IT budget optimisation: reduce software spend without cutting capability.
Enterprise software budgets grow by default — annual escalations, user growth, shelfware and uncontrolled SaaS proliferation add 10–15% a year. Reversing that is not a one-time cost cut but a systematic programme built on five savings levers: shelfware elimination, tier-down, renewal renegotiation, consolidation and cloud commitment discipline.
There is no single lever that fixes an inflated software budget. The organisations that sustain 20–35% reductions attack all five levers at once — then defend the gains with governance. The biggest mistake is confusing cost cutting with optimisation: cancelling tools users still need simply moves the cost downstream.
01 Key findings
Optimisation is not cost cutting. Cutting contracts, seats and renewals delivers short-term savings but creates larger downstream costs when tools must be replaced, users work around restrictions, or deferred renewals hit penalty clauses. Genuine optimisation eliminates waste while protecting capability.
Waste is structural, not occasional. The average enterprise runs 200–400 applications; 25–35% deliver minimal value relative to cost, and 28% of SaaS spend is typically identified as wasted in a portfolio review. It accumulates because budgets grow 10–15% a year unless actively managed.
Cloud is the single largest recoverable pool. For estates that have migrated at scale, 20–30% of cloud spend is recoverable through active FinOps — on a $10M cloud bill, $2–3M a year, much of it needing no vendor negotiation at all.
Auto-renewal is where budgets quietly erode. Enterprises renew 15–25% of the SaaS portfolio on auto-pilot, including tools that would fail a value test. A renewal calendar with review triggers is the cheapest structural fix available.
Savings decay without governance. Point-in-time exercises creep back within 18–24 months. Only a standing programme — software asset management, acquisition control and a vendor cadence — converts one-off savings into managed cost performance.
02 The savings levers
A complete programme works five distinct levers, each targeting a different form of waste and demanding a different action. Most organisations pull only one or two — which is why realised savings rarely reach their potential.
| Lever | What it targets | Typical saving | Effort & lead time | Risk |
|---|---|---|---|---|
| Shelfware elimination | Unused and under-used licences (<20% active users) | 15–25% of seats | Low · 30–90 days | Low |
| Tier-down & support swap | Over-provisioned editions; third-party support for stable estates | 40–50% on support | Medium · 2–4 weeks to assess | Low–Medium |
| Renewal renegotiation | Above-market rates and compounding escalators | 10–30% at renewal | Medium · 9–12 mo lead | Low |
| Consolidation | Duplicate and overlapping tools per function | 30–50% of parallel spend | Medium–High | Medium |
| Cloud commitment discipline | Idle resources, oversized instances, uncommitted usage | 20–30% of cloud spend | Low–Medium · ongoing | Low |
Two levers — shelfware and cloud — require little or no vendor negotiation and can move within a quarter. The SaaS rationalisation and cross-vendor consolidation guides cover the harder, higher-value work.
03 Savings potential
Recoverable share differs sharply by lever. The bars below show the typical proportion of the relevant category spend that a disciplined programme recovers — not a share of the total budget.
Percentages are not additive across the whole budget. A support swap saving 45% applies only to the maintenance line; cloud discipline applies only to cloud. The programme value comes from applying each lever to its own base — which is why total budget impact lands at 20–35%, not the sum of the bars.
04 The false-economy trap
The CFO asks the CIO to take 15% out of software spend without hurting the business. The instinctive response — cancel contracts, cut seats, defer renewals — is the one that fails. Each move looks like a saving on this year's budget and becomes a larger cost on next year's.
Cutting capability is not optimisation. A cancelled tool that has to be re-bought, a restricted licence that spawns shadow-IT workarounds, a deferred renewal that triggers a penalty clause — each converts a visible saving into a hidden, larger cost. The discipline is to eliminate waste (unused licences, redundant products, idle cloud) while protecting, and sometimes expanding, the capability that generates value.
The cleanest guard against false economy is a single metric — cost-per-capability — covered under quantification below. A programme that lowers cost while holding capability steady is a genuine success; one that lowers cost by removing capability is simply deferred spend.
05 Quantifying the opportunity
Optimisation programmes need metrics that track sustainable commercial performance, not just this quarter's cut. Four measures separate real savings from cosmetic ones.
| Metric | What it measures | Target |
|---|---|---|
| Cost-per-capability | Total software cost divided by capability delivered (active users, processes, revenue) | Falling while capability holds |
| Licence utilisation | Licences actively used versus licences paid for | Above 80% across the portfolio |
| Discount vs benchmark | Achieved discount against market best-achievable on major renewals | Within 10% of best rate |
| Renewal runway | Share of large renewals initiated well ahead of expiry | 100% at 9+ months for deals >$500K |
A 1,000-seat deployment carrying 20% shelfware is a concrete example: right-sizing at renewal removes roughly $360K–540K a year with zero capability loss. Feed each upcoming renewal benchmark data and rank by the gap between current and market price.
06 Prioritisation framework
With five levers and a finite team, sequencing decides how fast value lands. Weight these four factors to order the work.
Size of the gap
Rank by the absolute dollar delta between current spend and benchmark or market rate. A 10% gap on a $5M contract beats a 40% gap on a $200K one.
Speed to capture
Front-load the no-negotiation moves — idle cloud termination, licence right-sizing — that land within 90 days and fund credibility for the harder work.
Reversibility & risk
Favour low-risk, reversible actions early. Consolidation and support swaps carry migration risk and belong after the quick, safe wins are banked.
Renewal timing
Renegotiation leverage is tied to the calendar. Begin 9–12 months out; a contract two years from renewal is a governance task, not this quarter's target.
07 Sustaining the savings
The hard part is not finding savings but keeping them. Costs creep back within 18–24 months as new SaaS is added ungoverned, cloud grows unoptimised and vendors are managed reactively. Three governance elements convert one-off cuts into managed cost performance.
| Element | What it does | Cadence |
|---|---|---|
| Software asset management | Maintains portfolio inventory, tracks usage and compliance, owns the renewal calendar | Continuous |
| Acquisition control | Reviews every new purchase above threshold against existing capability before approval | Per request · $5K–25K gate |
| Vendor relationship cadence | Benchmarks top contracts, starts renewal prep early, drives executive engagement | Quarterly for top 10 by spend |
SaaS auto-renewal clauses trigger 30–90 days before the anniversary. Any renewal above $25K should require a usage-and-value justification at least 60 days out — the single control that stops the largest recurring source of quiet budget erosion.
08 Where to start
Run cloud cost tools and pull 90-day usage. Terminate idle resources and right-size seats — 10–25% recoverable with no negotiation, funding organisational support for the rest.
Where a handful of vendors dominate, benchmark them and open renewals 9–12 months early. Mid-contract leverage is real when the vendor's alternative is losing the account.
Stand up SAM, an acquisition gate and a vendor cadence. This is what stops the 18–24-month decay and turns a one-time cut into a managed cost curve.
Find where your budget is leaking
Our advisors run structured portfolio reviews that surface 20–35% of savings potential in 4 to 6 weeks — then lead the negotiations.
The Licensing Edge
Weekly cost optimisation tactics, benchmark data and contract intelligence for enterprise IT leaders. 3,000+ subscribers.