Every CIO walks into a board meeting with a version of the same problem: the numbers are big, the breakdown is vague, and the questions are pointed. “What does IT actually cost us?” “Why did it go up?” “What are we getting for it?”
The answers exist — buried across General Ledger exports, cloud billing statements, HR data, and vendor contracts. But assembling them into a defensible cost model takes weeks of manual classification, spreadsheet gymnastics, and cross-functional reconciliation. By the time the model is ready, the quarter is over and the numbers are stale.
The GL-to-cost-category gap
A typical mid-market enterprise has 500 to 5,000 GL account lines that touch IT. Each line has an account code, a description, a vendor name, and a cost center label. The challenge is mapping each of those lines to an IT Cost Framework — Infrastructure, Application, End User, Security, Management, People, Shared Services — in a way that is consistent, auditable, and defensible.
Most organisations do this manually. An IT finance analyst opens the GL export in Excel, reads each row, and assigns a category based on experience. The logic lives in their head. When they leave, the logic leaves with them. When the next quarter's GL arrives, the same work starts from scratch.
This is not a technology problem. It is a governance problem. And it becomes a board-level problem the moment someone asks “why does Infrastructure cost 40% of our IT budget?” and no one can trace the number back to the GL.
Why spreadsheets fail at scale
Spreadsheets are flexible. That is both their strength and their failure mode for cost intelligence work. Three problems compound over time:
Consistency. Different analysts classify the same vendor differently. “Microsoft Azure” might appear as Cloud in one quarter and Application in the next, depending on who does the work.
Auditability. There is no audit trail. No record of why a row was classified a certain way. No confidence score. No method indicator. When a classification is questioned, the only answer is “that's how we've always done it.”
Speed. Manual classification takes weeks. By the time the model is assembled, the opportunity to act on it has passed. Strategic IT decisions are made on intuition instead of data.
What “cost intelligence” actually means
True IT cost intelligence is not just knowing the total. It is three things:
Classification. Every GL line is mapped to a cost category and subcategory, with a method indicator (rule, AI, or reference) and a confidence score. The CFO can see the number. The IT finance manager can see why.
Allocation. Shared IT costs are distributed to business units using transparent, auditable drivers — headcount, revenue, cloud usage, ticket volume. BU leaders can trace every dollar to its source.
Context. Cost figures are benchmarked against industry data — Gartner, McKinsey, Deloitte, Forrester — so stakeholders know whether their spend is high, low, or typical for their segment.
A different approach
Inturol starts from the GL export your finance team already produces. No new data collection. No rip-and-replace. The platform applies a rules engine first — deterministic pattern matching for known accounts — then AI for the ambiguous rows. Every classification includes a confidence score and plain-English reasoning.
Critically, the Chart of Accounts becomes a persistent reference master. Classify an account pattern once, and every future upload reuses that classification automatically. The cost of classification drops to near zero over time.
Your team reviews, corrects, and locks the model. Corrections feed back as new rules. The system learns. The model gets more accurate with every cycle — not because the AI gets smarter, but because your reference data gets richer.
The result is an IT cost model that is defensible, auditable, and ready for the boardroom. Not in months. In weeks.