SaaS Optimisation Guide: Shelfware Elimination and Cost Recovery

The SaaS Waste Problem Is Structural, Not Incidental

Enterprise SaaS spending grew by 9.3 percent year-over-year in 2024, bringing average annual SaaS expenditure for large organisations to approximately $49 million. Gartner forecasts worldwide SaaS spend to reach $299 billion in 2025. Yet independent research consistently shows that between 42 and 50 percent of enterprise SaaS licences go unused — representing $127 million or more in annual wasted spend across a typical large enterprise portfolio.

These are not marginal inefficiencies. They are the product of structural dynamics that are inherent to how SaaS is procured, deployed, and managed in enterprise environments. Understanding those dynamics is the starting point for any effective optimisation programme, because tactical approaches that do not address the underlying causes will see waste return within twelve to eighteen months.

Why SaaS Shelfware Accumulates

SaaS shelfware accumulates through four primary mechanisms. The first is over-provisioning at procurement: SaaS contracts are often sized at peak headcount or future growth projections, and licences are rarely reduced when deployments fall short of expectations. A contract signed for 5,000 users at a time when only 3,500 active users are expected will carry 1,500 unused licences from day one, with no mechanism to reduce until renewal.

The second mechanism is organisational change without corresponding licence management: headcount reductions, departmental restructuring, and role changes leave licences assigned to inactive accounts or to employees who no longer need the tool. Without automated de-provisioning and regular usage audits, these licences accumulate silently for months or years.

The third mechanism is shadow IT and decentralised procurement: Gartner estimates that by 2027, 75 percent of employees will acquire, modify, or create technology outside IT oversight. When business units, project teams, or individual managers can purchase SaaS tools on corporate credit cards or through expense claims, the organisation rapidly accumulates duplicate tools serving overlapping functions — a dynamic that research shows produces an average of 7.6 duplicate SaaS subscriptions per functional category.

The fourth mechanism is vendor pricing model complexity. Modern SaaS vendors have replaced simple per-user pricing with consumption-based models, credit systems, feature tier structures, and usage limits that make it genuinely difficult to determine whether the organisation is getting value from its spend. When the commercial model is opaque, waste goes undetected.

Phase 1: Discovery — Building Complete SaaS Inventory Visibility

The foundational principle of SaaS optimisation is that you cannot manage what you cannot see. Yet most enterprises do not have an accurate, complete inventory of their SaaS subscriptions. Finance sees contract commitments but not usage. IT sees corporate-procured tools but not shadow IT. Procurement sees renewal dates but not consumption patterns. The starting point for any optimisation programme is building an integrated, single-source view of the complete SaaS estate.

Identifying SaaS Subscriptions Across Sources

A complete SaaS discovery exercise draws from multiple data sources simultaneously. The primary sources are accounts payable and corporate credit card data (which captures all payment flows to SaaS vendors, including shadow IT purchases), software licence agreements and contract repositories (which capture formally contracted commitments), SSO and identity provider logs (which show which SaaS applications employees are actually accessing), and expense management platforms (which capture individual SaaS purchases that bypass procurement).

Each source captures a different segment of the SaaS estate. AP data reveals financial commitments but not usage. SSO logs reveal active usage but not contractual scope. Contract repositories reveal formal entitlements but not shadow subscriptions. The intersection and reconciliation of these data sources produces the first genuine visibility into the full SaaS landscape.

In our advisory work, the discovery phase routinely surfaces 20 to 35 percent more SaaS tools than IT leadership believed existed in the environment. The gap between perceived and actual SaaS portfolios is one of the most consistently surprising findings in SaaS optimisation programmes — and it is the gap that makes governance so important, because tools that are not visible cannot be managed or eliminated.

In one engagement, a professional services firm with £32 million in annual technology spend undertook a complete SaaS discovery audit. Redress conducted cross-functional interviews, analysed AP, corporate card, and SSO data, and identified 187 SaaS subscriptions. Finance initially believed the organisation carried 118. After elimination of 41 redundant tools, consolidation of 23 overlapping products into 6 category winners, and renegotiation of 18 major vendor contracts at renewal, the organisation reduced annual SaaS spend to £21.6 million while improving user experience through tool consolidation. The engagement fee was less than 2% of the first-year recovery.

Usage Data Collection

Once the inventory is complete, usage data collection begins. For SaaS tools with SSO integration, login frequency and feature usage data can be extracted from IdP logs and, where vendor APIs permit, from the SaaS platform itself. For tools without SSO integration, browser extension-based monitoring, network traffic analysis, or periodic user surveys provide usage intelligence.

The critical metrics for shelfware identification are login frequency (the simplest proxy for active usage), feature utilisation breadth (whether users are accessing the core functionality they were licensed for, or only marginal features), and seat utilisation rate (the percentage of provisioned seats with meaningful recent activity, typically defined as login within the preceding 30 or 90 days depending on tool type).

A seat utilisation rate below 60 percent is the threshold we use to flag a tool for rationalisation analysis. Below 40 percent, the tool is classified as high-priority shelfware with a strong presumption toward elimination or significant seat reduction at next renewal.

Phase 2: Rationalisation — Eliminating Shelfware and Redundancy

The rationalisation phase applies the discovery findings to make structured decisions about which tools to eliminate, which to retain and resize, and which to consolidate. This phase requires cross-functional input — IT, Finance, Procurement, and business unit stakeholders — because the business consequences of tool elimination are not visible to any single function.

The Rationalisation Decision Framework

For each SaaS tool in the estate, the rationalisation analysis applies a four-question framework. First: is this tool genuinely used? Tools with seat utilisation below 40 percent should be placed on a deprecation track absent a compelling business justification for retention. Second: is this tool duplicated by another subscription in the estate? Where two or more tools serve the same function, the lower-utilisation tool is eliminated in favour of the higher-utilisation alternative unless there is a functional or contractual reason to maintain both. Third: is this tool contracted at the right size? Tools with 40 to 60 percent seat utilisation are retained but flagged for licence reduction at renewal. Fourth: is this tool generating the expected business value? Low utilisation that reflects poor adoption rather than lack of need may indicate a training or change management issue that is worth addressing before elimination is decided.

The output of the rationalisation analysis is a tiered list: immediate eliminations (tools below 40 percent utilisation with no business justification), renewals requiring licence reduction (tools at 40 to 75 percent utilisation), consolidation candidates (tools with high redundancy), and renewals requiring renegotiation (tools above 75 percent utilisation but with commercial terms that do not reflect current market pricing).

Managing Vendor Contracts During Rationalisation

The commercial execution of shelfware elimination requires careful attention to contract terms. Most SaaS contracts do not permit mid-term seat reductions — licences are committed for the contract term and cannot be reduced until renewal. This means the financial realisation of shelfware elimination is tied to renewal dates, and the optimisation programme must be planned with renewal calendars in mind.

For tools flagged for elimination, the notification timeline is critical. Most SaaS contracts require 30 to 90 days' notice before the renewal date to prevent auto-renewal. Failing to provide timely notice results in another year of commitment on a tool marked for elimination. Organisations that do not systematically manage renewal calendars consistently auto-renew tools that have been internally approved for elimination — a failure mode that is avoidable with basic calendar management but surprisingly common.

For tools flagged for licence reduction, the rationalisation data — specifically the usage metrics demonstrating the gap between contracted and active seats — is the negotiation foundation. Vendors are more likely to agree to seat reductions when the buyer can demonstrate usage data, and more likely to push back (or offer a partial reduction) when the request is made without evidence. Building the usage evidence into the renewal negotiation package is the mechanism that converts discovery data into commercial outcomes.

Phase 3: Vendor Negotiation — Recovering Value at Renewal

Renewal negotiation is the commercial execution phase of SaaS optimisation. It is also where many organisations fail to recover the full value identified during the discovery and rationalisation phases. The primary failure mode is inadequate preparation: renewal teams approach vendors with a request to reduce licences but without the market pricing data, usage evidence, and competitive alternative awareness needed to negotiate effectively.

The 120-Day Rule

Effective SaaS renewal negotiation starts 120 days before the renewal date — not 30 days. The 120-day window provides sufficient time to: compile and analyse usage data, identify the gap between contracted and active seats, benchmark current pricing against market reference data for comparable organisations, research competitive alternatives and build credible alternatives into the negotiation position, and engage the vendor in a structured commercial conversation before they present a renewal proposal.

Research from SaaS procurement specialists confirms that 83 percent of successful renewal negotiations that achieved meaningful price reductions started engagement at least 120 days before renewal. Negotiations initiated within 30 days of renewal — which is when most organisations start — consistently achieve inferior outcomes because the buyer is perceived as committed and the vendor has no incentive to negotiate aggressively.

Negotiation Tactics That Work

For tools being reduced in scope, the negotiation approach is straightforward: present the usage data, specify the new seat count, and request a proportional price reduction. Vendors may resist proportional reductions by citing minimum commitment tiers or volume discount structures — these are commercial positions that can often be negotiated when the buyer demonstrates awareness of market pricing and signals willingness to consolidate remaining spend or explore alternatives.

For tools being retained at current scope, the optimisation opportunity is pricing renegotiation. SaaS pricing has increased at 4.5 times the rate of general inflation in recent years, but buyers who established contracts two or three years ago may be paying above current market rates for their tier. Presenting benchmark data — the price comparable organisations pay for similar seat counts and functionality — creates leverage for pricing reduction even without scope changes.

Consolidation of SaaS spend provides a third negotiation lever. Organisations that have fragmented SaaS budgets across multiple business units often find that aggregating spend across the enterprise creates volume that justifies a more favourable commercial tier. Presenting consolidated spend figures — even if procurement has historically been decentralised — enables the vendor conversation to shift from a unit-price discussion to a total-relationship-value discussion.

Phase 4: Governance — Preventing Shelfware from Returning

The governance phase is the most frequently neglected element of SaaS optimisation programmes, and the reason that the identified savings often erode within two years of implementation. Shelfware is not a one-time problem to be solved and forgotten — it is a recurring consequence of how enterprise SaaS is procured and managed. Without governance structures that address the root causes, waste returns.

Centralised Procurement and Approval Governance

The most effective governance mechanism for preventing shadow IT and uncontrolled SaaS proliferation is a mandatory approval process for all new SaaS purchases above a defined threshold — typically $1,000 annually, which captures the vast majority of SaaS tools while avoiding excessive friction for trivial subscriptions. The approval process should require: identification of existing tools in the estate that serve the same or similar function, business case justification demonstrating why the existing tool is insufficient, IT security review (to ensure the new SaaS vendor meets data handling and compliance requirements), and Finance sign-off on the budget commitment.

This process does not prevent business units from adopting new SaaS tools — it creates visibility and accountability for the decision, and it forces explicit comparison against existing tools before new commitments are made. The duplicate subscription rate (7.6 tools per functional category in unmanaged environments) drops to near zero in organisations with effective centralised procurement governance.

Automated Licence De-Provisioning

The most impactful technical governance control is automated de-provisioning of SaaS licences linked to the HR offboarding and role-change process. When an employee leaves the organisation or changes roles, their SaaS licences should be automatically suspended and flagged for reassignment or termination within 24 to 48 hours. In organisations without automated de-provisioning, former employees' licences accumulate for months — sometimes years — before anyone notices they are still being paid for.

Integration between the HR system (as the authoritative source for employee status and role) and the identity provider (as the mechanism for application access control) is the technical foundation of effective de-provisioning. Most modern IdP platforms (Okta, Azure AD, Google Workspace) support automated suspension triggered by HR status changes, and the configuration effort is modest relative to the ongoing savings from immediate licence recovery.

Quarterly Usage Review Cadence

Governance is sustained by regular review, not by one-time interventions. We recommend a quarterly SaaS usage review that covers the top 20 SaaS tools by spend, tracks seat utilisation trends over the preceding 90 days, identifies any new shadow IT purchases captured through the AP monitoring process, and flags upcoming renewals in the 90 to 120 day window. The quarterly review keeps the optimisation programme active and ensures that shelfware is identified before renewal rather than after auto-renewal has committed another year of spend.

The quarterly review should be owned by a designated SaaS programme manager — a role that sits at the intersection of IT, Finance, and Procurement and has the authority to act on findings across those functions. Without clear ownership, review findings sit undone between meetings, and the governance programme becomes a reporting exercise rather than a cost management mechanism.

The Financial Case for SaaS Optimisation Investment

SaaS optimisation programmes require investment — in tooling (SaaS management platforms), in programme management (the quarterly review cadence and governance overhead), and in advisory support (for benchmark data and vendor negotiation expertise). The financial case for this investment is robust at virtually every scale of enterprise SaaS spend.

For an organisation spending $10 million annually on SaaS, a conservative optimisation assumption of 25 percent recoverable waste yields $2.5 million in annual savings. At $49 million (the large enterprise average), the same recovery rate produces $12.25 million per year. Even at a 15 percent recovery rate — achievable without aggressive negotiation or significant consolidation — the returns from SaaS optimisation dwarf the programme investment by a factor of ten to twenty.

The organisations that capture these savings are not doing anything exotic. They are applying systematic visibility, structured decision frameworks, timely renewal engagement, and consistent governance to a spending category that has grown faster than their management practices have adapted. The gap between the best and worst performers on SaaS cost management is not a technology gap — it is a process and governance gap that any organisation can close with appropriate focus and expertise.