Most IT Finance organizations are past the “AI curiosity” stage. You’ve seen the demos. You’ve attended the conferences. You may even have a slide titled “AI in IT Finance” in your strategy deck.
And yet… the gap between planning and piloting AI within IT Finance is where many teams are stuck today. Not because of a lack of ideas, but because four challenges keep getting in the way:
- Data that’s “almost” ready – but not quite
- Governance and risk concerns
- Skills and cultural readiness
- Unclear funding and ownership
In a video I discussed some of these topics with two great thought leaders. See here:
For now, let’s look at each of the challenges briefly. But more important, I want to outline what a pragmatic next step could be around each of the four topics.
1. Data: From “We Need to Fix Everything” to “Good Enough for One Use Case”
When IT Finance teams think about AI, they often start with a giant data problem.
- Multiple systems (GL, CMDB, cloud bills, HR, asset data, spreadsheets)
- Inconsistent cost models and service definitions
- No single owner for data quality across IT, finance and business
The result? AI becomes “something we’ll do after we clean up the mess” – which means it never really starts.
A better approach:
Pick one high-value, well-bounded use case - (for example: automated variance explanations, AI-assisted budgeting for a single domain, or service costing for workplace services). Use proven cost and value frameworks like Technology Business Management (TBM) to define which cost pools, towers and services are in scope for the first slide.
Define the minimal data slice needed – using a reference model instead of inventing your own. Rather than starting from a blank sheet, align your data structures with a standardized model such as Serviceware’s Serviceware’s Digital Value Model® (DVM), which combines a TBM-optimized data model with ITFM best practices for planning, forecasting, allocation and cost transparency.
Invest in making that data “fit for purpose” - not perfect. Give this first use case clear structure (services, cost pools, consumers), stable identifiers, and basic quality checks along the IT value chain. ITFM frameworks like TBM, FinOps and DVM are all designed to support exactly this kind of value-driven, incremental approach to IT cost and value management.
Bottom line: You’re not solving data for everything. You’re solving it for something that matters – and learning as you go.
2. Governance & Risk: Letting AI In Without Losing Control
The second blocker is governance. The core worry is simple: “If AI touches financial processes, who is accountable when it’s wrong?”
Here, the answer is not a brand-new AI governance tower. It’s integrating AI into the control structures you already have.
Treat AI as a new actor inside existing financial governance, not an exception outside of it. Align AI-supported processes with the same principles you already apply for financial controls (e.g., accountability, documentation, review) instead of inventing a parallel world just for AI.
Define clear guardrails: Decide where: AI can fully automate, can only recommend and humans approve, and where AI is not allowed to act at all.
Ensure your pilots include role-based access and segregation of duties, audit trails for AI-supported decisions and changes, and a straightforward way to review, challenge, and override results.
Bottom line: The goal is not “trust the model blindly,” but to give AI a clearly defined role in a controlled environment – one that strengthens your governance.
3. Skills & Culture: Preparing the Team, Not Just the Tech
You can have great data and solid governance – and still fail if your staff members don’t come along.
In many IT Finance teams (and other teams faced with AI), the pattern is familiar: high curiosity about AI, equally high concern about being replaced or losing control, and training that is too generic (‘What is AI?’) and not concrete enough for the teams daily work. Here are the things you must do:
Shift from abstract discussions to role-specific scenarios. The point here is to show what AI can do in the team members’ job. Examples are “How AI helps you prepare a budget narrative,” “How AI assists with monthly variance explanations,” “How AI flags anomalies in cloud spend”.
Change comes from teams so build mixed teams for pilots. While the hero might exist, these people should not work alone. Combine team members from IT Finance, IT, and data/AI – so the ability is shared and nobody feels isolated.
Celebrate, celebrate, celebrate …small, credible wins. Highlight the outcomes such as hours saved, fewer manual reconciliations, faster forecast cycles, better conversations with business stakeholders and more. This builds confidence and momentum.
Bottom line: Your team’s culture will shift when people experience AI as a helper in their daily work – not as a threat in a presentation. Leadership here means to make AI feel like a trusted co-pilot, while team members stay firmly in the driver’s seat.
4. Funding: Positioning AI in IT Finance as a Strategic Capability
And, finally, there’s the money question: “Who pays for AI in IT Finance – and how do we justify it?” Here, framing makes a significant difference – or simply said AI in IT Finance should not be positioned as a tool but rather as an accelerator for better decision making. Here is what I would suggest in how to position it.
A cost-optimization and productivity lever – specific use cases like automated reconciliation, smarter cloud cost management, or streamlined budgeting are measurable. This is why you start with pilots which then can clearly show how hours saved, work was reduced, or resources and budgets have been used better. These savings can then be used for the next pilot on your journey towards AI Powered IT Finance.
How about a shared investment with the business? As better show back/chargeback, service costing and decision support improves how business teams plan and spend their money across their technology and service towers, they become direct beneficiaries of your AI adoption within IT Finance. That gives you convincing arguments for co-funding such efforts (of course you have proven that the above-mentioned situation happened in your pilot).
Bottom line: The important shift is this: AI in IT Finance is not “one more tool.” It’s an investment in how you steer technology spending and value creation (your financial GPS) – with sharper insights, higher speed, and greater credibility in every financial conversation.
Please do not miss the AI for the IT Finance Practice whitepaper, where Dr. Alexander Becker and Nancy Braun, describe the AI-Powered IT Finance further.
