How to Control Software Costs in 2026: A CIO's Practical Playbook

Discover how ARDURA Consulting SAM specialists help CIOs and CFOs cut software spend by 25-40% through disciplined Software Asset Management and contract renegotiation.

Read also: IT Budget Planning Checklist 2026 and SAM Implementation Cost and ROI Guide

The 2026 software cost crisis is real, and it is structural

In 2025, enterprise software spend grew approximately 31 percent year over year, according to combined data from Gartner and IDC, while average IT operating budgets grew only 4 percent. That gap is not a temporary anomaly tied to a single product cycle. It reflects a structural shift in how the software industry monetizes its customers, and it puts every Chief Information Officer in 2026 in the same uncomfortable position: deliver more digital capability with a budget that simply cannot keep up with vendor invoices.

Three forces are driving the squeeze. First, SaaS sprawl has exploded. The average mid-market enterprise now operates between 250 and 300 distinct SaaS applications, compared with roughly 80 in 2018, based on benchmarks published by Productiv and BetterCloud. Second, the largest vendors, including Adobe, Microsoft, Salesforce, Oracle, IBM, and ServiceNow, have all pushed through price increases between 8 and 22 percent across their major product lines during 2024 and 2025, often combined with mandatory bundling. Third, generative artificial intelligence add-ons, including Microsoft 365 Copilot, Salesforce Einstein, Adobe Firefly, and Google Workspace Gemini, are being positioned as premium tiers that customers are pressured to adopt as a matter of competitive necessity.

The result is a financial environment where doing nothing means a 15 to 25 percent annual increase in software spend, every year, with no corresponding increase in business value. This playbook explains how to break that pattern. It is written for CIOs, IT finance leaders, and heads of procurement who need to take back control in 2026.

Why software costs are spiraling

Understanding the dynamics behind the crisis is essential before any cost program can work. Four mechanics are at play in 2026.

SaaS sprawl and shadow IT. The democratization of procurement that came with cloud-delivered software has been a productivity win, but a cost disaster. Lines of business sign up for tools with corporate credit cards, often without involving IT or procurement. The Flexera 2025 State of ITAM report found that the average enterprise had 40 to 60 percent more SaaS subscriptions than the central IT inventory tracked. Each unmanaged subscription is a renewal liability, a security risk under NIST SP 800-171, and a data residency problem under the General Data Protection Regulation. Multiply that by hundreds of applications and the cost picture becomes unmanageable.

Vendor consolidation and pricing power. Over the past decade, software has consolidated dramatically. Salesforce acquired Slack and Tableau. Microsoft acquired GitHub, LinkedIn, and Activision. Adobe acquired Figma’s competitors and tightened its Creative Cloud bundles. Oracle continues to roll up niche enterprise applications. When a vendor controls three or four critical platforms in your stack, your negotiating leverage drops to near zero unless you have a credible alternative ready. Gartner has been publishing annual warnings that the top five software vendors will control more than 40 percent of enterprise software spend by 2027.

The perpetual-to-subscription shift. The classic perpetual license, paid once and amortized over five to seven years, is nearly extinct in enterprise software. Subscription pricing has converted what used to be capital expenditure into a never-ending operating expenditure. The total cost of ownership over a seven-year window for a typical enterprise application is now 2.5 to 3.5 times what it was under the old perpetual model. Subscription pricing also gives vendors annual reset opportunities to raise prices, restructure tiers, and force migrations to new editions.

Bundling pressure on AI features. The current wave of AI add-ons follows a predictable playbook. The vendor announces a flagship AI capability. It is initially positioned as optional. Then it is bundled into the next tier up. Then the previous tier loses key features. Within 18 months, what was an optional 30-dollar-per-user add-on becomes a mandatory 50-dollar-per-user inclusion in a tier upgrade. Microsoft 365 Copilot, Salesforce Einstein GPT, and ServiceNow Now Assist all follow this pattern.

The 30/30/30 framework: where the waste actually lives

Before you can cut software costs, you have to know where the waste is. After years of working with mid-market and enterprise clients on Software Asset Management programs, ARDURA Consulting has observed a consistent pattern, which we call the 30/30/30 framework. Approximately 30 percent of typical waste comes from unused licenses, another 30 percent from oversized editions, and another 30 percent from redundant SaaS applications. The remaining 10 percent comes from contractual inefficiencies such as missed termination windows and unused support credits.

Bucket one — unused licenses (shelfware). These are licenses that have been purchased and assigned but show no meaningful usage. They include accounts for former employees that were never deactivated, departments that bought capacity assuming growth that never happened, and licenses for legacy products still rolling forward on auto-renewal. Flexera consistently reports that 30 to 35 percent of all software licenses in mid-size organizations are effectively shelfware.

Bucket two — oversized editions. This is the most overlooked category. An organization pays for Adobe Creative Cloud All Apps when 70 percent of users only need Acrobat. It pays for Microsoft 365 E5 when E3 would suffice for two thirds of the workforce. It pays for Salesforce Unlimited when most users live inside three or four objects that Enterprise edition handles fine. Edition right-sizing can recover 12 to 20 percent of spend on the affected platforms without removing a single user.

Bucket three — redundant SaaS. This is the SaaS sprawl problem made concrete. The organization has Asana, Monday, Smartsheet, Trello, and Jira simultaneously, because different teams chose different tools without coordination. It pays for Zoom, Microsoft Teams, Webex, and Google Meet because no one ever consolidated. Discovering and rationalizing this redundancy is the single highest-velocity cost reduction available to most enterprises.

Measuring all three buckets requires actual telemetry, not procurement spreadsheets. That measurement is the foundation of every credible savings program.

Quick wins (Days 1 to 90)

The first 90 days of any software cost program should focus on quick wins that fund the rest of the work. Five tactics produce the bulk of these early savings.

SaaS discovery and rationalization. Begin with a complete inventory of every SaaS subscription, sourced from expense reports, single sign-on logs, and DNS traffic analysis. Tools like Productiv, BetterCloud, and Zylo accelerate this process; ServiceNow SAM Pro includes SaaS discovery modules as well. Once you have the inventory, categorize every application as keep, consolidate, or eliminate. The typical outcome is that 20 to 30 percent of discovered SaaS spend can be eliminated immediately, with another 10 to 15 percent recovered through consolidation onto preferred platforms.

License harvesting. Harvesting is the practice of reclaiming licenses from inactive or terminated users and either retiring them or reassigning them. For perpetual or true-up-based licensing, every harvested license avoids a future purchase. For subscription products, harvesting reduces the next renewal quantity. Done systematically through quarterly reconciliation tied to the Human Resources system, harvesting alone typically recovers 5 to 10 percent of software spend. The 20-step license optimization checklist provides the operational sequence.

Deactivation of dormant users. Define dormant as no meaningful login in 60 days. Every dormant account is a candidate for deactivation, with the user notified first. This works particularly well for analytics platforms, marketing automation, and developer tools where adoption is patchy. Tableau, Power BI, GitHub Enterprise, Atlassian Jira, and similar platforms commonly show 25 to 35 percent dormancy rates.

Effective License Position (ELP) discovery. An ELP reconciles what you own, what you have deployed, and what you are using, for each major vendor. The gaps between these three numbers reveal both compliance risk (deployed more than owned) and waste (owned more than used). Most organizations have never produced a true ELP for their major vendors. The first one always surfaces surprises.

Cancel before auto-renewal. Most SaaS contracts auto-renew 30 to 60 days before the term end if you do not provide written notice. Build a renewal calendar with explicit decision deadlines 90 days before each contract end. Apply a default policy: if usage is below 70 percent of contracted volume, the renewal is paused for renegotiation. This single discipline saves 3 to 5 percent of total software spend by itself.

Strategic wins (months 4 to 18)

After the quick wins, the real savings come from strategic work that takes longer to execute but produces structural change. Four strategic moves matter most.

Contract renegotiation. Every major contract should be renegotiated, not renewed. Renewal accepts the vendor’s terms with marginal adjustments. Renegotiation rebuilds the contract from your business needs. Critical levers include unit price reductions, term length flexibility, true-up versus true-down rights, audit clause limitations, transferability of licenses across entities, exit assistance, and price protection caps. Major Enterprise Agreements with Microsoft, Oracle, IBM, and SAP typically have 12 to 24 month negotiation windows; starting early is the single biggest predictor of savings. Industry benchmark: well-prepared renegotiations recover 18 to 25 percent on average for large Enterprise Agreement renewals.

Vendor consolidation. If you operate five collaboration platforms, three CRM platforms, and four observability tools, your aggregate spend is far higher than necessary, and your data is fragmented. Consolidation onto two or three strategic vendors per category produces volume discounts, integration savings, and operational efficiency. The trap is moving too fast: consolidation requires migration planning, change management, and contractual exit terms from the platforms being retired.

Edition right-sizing. Right-sizing means moving users to the smallest edition that meets their actual usage profile. Microsoft 365 right-sizing alone can save 10 to 15 percent of the Microsoft 365 bill in most enterprises. Adobe Creative Cloud right-sizing typically saves 25 to 30 percent because the All Apps versus single-app spread is so large. Salesforce edition right-sizing is harder because the editions are functionally entangled, but moving heavy users to Enterprise from Unlimited and light users to Essentials can save 15 to 20 percent.

Audit defense playbook. Every major vendor — Oracle, Microsoft, IBM, SAP, Adobe, Autodesk, Salesforce — will eventually audit your usage. According to Gartner, 68 percent of enterprises face a vendor-initiated software audit every 24 months. Treat audits as inevitable, not as exceptional events. The defense playbook includes continuous internal audits aligned with ISO 19770 (Software Asset Management standard), clean ELP documentation, defined audit response procedures, and an external SAM partner on retainer for complex audits. A mature audit defense reduces audit settlements by 60 to 80 percent. See our license audit preparation checklist for the operational detail.

For organizations early in the journey, the SAM maturity assessment framework provides a structured way to benchmark current state and prioritize the strategic moves above.

Building a SAM and FinOps practice

Tactics alone do not produce sustainable savings. To lock in the gains, you need a discipline — a permanent organizational capability that owns software cost management as a function. In 2026, that discipline is increasingly a fusion of traditional Software Asset Management and FinOps.

Governance. Establish a cross-functional steering committee with representation from IT, finance, procurement, and security. The committee owns the software cost reduction targets, reviews major contracts before signature, and approves new software acquisitions over a defined threshold. Without this governance, savings leak back through uncontrolled purchasing.

Tooling. The three leading SAM platforms are Flexera One, Snow Software (now part of Flexera since the 2024 acquisition), and ServiceNow SAM Pro. Each provides discovery, license entitlement management, and optimization analytics. For SaaS-specific management, Productiv, BetterCloud, and Zylo are the dominant choices. For FinOps overlay, Apptio, Cloudability (acquired by IBM), Vantage, and CloudHealth (acquired by VMware/Broadcom) are commonly used. Tool selection should follow process maturity, not lead it.

Metrics that matter. Track at least the following: total software spend as a percentage of revenue (industry benchmark is 4 to 6 percent for non-tech enterprises), spend per employee, percentage of spend under contract, percentage of contracts renegotiated within 90 days of expiry, license utilization rate by major vendor, shelfware percentage, and audit settlement amounts. Each metric should have a target and a trend line, reviewed monthly.

Internal versus external capability. Most mid-market enterprises (1,000 to 5,000 employees) cannot economically justify a full internal SAM team. The hybrid model works better: a small internal team for daily operations and stakeholder management, paired with external specialists for high-leverage moments such as Enterprise Agreement renewals, vendor audits, complex license model migrations, and tool implementation. ARDURA Consulting provides this external capability for clients across Poland and Europe, with senior SAM consultants who have led negotiations against every major vendor.

Negotiation tactics that work in 2026

Vendor negotiations are not symmetric. Vendors negotiate the same contracts hundreds of times per year. Customers negotiate them once every three years. Closing that asymmetry requires preparation. Four tactical principles matter most.

Never let an auto-renewal trigger. Every contract should have a calendared decision review at least 90 days before the auto-renewal date. By that point, you should know whether you intend to renew, renegotiate, or replace. Once auto-renewal triggers, your leverage drops to zero.

Benchmark before you negotiate. Pricing transparency in enterprise software is poor by design. Before negotiation, gather comparable deal data from advisory firms such as Gartner, Forrester, NPI Financial, or UpperEdge. Even directional benchmarks shift vendor pricing meaningfully. A unit price that is 30 percent above benchmark is a clear conversation starter.

Build credible multi-vendor leverage. Vendors only discount when they believe you can walk away. That belief has to be substantiated. Run a real evaluation of alternatives, including reference calls and pilot conversations, before each major renewal. Document the alternative case. Share enough of it with the incumbent to demonstrate seriousness — without bluffing, because experienced sales teams detect bluffs immediately.

Prepare your BATNA. BATNA — Best Alternative To a Negotiated Agreement — is the term from Roger Fisher and William Ury’s classic negotiation framework that defines what you do if the deal falls through. A strong BATNA includes a defined fallback plan, a timeline, and an estimated cost. With a strong BATNA, you can accept walking away. Without one, every concession the vendor demands is leverage they keep.

Common vendor tactics to counter

Sophisticated vendor sales teams have a well-worn playbook for extracting more revenue from enterprise customers. Recognizing the tactics is half the battle.

The audit threat. Some sales conversations open with subtle references to compliance gaps the vendor has noticed. This is a soft audit threat designed to push the customer into a larger purchase. Counter by separating the audit conversation from the commercial conversation, demanding a formal audit notice if applicable, and never agreeing to commercial terms under audit pressure.

Bundling pressure. Vendors push bundles because bundles obscure unit pricing and create dependency. Counter by demanding line-item pricing for every component of any bundle, then evaluating whether you actually need each component.

FUD on AI add-ons. The current wave of FUD — Fear, Uncertainty, and Doubt — focuses on AI capabilities. Counter by separating proven business value from speculative future value, piloting AI add-ons in small departments before enterprise-wide rollouts, and refusing to pay for capacity you cannot consume.

End-of-life upgrade pressure. When a vendor end-of-lifes a product or version, customers feel forced to migrate on the vendor’s timeline at the vendor’s price. Counter by demanding written end-of-life roadmaps with at least 24 months of notice, exploring third-party support options (Rimini Street, Spinnaker Support) for products like Oracle and SAP, and using end-of-life events as negotiation triggers, not surrender moments.

Conclusion: discipline beats luck

The 2026 software cost crisis is not a problem any organization will solve with a single procurement event or a one-time renegotiation. It is a structural challenge that requires permanent organizational discipline — a fusion of SAM, FinOps, procurement, and governance, anchored to measurable targets and reviewed monthly.

The good news is that the savings are large and accessible. Organizations that have built mature SAM and FinOps practices consistently reduce software spend by 25 to 40 percent within 18 months of starting, while improving compliance, reducing audit exposure, and freeing budget for genuine innovation. The bad news is that the savings only come to organizations that do the work — and the vendors are betting that most organizations will not.

If you are starting that work, ARDURA Consulting provides senior Software Asset Management consultants, audit defense specialists, and contract renegotiation advisors who work with CIOs across Europe to close the software cost gap. Whether you need a one-time Enterprise Agreement renegotiation, a structured 12-month SAM program, or ongoing FinOps overlay for software spend, our team can help you build the capability and capture the savings. Learn more about our approach on the ARDURA Consulting SAM page, explore the FinOps for software licenses 2026 guide, or contact us for a confidential discussion of your specific situation.

Software costs in 2026 will not control themselves. With the right playbook and the right partners, you can.