The problem with measuring everything
Ask most facility managers what metrics they track for vendor performance, and you'll get one of two answers: either a very short list (usually just complaints and cost), or a very long one (a spreadsheet with 30 indicators that nobody reviews regularly).
Neither approach works. Tracking too little leaves you blind to problems until they escalate. Tracking too much creates noise that obscures the signals that actually matter, and the effort of maintaining a complex system usually means it's abandoned after a few months.
The most effective vendor KPI frameworks are focused and sustainable. These KPIs are also the foundation of a working vendor accountability framework. This article identifies the metrics that consistently predict vendor performance in facility management contexts, and the ones that look important but rarely are.
The distinction between leading and lagging indicators
Before choosing KPIs, it helps to understand the difference between leading and lagging indicators.
Lagging indicators tell you what has already happened, complaint rates, incident counts, failed inspections. They're important for accountability and trend analysis, but they measure failure after the fact.
Leading indicators predict what is likely to happen, response time trends, schedule adherence rates, near-miss reporting frequency. They give you an opportunity to intervene before performance deteriorates significantly.
A good KPI framework includes both. If you're only tracking lagging indicators, you're always reacting. If you're only tracking leading indicators, you don't have a baseline measure of actual outcomes.
KPIs that consistently matter across facility services
Quality score (from structured inspections)
The most direct measure of service quality is a structured inspection score, ideally across multiple zones or categories, aggregated into an overall score using a vendor scorecard for facility services. Quality scores should be tracked over time to identify trends, not just assessed in isolation.
What it tells you: Whether the work being delivered meets the agreed standard.
Why it's a leading indicator: A declining score trend predicts complaints and incidents before they occur.
Schedule adherence rate
What percentage of scheduled tasks is completed on time and in full? For services with a regular cadence, cleaning, landscaping, preventive maintenance, schedule adherence is a fundamental performance indicator.
What it tells you: Whether the vendor is delivering the contracted scope of work.
Why it matters: Low schedule adherence rates are often the root cause of quality complaints. You can't maintain quality if you're missing tasks.
Response time to reactive requests
How quickly does the vendor respond when an unplanned issue arises? Track from the time the request is made to the time work begins, not just to the time of acknowledgement.
What it tells you: Whether the vendor can be relied on when you need them outside of scheduled work.
Segment by priority: Track response times separately for emergency, urgent, and routine requests. A vendor with good routine response times but poor emergency response has a specific and serious gap.
Complaint rate per site (or per 1,000 sq ft)
Normalising complaint rates by facility size or site count makes them comparable across your portfolio. A vendor managing a 50,000 sq ft building will generate more raw complaints than one managing 5,000 sq ft, normalisation removes that distortion.
What it tells you: The occupant experience of the vendor's work, separate from your own inspection results.
Watch for divergence: If your inspection scores are high but complaint rates are rising, either your inspection standard is too lenient or complaints are coming from areas you're not inspecting.
Self-reported incident rate
How frequently does the vendor proactively report problems, near-misses, or service failures before you discover them? This is a useful proxy for vendor transparency and operational maturity.
What it tells you: Whether the vendor takes ownership of their performance, or whether your oversight is the only mechanism catching issues.
A low rate isn't always good: If a vendor is never self-reporting incidents, it may mean nothing is going wrong, or it may mean issues aren't being surfaced.
KPIs that seem important but rarely are in isolation
Cost per service visit. Cost is important in budget management, but as a standalone KPI it doesn't tell you whether you're getting value. A cheaper vendor with poor quality scores costs more in the long run. Track cost alongside quality metrics, not separately.
Number of staff on site. Staffing levels are a contractual compliance metric, not a performance metric. Meeting the minimum staffing requirement doesn't mean the work is being done well.
Years of relationship. Tenure can create a false sense of accountability. A vendor who has been in place for five years and has never been formally evaluated may be significantly underperforming relative to what the market can offer.
Building a KPI dashboard that actually gets used
The key to a useful KPI framework is making it easy to review and act on. A well-designed vendor dashboard should:
- Show performance for all your vendors in a single view
- Highlight which vendors are trending in the wrong direction
- Be updated automatically or with minimal manual effort
- Feed directly into your contract review and renewal process
If your KPI review is a monthly exercise in pulling data from multiple sources into a spreadsheet, it will be the first thing that gets cut when you're busy. The system needs to fit into your workflow, not create a separate one.
Evalystar is designed around this problem, giving facility teams a real-time view of vendor performance across all their sites, with inspection scores, complaint data, and response time metrics in a single dashboard. See the platform.