Half the waste sits in three vendors, not spread evenly across the portfolio
Shelfware is the share of paid licences delivering no business value, and it falls into three patterns: purchased and never deployed, deployed and never used, and deployed then abandoned after a pilot. The audit finds the volume, the quantification turns licence counts into recoverable dollars, and the recovery moves turn quantified savings into a reduced renewal commitment. None of it works in the other order.
Prepared by Redress Compliance · August 10, 2026 · Spend advisory. Based on 60 to 80 software estates audited for shelfware, 2024 to 2025.
Executive summary
Shelfware runs at 12 to 30 percent of annual licence spend, and concentration is the useful finding.
About half of the waste sits in three vendors rather than spread evenly across the portfolio, which means an audit that starts with the top publishers by spend finds most of the money in a fraction of the effort.
Three patterns cover almost every finding: purchased and undeployed, deployed and unused, and abandoned after a pilot, and the pattern dictates which recovery move applies.
Each pattern has a root cause in process rather than in negligence. Purchased and undeployed comes from discount window over commitment, where a vendor pushes a larger commitment for a deeper headline discount.
Deployed and unused comes from provisioning without offboarding discipline, so licences assigned at onboarding are never reclaimed. Abandoned after pilot comes from pilots with undefined success criteria and contracts that auto renew.
Fix the cause or the audit repeats annually with the same findings.
The method is mechanical and reproducible, which is what makes the finding defensible.
Six steps: scope by vendor, licence type, and a trailing twelve month window; reconcile contracted counts per SKU, vendor, and cost centre; prove deployment from provisioning and identity records; pull usage telemetry over the window; get business owner sign off on each unused set.
And recommend a move.
The sign off step is what converts an analyst's spreadsheet into a decision somebody owns.
Reclaimed and renegotiated spend returned 8 to 18 percent of the budget within one renewal cycle.
Quantification has three layers, the annual cost of the shelfware, the portion actually recoverable at the next renewal, and the multi year value of not carrying it forward, and only the middle layer is negotiable this year.
The audit cycle runs about ninety days from scope to action, which means it has to start well before the renewal it is meant to inform rather than alongside it.
Where shelfware hides, publisher by publisher
| Vendor | Common shelfware hotspot | Primary recovery lever |
|---|---|---|
| Microsoft | Top tier over assigned, per user platform idle | Mix shift to a lower suite, per app licensing |
| Oracle | Unlimited agreement over deployed at certification, options enabled unused | Right size at certification, deselect unused options |
| SAP | Indirect access named users, conversion buffer | Conversion credit, contract reduction right |
| Salesforce | Top tier users without login in 90 days | Reduce at anniversary, downgrade the SKU |
| IBM | Measured capacity spread over unused virtual machines | Sub capacity rebaseline, deactivate workloads |
| Broadcom and VMware | Cores over committed, storage entitlement excess | Renewal reduction, alternative platform |
The concentration finding is the one that changes how the audit is run.
Because roughly half the waste sits in three vendors, an audit sequenced by spend rather than by convenience finds most of the recoverable money in the first fortnight, which matters when the ninety day cycle has to fit inside a renewal calendar somebody else set.
It also changes who needs to be in the room: three publisher relationships rather than thirty, each with a named business owner who can sign off on an unused set without escalation.
Sequence by spend, prove deployment before arguing usage, and get the sign off in writing, because the recovery move at renewal depends on evidence that survives being challenged by an account team.
The six step method and the evidence each step needs
- Scope. Vendor list, licence types, and a time window, typically the trailing twelve months, agreed before any data is pulled.
- Entitlement reconciliation. Contracted licence count per SKU, per vendor, per cost centre, taken from the contracts rather than from the vendor's portal.
- Deployment proof. Provisioning records, identity provider exports, and vendor admin console exports, which establish what was assigned rather than what was bought.
- Usage telemetry. Login records, API call counts, and feature usage flags across the window, which separate deployed and used from deployed and idle.
- Business owner sign off, then recommendation. A manager confirms each unused set, and the finding becomes reduce at renewal, mix shift, cancel, or redeploy. The engagement wrapper sits with spend management.
The enterprise FinOps and cost optimization brief
Shelfware recovery as one of the four workstreams, with the audit method, the quantification layers, and the recovery moves that survive a renewal.
Get the white paper →Turning quantified waste into a smaller commitment
Quantification has three layers and confusing them is how shelfware programmes lose credibility with finance. The first is the annual cost of the shelfware, which is the headline number and the least actionable, because most of it is contractually committed for the current term.
The second is the portion actually recoverable at the next renewal, which is the only number worth promising, and it depends on the contract shape rather than on the size of the waste: a licence that can be reduced at anniversary is recoverable.
One inside a multi year commitment with no reduction right is not.
The third is the multi year value of not carrying the waste forward, which is the number that justifies fixing the process rather than repeating the audit. Five recovery moves cover the findings. Reduce at anniversary, where a reduction right exists.
Mix shift, moving users from a high tier they do not exercise to a lower one that covers their actual behaviour, which is frequently larger than the seat count suggests. Sublease internally, reallocating entitlement to a team that would otherwise buy.
Cancel, where the contract permits and the business owner has signed off. And redeploy, where the capability is genuinely wanted somewhere it was never provisioned.
Match the move to the pattern rather than defaulting to cancellation, because the mix shift and redeploy routes preserve capability while removing cost, which is what makes them survivable internally. The assessment side sits in the software spend assessment.
- Percentile standing for your exact deal size and industry, from real closed transactions
- Scenario simulation before the call: test alternative terms and see the financial impact of each
- A negotiation playbook, talking points, and a two page executive brief on day one
What we saw across shelfware audits, 2024 to 2025
Across roughly 60 to 80 software estates we audited for shelfware between 2024 and 2025, unused entitlement was never a rounding error:
The range shelfware occupied across the estates audited, higher on the large per seat platforms than on the infrastructure side.
Budget recovered through reclaimed and renegotiated spend inside a single renewal cycle, which funds the programme many times over.
Three patterns recurred: shelfware running at 12 to 30 percent of annual licence spend across the estate, about half of the waste sitting in three vendors rather than spread evenly, and reclaimed and renegotiated spend returning 8 to 18 percent of the budget within one renewal cycle.
The buyer side move is to sequence the audit by spend so the concentration works for you, run the six steps mechanically so the finding survives challenge, get business owner sign off in writing, and match the recovery move to the pattern rather than defaulting to cancellation.
Then fix the process behind each pattern, or the same audit produces the same findings next year.
Your first five moves
- Sequence the audit by spend, starting with the top three publishers, because roughly half the waste concentrates there and the ninety day cycle has to fit a renewal calendar.
- Reconcile entitlement from the contracts rather than the vendor portal, per SKU, per vendor, per cost centre, because the portal shows what is provisioned rather than what was bought.
- Separate deployment proof from usage telemetry, since assigned and idle is a different finding from never assigned, and the two carry different recovery moves.
- Get business owner sign off in writing on each unused set, which is what converts an analyst's spreadsheet into a decision that survives an account team challenge.
- Fix the process behind each pattern, the discount window over commitment, the missing offboarding discipline, and the pilot with no exit criteria, or the audit repeats annually. The spend practice runs the cycle with you.
Frequently asked questions
What counts as shelfware?
The share of paid software licences that delivers no business value. Three patterns cover almost all of it: licences purchased and never deployed, licences deployed and never used, and licences deployed for a pilot that ended while the licences remained on the bill.
Together they regularly account for 12 to 30 percent of an enterprise software estate.
Where is shelfware concentrated?
About half the waste sits in three vendors rather than spread evenly across the portfolio, which is the most useful finding for planning an audit.
Sequencing by spend rather than by convenience means most of the recoverable money surfaces in the first fortnight, inside the ninety day cycle the audit has to complete in.
What does a shelfware audit actually involve?
Six mechanical steps: scope by vendor, licence type and time window; reconcile contracted entitlement per SKU, vendor and cost centre; prove deployment from provisioning and identity records; pull usage telemetry across the window; obtain business owner sign off on each unused set.
And issue a recommendation of reduce, mix shift, cancel, or redeploy.
How much is realistically recoverable?
Between 8 and 18 percent of the budget within one renewal cycle across the estates we audited. That is smaller than the headline shelfware figure, because most shelfware is contractually committed for the current term.
The recoverable portion depends on contract shape: a reduction right at anniversary makes waste recoverable, a multi year commitment without one does not.
Why does business owner sign off matter?
Because it converts an analyst's spreadsheet into a decision somebody owns. An account team will challenge a usage based finding, and a named manager confirming that a licence set is genuinely unused is what makes the reduction survivable at the renewal table.
Without it the audit produces analysis rather than savings.
What are the recovery moves?
Five. Reduce at anniversary where a reduction right exists. Mix shift, moving users from a tier they do not exercise to one matching their behaviour. Sublease internally to a team that would otherwise buy. Cancel where the contract permits and the owner has signed off.
And redeploy where the capability is wanted somewhere it was never provisioned. Match the move to the pattern.
Why do the same findings reappear each year?
Because the audit removes the waste without fixing the cause. Purchased and undeployed comes from discount window over commitment, deployed and unused from provisioning without offboarding discipline, and abandoned after pilot from undefined success criteria and auto renewal.
Address those three processes or the next audit produces an identical report.
