MagicOrange Wants to End the Annual IT Budget Squeeze

The orange was being squeezed again. At a conference, a chief information officer described the annual ritual: the finance chief wanted another 10% cut, with services intact. According to MagicOrange’s account, its founders heard the complaint and kept the metaphor. An IT budget apparently had to behave like a piece of fruit that never ran out of juice.

The company built around that image deals in a less entertaining substance: the evidence behind an enterprise technology bill. MagicOrange combines IT financial management, cloud financial operations and shared-services costing. Its proposition is that a business should be able to follow an expense from the original input to the department, service or product that consumes it.

That sounds tidy. Inside a large organization, it can be a political undertaking. A finance team sees accounts. An infrastructure team sees machines and workloads. A business manager sees a charge arriving from somewhere upstairs. MagicOrange’s opportunity sits in the distance between those views.

The bill is easy. Explaining it is harder.

Consider a hypothetical customer-facing application. Its cloud invoice is only part of its cost. People maintain it, software licenses support it and shared infrastructure keeps it available. Looking at the invoice alone can make the application appear cheaper than the resources it actually consumes. Allocating every shared expense by headcount can produce a different distortion.

MagicOrange brings financial and operational records together to model those relationships. Its integration offering covers systems such as SAP, Oracle, Microsoft Dynamics 365, ServiceNow and Jira, alongside cloud-provider data. It supports FOCUS, a common format for technology billing data. Imports can run at a chosen frequency, including daily, weekly or monthly: the freshness of the answer still depends on the inputs.

The result can be a bill of IT that a department can inspect. Showback displays the assigned costs; chargeback uses allocations to recover them financially. The difference matters. A manager who receives an informative dashboard and a manager whose budget absorbs a charge have different reasons to scrutinize the arithmetic.

The platform supports the Technology Business Management taxonomy, but also lets customers configure their own models. Its All-Back feature traces allocations through the model, helping users investigate how a charge came about. Forecasting, scenario analysis and reporting turn that historical explanation into a planning exercise: what would a change in demand, infrastructure or service delivery do to the numbers?

At SPAR, the trouble was in the tags

SPAR’s published customer story makes the mechanics concrete. The retailer wanted detailed cost understanding and equitable recovery from its divisions. MagicOrange describes four models: VPN actuals, VPN current recoveries, Azure cloud costs, and actuals against budget. These gave the business several views of spending within a common financial approach.

One issue was wonderfully ordinary. Azure resources needed tags identifying the country using them. Inaccurate tags interfered with allocation, so the work included identifying and correcting them. The lesson is portable: before arguing over the charge, check whether the resource belongs to the department receiving it.

The same case study reports reserved-instance coverage rising from 12% to 30%. That measures coverage, not a reduction in the whole IT budget. The practical sequence is more useful than a sweeping savings promise: establish ownership, examine consumption, then consider purchasing changes.

A separate bank example shows the administrative stakes. In a case study published in January 2026, MagicOrange says an unnamed international bank replaced 180 Excel-based models with one automated platform in 90 days. It reports calculation runs reduced to minutes and chargebacks feeding directly into the general ledger. This is a vendor-reported result for that engagement, rather than an implementation timetable every buyer should expect.

The cost-control company had a cost problem

MagicOrange has had to confront some of the same complexity it sells against. In a technical retrospective, its architecture team described starting data-platform work in 2015 with Azure services suited to the volumes anticipated then. More customers brought more data, and the team began reconsidering how ingestion, storage and analysis fitted together.

The concern was the expense and management work involved in stitching together specialist tools. After evaluating options including Snowflake, the company chose Azure Databricks. The account describes faster, cheaper data-processing tasks and a platform that combined several functions. Growing scale changed the design calculation; this was a response to operational pressure.

That decision fits a wider product habit. Explaining the Microsoft relationship, sales executive Stephen Coull said, “We chose the partner route.” MagicOrange embedded Power BI and used Microsoft data tools, concentrating its own effort on costing, allocations and profitability calculations. For buyers with those skills already in-house, familiarity can reduce one source of friction.

“We chose the partner route.”

Stephen Coull, on MagicOrange’s Microsoft strategy

Founded in 2014, according to funding coverage, MagicOrange counts David Harding, Greg Guye and Matthew Mottram among its founders. Harding is CEO and Guye leads product. The company emphasizes its leadership’s experience on the customer, integrator and software-vendor sides of enterprise IT. Its careers page describes ongoing technical training and remote and hybrid arrangements, a picture of the employer it aims to be.

Now the spreadsheet answers back

Databricks also underpins MagicOrange’s move into conversational analysis. Its AI/BI Genie integration lets users question financial datasets in natural language. In Databricks’ customer account, the underlying problem includes dependence on technical teams whenever somebody needs an unfamiliar answer. Bringing that exchange closer to the meeting can shorten the journey from question to investigation.

The partner case study highlights more than $100,000 in savings from choosing Genie over alternative generative-AI tools. That is a reported outcome for MagicOrange’s own implementation, not a customer savings guarantee or a price comparison that applies to every AI workload. The company also uses Unity Catalog for access control and Delta Sharing for sharing data.

Its current AI offering includes executive summaries, assistance with allocations, anomaly detection and recommendations about usage and cost. These are relevant tools for finance teams, provided the questions remain attached to a sound model. A fluent answer cannot settle whether headcount, storage consumed or transactions processed is the fairest allocation driver. People still own that choice.

There is a useful symmetry here. MagicOrange uses AI to help explain technology spending while expanding its pitch to include the economics of AI itself. A rise in token consumption becomes another financial event to connect to owners, activities and outcomes. The interesting unit is the business service delivered, alongside the technical resource consumed.

What the meter costs

MagicOrange sells enterprise SaaS through direct sales, channel partners and Microsoft Azure Marketplace. Its G-Cloud 14 pricing document provides a concrete view of the commercial model. ITFM fees depend on IT or shared-services spend under management, with volume discounts and minimums. FinOps fees range from 0.6% to 1.5% of cloud spend, depending on the granularity of analysis, also with minimums.

For illustration, applying those percentages to £1 million of annual cloud spend gives £6,000 to £15,000. That is arithmetic using the published schedule, not a quote. Scope, applicable minimums and negotiated terms affect an actual agreement. The commercial question is whether the work saved and decisions improved justify the cost of watching the bill.

Reported financing provides another number: £7 million in April 2025, led by Octopus Ventures, with coverage also naming 6 Degrees Capital and Peter Bauer. In June 2026, MagicOrange announced that it had been named a Leader in Forrester’s second-quarter evaluation of IT financial management software. Both developments help explain why it belongs in the enterprise ITFM conversation.

A fairer fight over the budget

IBM Apptio is a direct alternative in that conversation, with cost allocation, planning and technology-finance capabilities; IBM Cloudability addresses cloud financial management. MagicOrange’s case rests on combining ITFM, FinOps and wider enterprise costs, permitting custom models and using familiar Microsoft analytics. Those are reasons to evaluate its fit, rather than proof that every competitor lacks the same capabilities.

Its customer material spans retail, data services, insurance and healthcare, with names including SPAR, Experian and eHealth Queensland. The common thread is organizational complexity: shared services, multiple consumers and a need to explain the distribution of costs. A small team with a simple cloud account has a different problem from a bank coordinating scores of allocation models.

The approach demands participation. It becomes less useful when resource ownership is wrong, usage records arrive too late or departments refuse to agree on allocation rules. Reserved commitments also need dependable usage assumptions. A better view of spending does not, by itself, authorize shutting a service down or make an unsuitable commitment sensible.

A reader can copy a modest part of the method before buying anything: choose one disputed service, identify its direct and shared costs, agree on the consumption measure and let the receiving team inspect the calculation. Reconcile the total to finance’s records. Then revisit the rule when consumption changes. That exercise makes the next software demonstration considerably more revealing.

The orange in the origin story was asked to keep its shape while losing its contents. MagicOrange’s more useful promise is an explanation of what each squeeze would change. In a budget meeting, that gives people something better to debate than another unexplained percentage.