow predictable patterns. Almost all of the cost drivers are visible before the project starts. The problem is that most teams find them mid-project, by which time the cost has already grown and the options for containing it have narrowed significantly.
This article covers the eight most common reasons ANZ mid-market migrations blow the budget, what the warning signs look like before each one becomes expensive, and what to do differently next time.
Planning a migration and want an experienced view of the budget risks before committing? Book a diagnostic call and we will scope the risks and give you a realistic budget in 30 minutes.
Why ITSM Migrations Go Over Budget
The overruns that appear most consistently in ANZ mid-market ITSM migrations are not technical failures. They are planning gaps that existed from day one but were not surfaced until the project was already underway. Each one is preventable. Each one is also significantly cheaper to address before the project starts than after configuration has begun.
The planning gap in practice
Research from Kaopiz Software found that 65% of failed migration projects spent less than 20% of their total timeline on planning. For ITSM migrations specifically, the cost drivers described below are almost all planning-phase problems. The projects that go over budget are those where the planning phase was compressed to create the appearance of faster progress.
1. Scope Was Not Defined
Scope creep is the top cost driver in ITSM migrations. The project starts with incident management, service requests, and change management. Problem management gets added in week three. Asset management in week five. The knowledge base scope doubles when someone notices articles need rewriting. Each addition is individually justifiable. Together they add weeks and tens of thousands in consulting time.
Fix: Define scope before work starts. Document what is in, what is out, and what is deferred to Phase 2. Get the sponsor to sign off. Any additions after sign-off go through change control with documented cost and timeline impact.
2. Data Quality Found Mid-Migration
Duplicate contacts, tickets assigned to agents who left two years ago, a 200-category structure nobody uses well, and custom fields that do not map to the target platform are each cheap to fix before migration and expensive after. A data audit in week one costs one to two days. Finding the same problems in week six costs three to four weeks of rework.
Fix: Run a data audit in week one. Find duplicates, inactive users, old categories, and bad field values. Clean them in the source system before migration begins, not after.
3. Process Design Was Skipped
Teams want visible progress and open the new platform and start configuring. Six weeks later they find the category structure does not work, SLA policies need a full rebuild, and change management needs reconfiguration. Skipped process design consistently adds three to five weeks and AU$20,000 to AU$50,000 in rework. In practice, process design takes two weeks. Rework takes four.
Fix: Complete process design before any configuration begins. Incident workflow, service catalogue, change model, SLA framework, priority tiers. Get each signed off before the next phase starts.
4. Integration Complexity Was Underestimated
The integration list looks simple at kick-off: connect to monitoring, link to HR, set up Teams alerts. In practice, each integration involves API authentication, field mapping, error handling, and testing. The reality of how production APIs behave is different from the documentation. A two-day integration takes five. A native connector needs custom middleware because the HR system is an old version.
Fix: Map every integration in week one. Document source version, target version, connector approach, and known limits. Add a 50% time buffer to every integration estimate as standard practice.
5. Testing Was Compressed
When the project runs late, testing is the first phase cut. The reasoning is that most things are working well enough. Defects are then found by users on day one rather than by agents during UAT. Fixing problems in production costs more and damages trust in a way that takes months to repair.
Fix: Protect testing time. If the project runs late, extend the go-live date rather than compressing the testing window. A one-week delay costs significantly less than a bad go-live.
6. Wrong Partner Selected
The wrong partner adds 30 to 60% to project cost through rework and delays. The warning signs are consistent: they configure before designing, they cannot explain their week-by-week delivery plan, they have no ANZ mid-market references from the past 18 months, and their quote is significantly lower than all other options.
Fix: Evaluate partners on references, methodology, and scope clarity — not price. Call a reference from a comparable ANZ project completed in the past 18 months and ask specifically about budget adherence.
7. Passive Executive Sponsor
Decisions stall. Scope disputes linger. Budget requests sit in inboxes. Stakeholders skip workshops. Every week of project delay from governance inaction costs AU$10,000 to AU$20,000 in idle consultant time. A sponsor who attends kick-off and then disappears is one of the most reliable predictors of project overrun.
Fix: Confirm the sponsor has both authority and time before the project starts. Document weekly check-in cadence and same-week decision turnaround requirements in the project charter and get explicit sign-off.
8. Training Was an Afterthought
Low adoption after go-live is a hidden cost that rarely appears in project post-mortems. Agents create tickets incorrectly, SLA timers are missed, change management goes unused, and the efficiency improvement that justified the migration never materialises. The cost of poor adoption frequently exceeds what training would have cost. According to Prosci, projects with excellent change management are six times more likely to meet their objectives than those with poor change management.
Fix: Budget for role-based training as a line item: agents, team leads, and end users. Two to three hours per session. AU$5,000 to AU$15,000. It is consistently the highest-ROI line item in any migration budget.
The Real Cost of Going Over Budget
Direct costs are visible: extra consulting days, longer timelines, rework. The indirect costs are often larger and never appear in the project accounting.
Every extra week on a migration project is a week the IT team is not fully running the service desk. Backlogs grow. Change discipline slips. An IT Director whose AU$120,000 project costs AU$185,000 faces harder scrutiny on every subsequent budget request. The real cost is not the AU$65,000 overrun. It is the friction on every future strategic initiative. A bad go-live reinforces the perception that IT is slow and reactive, and that trust takes months to rebuild through operational consistency.
Prevention Checklist: Eight Cost Drivers and When to Address Each
| Cost Driver | Prevention | When |
|---|---|---|
| Undefined scope | Scope document signed off before configuration begins | Week 1 |
| Data quality | Data audit: duplicates, inactive users, bad categories | Week 1 |
| No process design | All process designs signed off before configuration | Weeks 1 to 2 |
| Integration underestimation | Integration map with 50% time buffer per integration | Week 1 |
| Compressed testing | Testing time protected; go-live date moves if needed | Planning phase |
| Wrong partner | Evaluate on ANZ references, methodology, and scope clarity | Before project start |
| Passive sponsor | Sponsor commitment and availability documented in charter | Before project start |
| No training budget | Training budgeted as a line item at project planning | Planning phase |
Budget Contingency: What to Plan For
Every ITSM migration budget should include contingency. For a standard ANZ mid-market migration covering 200 to 500 employees, four to six practices, and three to five integrations: 15 to 20% on the consulting component. For complex migrations with heavy customisation, poor data quality, or many integrations: 25 to 30%.
Contingency is a risk allocation, not a slush fund. Document the specific conditions under which it can be drawn: scope additions approved through change control, or problems that were genuinely not visible at project start with documented evidence. Contingency drawn for problems that were visible and ignored is a governance failure, not a risk event.
What a Well-Budgeted ITSM Migration Looks Like
The migrations that come in at or under budget share one consistent characteristic: the planning phase was treated as delivery work, not overhead. Scope was defined and signed off. Data was audited and cleaned before migration. Process designs were completed before configuration. Integrations were mapped with realistic time estimates. The sponsor was active. Training was budgeted from the start.
Village Roadshow: ServiceNow to Freshservice in six weeks, on budget
Village Roadshow, the iconic Australian entertainment company operating 37 businesses and serving 22 million customers, migrated from ServiceNow to Freshservice with thorough planning covering scope definition, data preparation, and process redesign before configuration began. The full implementation took six weeks. The outcome: 60% reduction in ITSM costs (approximately AU$500,000 per year), 25% improvement in average ticket resolution time, and 25% higher employee satisfaction scores. The migration came in on time and produced immediate measurable results because the planning work was done before the project started. Source: Freshworks customer case study.
Our ITSM Platform Migration service covers migration scoping, budget modelling, and risk identification as core components for ANZ mid-market teams. You can also read our complete ITSM migration checklist for the full preparation process that prevents the eight cost drivers described in this article, and our article on ITSM implementation mistakes for the structural decisions that most directly determine whether a migration delivers its intended outcomes.
Book a 30-minute diagnostic call. We will tell you honestly what is broken, what is not, and what to fix first.
Frequently Asked Questions
Skipping process design. Teams jump to configuration and discover three to six weeks later that the category structure does not work, SLA policies need rebuilding, and workflows need reconfiguration. The rework consistently takes longer and costs more than the design phase would have. The second most common cause is data quality problems found during migration rather than before it, which adds two to four weeks of remediation to the project timeline.
15 to 20% on the consulting component for standard migrations covering 200 to 500 employees with four to six practices and three to five integrations. 25 to 30% for complex migrations with heavy customisation, poor data quality, or many integrations. Document contingency as a risk allocation with specific draw conditions, not as a buffer for scope additions that should go through change control.
Yes, but it adds cost. A reset returns to week one planning, completes the skipped steps, and builds a new delivery plan. It typically costs 30 to 50% of the original budget on top of what has already been spent. Teams that resist the reset and continue on the failing path consistently spend more in total than those that reset promptly. The earlier the decision to reset is made, the lower the additional cost.
Ask three questions of any quote. What is specifically in scope and what is explicitly out? What is the data quality assumption, and what happens to the timeline and cost if data quality is worse than assumed? What is the integration approach per connection and how was the time estimate for each one set? Quotes that cannot answer these three questions clearly are carrying undisclosed risk that will surface as overrun during the project.
Agent and end-user training. It is consistently the most under-budgeted line item and the one whose absence produces the most visible post-go-live problems. Low adoption after go-live means agents create tickets incorrectly, SLA timers are missed, and the efficiency improvement that justified the migration never materialises. Role-based training covering agents, team leads, and end users typically costs AU$5,000 to AU$15,000. The cost of not doing it, measured in rework, low adoption, and a CSAT that does not improve, consistently exceeds that investment within the first 90 days.