Legal functions negotiate hard for every clause: the price escalation mechanism, the service credits, the volume commitments, the termination rights. Then the contract is signed, and WorldCC, the association that has studied commercial contracting for decades, measures what happens next. Its long-standing benchmark puts average value leakage at 9.2% of contract value, and more recent WorldCC-derived estimates run higher still. On £1bn of annual contract spend, a sustained loss of roughly £92 million, agreed to in advance, clause by carefully negotiated clause.
The mechanism is not mysterious, and the same research tradition documents it: the large majority of companies do not systematically track the obligations in their contracts after signature, and many report struggling to even locate the contracts themselves when a question arises. The expertise flows into the words. The words go into a repository. The commitments go into institutional memory, which is to say, they leave.
What a contract actually contains
Read any substantial commercial agreement as an engineer rather than a drafter and it resolves into a machine for generating future events, each with a date, a condition, and money attached:
- Obligations with deadlines: deliverables, notice periods, reporting duties, on both sides.
- Rights with windows: the price escalation invocable from a date, the benchmarking right, the audit right, each worthless if the window passes unnoticed.
- Thresholds with consequences: service levels that trigger credits, volumes that trigger rebates, delays that trigger penalties, each requiring someone to compare an actual against a commitment.
- Anniversaries with defaults: renewals that auto-roll on legacy terms, indexation that should have been applied, expiries that quietly extend.
The 9.2% is what these events cost when nobody operates them. The service credit not claimed because nobody joined the SLA data to the contract term. The escalation not invoked because the window lived in a paragraph, not a calendar. The rebate never reconciled because the volumes sat in an ERP the legal team has never queried.
Every leaked pound had a clause protecting it. The clause worked. The operation of the clause never existed.
From repository to instrument
Fixing this is unglamorous in the way profitable things usually are. The contract estate has to be converted from documents into structured commitments, and the commitments wired to the data that tests them:
- Extract the events. Every obligation, right, threshold and anniversary, with dates, parties, amounts and the clause it came from, so the portfolio becomes a queryable calendar of commitments rather than a folder of PDFs.
- Join them to operational data. Service levels against the monitoring systems, volumes against the ERP, prices against the invoices. A threshold clause is only operated when actuals flow against it continuously.
- Route each event to an owner. The escalation window opening, the SLA breach accruing toward a credit, the renewal approaching on stale terms: each becomes an alert to a named person, with the money at stake and the deadline attached, not a line in a quarterly report.
- Score the recovery. Track claimed credits, invoked rights and repriced renewals against the leakage baseline, so the programme has a number and the number compounds.
The sequencing matters for credibility: start with the highest value contracts and the event types that leak most, typically renewals and service credits, and let the recovered value fund the long tail. This is one of the few compliance-adjacent programmes that is self-funding from the first quarter, because the money is not hypothetical risk avoidance. It is margin already earned and currently donated.
Nine percent of contract value is not a rounding error. It is the best-documented recoverable loss in the enterprise.
Turning contracts into monitored commitments joined to operational data is what the Prophesee Compliance Suite's Legal module does. Measure your own leakage first. Start here.