An organization running three contracts can track them through meetings. An organization running thirty contracts across forty centers cannot — not because the work differs in kind, but because human attention is finite. At thirty contracts, whatever does not appear in a report does not exist.
Contract performance KPIs are the mechanism that turns attention from random to directed: instead of reviewing every contract equally, review goes where the data indicates it is needed.
Why Tracking Fails at Scale
Three causes appear in sequence:
Divergent standards. Each contract was entered differently with different categories, making comparison impossible despite the data existing.
No weighting. A high-value contract and a small one appear at the same size in the list, so priority disappears.
Delayed reports. A report compiled manually each quarter reveals a problem that is three months old — past the point where correcting it was possible.
The answer is not more reports but a limited set of metrics calculated automatically and standardized across every contract.
The Six Core Metrics
| Metric | How it is calculated | What it reveals |
| Clause completion rate | Delivered ÷ required in the period | Overall scope compliance |
| Frequency adherence | Rounds within window ÷ total rounds | Whether delivery is regular or bunched |
| Violation density | Violations ÷ number of centers | Where weakness concentrates |
| Remedy time | Average duration from notice to closure | Contractor responsiveness |
| Recurrence rate | Repeat violations ÷ total | Structural defect or isolated errors |
| Deduction rate | Deductions ÷ entitlement | The actual financial effect of shortfalls |
The last two matter most in renewal decisions. The first four are diagnostic and used during delivery.
The Weighting Rule: Do Not Treat All Contracts Equally
Raw metrics mislead in comparison. A contract with two violations covering one center is worse than one with ten violations covering twenty centers.
Correct weighting rests on three factors: number of covered centers, financial value, and service criticality (a security or safety contract is not a landscaping contract).
Without weighting, the dashboard directs attention toward large ordinary contracts instead of small critical ones.
The Multi-Contract Dashboard
A useful dashboard presents four layers only:
Layer one — contracts needing a decision now. Expiring within 90 days, approaching the deduction ceiling, or carrying clauses undelivered across two consecutive periods.
Layer two — weighted ranking. Contracts ordered by a composite metric rather than alphabetically or chronologically.
Layer three — geographic distribution. Violations mapped across centers, revealing that the problem may lie with the location rather than the contractor.
Layer four — trend over time. Is the metric improving or deteriorating across periods? Direction matters more than absolute value.
The fourth layer is the most neglected despite being the most useful: a contractor at 85% and improving is preferable to one at 90% and declining.
Comparing Contractors: Conditions for Validity
Contractor comparison is useful under two conditions. First: standardized categories and criteria across all contracts, otherwise you are comparing two different scales. Second: accounting for differing circumstances — a high-traffic center naturally generates more violations, and charging that to the contractor without adjustment produces a false result.
The fairest comparison sits within a category: cleaning contractors against each other, maintenance contractors against each other, with results normalized for site size and type.
Metrics That Get Calculated but Never Used
Some metrics look reasonable but lead to no decision:
- Total violation count without context or weighting — a number that moves with activity volume rather than quality
- Number of rounds completed without a ratio to the required figure — it may rise simply because scope expanded
- Average performance across all contractors — it conceals the outliers, which are exactly what needs intervention
The rule: any metric where you cannot name the decision it drives should be removed from the dashboard.
From Single Contract to Portfolio
Managing a contract portfolio differs from managing a contract. New questions appear at portfolio level: should we concentrate contracts with fewer contractors or distribute risk? What proportion of contracts expire in the same quarter, which is an operational risk if they coincide? Are there duplicated clauses across different contracts that could be consolidated?
These questions only surface when every contract sits in one place under standardized categories — and they are usually the source of the largest portfolio saving, larger than any penalty.
How Masharef Supports Multi-Contract Management
The Masharef contract operations module manages operational contracts and their rounds with financial item and violation tracking, while instant reports and analytics provide the measurement layer without manual compilation.
Center management distributes performance across locations through Google Maps integration, making the dashboard’s geographic layer practically possible. User management controls who sees the full contract portfolio and who sees only their scope, through Microsoft Azure integration, and interactive notifications cover the layer of contracts needing a decision now.
Conclusion
At three contracts, metrics are a luxury. At thirty, they are the only way to know where to look. And a good metric is not the most computationally precise one — it is the one that leads to a decision: escalate, revise scope, or decline renewal.
Frequently Asked Questions
How many metrics should we start with?
Three: clause completion rate, remedy time and recurrence rate. Together they cover compliance, responsiveness and pattern, and the rest can be added once data quality has been stable for a sufficient period.
Do the same metrics apply to every contract type?
The core metrics yes, but thresholds differ. A remedy time acceptable in a landscaping contract may be entirely unacceptable in a safety contract. Standardizing the metric while varying the threshold is the correct approach.
When should the metrics themselves be reviewed?
Annually, or whenever the operating model changes. A metric that has driven no decision in a year should be removed, because its presence consumes attention with no return.
How do we handle a contractor with good metrics but poor field impressions?
By reviewing what is not being measured. Usually the clause causing the poor impression is undecomposed in the system or lacks an acceptance criterion, so it appears in no metric despite being felt daily.