Facility Management Insights

You inherit a facilities portfolio and discover that “vendor performance” means whatever the loudest stakeholder remembers from the last meeting. Janitorial complaints sit in email threads, HVAC response times live in a CMMS export, and landscaping reviews depend on whoever walked the site that morning. Renewal discussions then become debates about relationships instead of evidence.

A vendor performance scorecard template fixes that gap only when it reflects how facilities work. The useful version defines the measurement rule, names the data source, assigns an owner, sets a review cadence, and links poor results to a contract action. A colorful spreadsheet without those controls is just another report nobody trusts.

Why Most Facility Vendor Programs Run Without a Real Scorecard

A facilities lead taking over a multi-site portfolio often discovers that one vendor program contains several unrelated evaluation systems. Occupant complaints and inspection notes shape the janitorial review. Work orders and emergency calls shape HVAC assessments. Landscaping performance is discussed after seasonal site walks. Each service may have useful evidence, but the definitions, time periods, and accountable reviewers do not match.

Contract language can create the same problem. “Maintain industry standards” does not tell a supervisor how to score restroom sanitation, preventive-maintenance completion, or irrigation reliability. By renewal time, evidence is spread across emails, invoices, meeting notes, and individual memory. A usable scorecard replaces that informal process with recurring measures that fit the service and the contract.

The inherited-program problem

The previous review may be little more than a checkbox. One building can have repeated missed cleans, another unresolved air-conditioning faults, and a third overgrown paths that create a slip or trip concern, while the vendor still receives a broadly positive assessment. A portfolio average can conceal the pattern that matters operationally.

A scorecard gives the FM lead a defensible basis for comparing similar services, documenting a renewal recommendation, and identifying risk before an emergency forces a decision. It also makes the vendor discussion fairer. The supplier can see the scoring rule, the supporting record, and the action required.

Practical rule: If two reviewers can score the same event differently because the template does not define the evidence, the KPI is not ready for a contract review.

Start with the operational plumbing. State whether a janitorial inspection uses a site checklist, whether an HVAC response clock begins at ticket creation or dispatch, and whether landscaping completion requires a signed visit record. A useful contract management approach for facilities teams connects those rules with renewal dates, service-level clauses, invoice approvals, and corrective-action records.

Multi-site scoring also needs a fairness rule. Compare like with like, record site-specific exceptions, and separate a vendor's controllable performance from access restrictions, tenant delays, or approved scope changes. A single portfolio average should support review, not erase differences between buildings or service conditions. Practical examples and cleaning operations perspectives are also available through the Arelli Cleaning blog.

For cleaning teams, the scorecard should protect public-health requirements rather than reward appearance alone. CDC guidance states that disinfectants must keep a surface wet for the full product contact time listed in the product directions and Safety Data Sheet CDC facility cleaning guidance. When disinfection is included in the contract, verify that rule through training records and spot audits.

The remedy is a controlled scorecard with a small core, service-specific measures, and documented exceptions. It turns informal feedback into evidence while recognizing that buildings and vendor risks differ.

Core KPI Categories and a Sensible Weighting Structure

A practical scorecard starts with five KPI categories, then adapts the detail to the contract. Common supplier scorecard structures use quality, delivery, cost, service, and compliance or risk, with the total weighting adding up to 100% Ivalua's vendor scorecard framework. The mistake is treating those categories as equal when the operational consequences aren't equal.

For a typical facilities contract, a useful starting point is quality at 30%, delivery and responsiveness at 25%, cost compliance at 20%, safety and risk at 15%, and service or relationship at 10%. This is a starting model, not a universal truth. A life-safety systems contractor may need more weight on compliance and risk, while a low-risk commodity service may justify greater attention to invoice accuracy and price stability.

KPI Category	Default Weight	Example Metrics	When to Increase
Quality	30%	Inspection pass rate, defect rate, first-pass completion, open corrective actions	Increase for hygiene, critical assets, or regulated work
Delivery and responsiveness	25%	On-time service, response time, lead-time variance, order-fill rate	Increase when missed visits or delays disrupt operations
Cost compliance	20%	Invoice accuracy, price variance, contract-rate compliance	Increase for commodity services with interchangeable providers
Safety and risk	15%	Training records, certificates, incident controls, financial health	Increase for emergency, life-safety, or high-hazard work
Service or relationship	10%	Communication, escalation handling, improvement participation	Increase only when the rubric is structured and auditable

Lock the rules before collecting data

Procurement, finance, operations, and the contract owner should approve the category weights before the first reporting period. If the team changes weighting after seeing a disappointing result, the scorecard becomes a negotiation tool rather than a measurement system.

Each KPI needs four fields beside its target:

• Measurement rule: State exactly what counts and what doesn't.
• Data source: Name the CMMS field, invoice record, inspection form, survey, or compliance register.
• Frequency: Specify whether the value is captured per visit, weekly, monthly, or during a formal review.
• Data owner: Assign one person to protect completeness and resolve disputes.

A technical scorecard can normalize each metric to a 0 to 100 score before applying the composite calculation. Guidance from ProcurementVMS describes a commonly used pattern of quality 30%, delivery 25%, cost 20%, service 15%, and risk or compliance 10%, with threshold logic defined in advance vendor scorecard template guidance. Your facilities version can use the alternative weighting above, but the principle remains the same: weights must be visible, stable, and tied to business impact.

For a broader facilities KPI reference, the facilities management KPI guide can help teams identify measures before trimming them to a usable set.

Customizing Metrics for Janitorial, HVAC, and Landscaping Vendors

The five-category spine should stay recognizable across the portfolio, but the individual KPIs must follow the service risk. A janitorial provider, an HVAC contractor, and a landscaping company can all score “quality,” yet the evidence behind that score is completely different.

Janitorial services

Janitorial quality should be grounded in structured inspections rather than general impressions. Useful measures include inspection pass rate, restroom audit results, supply replenishment compliance, completion of scheduled areas, and response to spot requests. A short occupant survey can add context, but it shouldn't replace inspection evidence. Keep questions tied to observable service, such as restroom condition, odor control, and availability of consumables.

Cleaning protocols also need a compliance field. If disinfection is required, the supervisor should verify product identity, application method, and wet contact time. CDC recommendations note that most EPA-registered hospital disinfectants have a label contact time of 10 minutes, while some pathogens may be addressed by products showing efficacy at at least 1 minute, so the product label, not a generic template, controls the audit CDC disinfection recommendations.

HVAC contracts

HVAC scoring should emphasize asset availability and preventive control. Track uptime, mean time to repair, preventive-maintenance completion, energy performance against an agreed baseline, and refrigerant-leak compliance. A ticket that closes quickly but leaves the same air-handling unit failing repeatedly shouldn't receive the same interpretation as a durable first-time repair.

Landscaping services

Landscaping measures should reflect scheduled work and site safety. Use visit completion records, seasonal quality audits, irrigation-system uptime, and incidents connected to overgrowth, blocked paths, or chemical application. Weather and site conditions need documented exception codes, otherwise the vendor gets penalized for events outside the agreed scope.

Category	Janitorial KPI	HVAC KPI	Landscaping KPI
Quality	Restroom audit score and inspection pass rate	Repair quality and asset condition after work	Seasonal appearance and horticultural audit
Delivery and responsiveness	Spot-request response and scheduled-area completion	Emergency response and PM completion	Scheduled visit completion and issue response
Cost compliance	Invoice line accuracy and supply-rate compliance	Approved parts pricing and scope variance	Visit billing and materials variance
Safety and risk	Chemical records, training, and disinfection procedure	Refrigerant compliance, permits, and technician credentials	Chemical application, path safety, and equipment controls
Service	Supervisor communication and occupant feedback	Escalation handling and root-cause reporting	Site communication and seasonal planning

Every row needs a named source and owner. The cleaning services contract resource is useful when translating janitorial expectations into measurable scope language rather than leaving them as broad service promises.

For gym, recreation-center, and campus facilities, add equipment and high-touch-area checks to the janitorial work order. Where a contract supplies disinfecting wipes, specify whether staff are expected to use gym equipment cleaning wipes, replenish a gym wipe dispenser, and document empty dispensers or damaged equipment. CDC guidance also says that products must be used according to manufacturer instructions for concentration, application method, and contact time CDC community-facility cleaning guidance.

How the Score and Thresholds Are Actually Calculated

A scorecard becomes credible when a vendor can trace every result back to a record. Start with the raw operational event, then convert unlike units into a common scale. Work-order timestamps can support response time, inspection forms can support cleanliness, invoices can support commercial accuracy, and tenant surveys can provide structured perception data.

Build the calculation in layers

For each KPI, choose either a target-based formula or a min-max method. Target-based scoring is usually easier to explain in a vendor meeting. For a measure where higher is better, such as on-time completion, a score can rise toward 100 as the target is met. For a measure where lower is better, such as lead-time variance or unresolved actions, the formula must reverse the direction. Set the threshold logic before the reporting period starts.

The composite is straightforward:

Category score × category weight, then add every weighted result.

A spreadsheet should show the raw value, normalized score, weight, weighted score, source record, and exception note. That audit trail matters more than a polished traffic-light graphic.

Worked example

The following example uses the requested calculation structure. It is an illustrative model, not a reported vendor result. The five KPI values are normalized against pre-agreed targets, and the category weights total 100%.

KPI	Raw Value	Normalized (0–100)	Weight	Weighted Score
Inspection pass rate	94%	92	30%	27.60
Scheduled-area completion	97%	95	25%	23.75
Invoice accuracy	98%	98	20%	19.60
Safety and training compliance	90%	88	15%	13.20
Supervisor service rating	4.4 / 5	66	10%	6.60
Composite score			100%	90.75

That example illustrates the mechanics, but it doesn't justify inventing a final score for a real contract. If your source data produces 87.4, the spreadsheet should show exactly which normalized values and weights produced that result. A score isn't defensible because it looks precise. It's defensible because another reviewer can reproduce it.

Thresholds and exceptions

A practical banding model can use Green at 90 or above, Yellow from 75 through 89, and Red below 75. Those bands should be paired with individual SLA triggers. A vendor might have a healthy composite while still breaching a critical requirement, such as a missed emergency response, expired certification, or repeated disinfection failure.

Handle missing data explicitly. Mark a KPI as not measured, exclude it only under a documented rule, and redistribute weight transparently if the contract permits it. Don't convert missing records into automatic success. Outliers should be investigated, not deleted, and partial-month results should be labeled as partial rather than presented as a full-period trend.

Reporting Cadence, SLAs, and Governance by Vendor Tier

A scorecard becomes useful when it fits the operating rhythm. Classify vendors by business impact, substitution difficulty, safety exposure, and service frequency. Then set the review cycle from those conditions, not from an administrative default.

Use three practical tiers:

• Strategic: Annual or business-critical contracts, high operational dependence, difficult substitution, or major safety exposure.
• Preferred: Recurring services with moderate risk and an established scope.
• Transactional: Low-spend, low-risk, or on-call work that requires review by exception.

Strategic vendors generally need a monthly operational review and a quarterly business review. Preferred vendors can use a bi-monthly operating check-in. Transactional vendors can be reviewed quarterly or after a threshold breach. A janitorial provider may need frequent inspection review, while an HVAC specialist may require closer attention after repeat failures. Landscaping schedules should account for seasonal deliverables and site conditions.

Turn scores into contract actions

Every scored KPI needs a measurement rule, a source, and an owner. Put the response-time definition beside the clock start and stop rules. Put first-time fix beside the closure standard. Put cleanliness results beside the inspection method, sample location, and evidence requirement. Without those details, two sites can report different results for the same service.

A Yellow result should create a corrective-action record before the next review. A Red result should trigger the escalation written into the contract, particularly when safety or compliance is involved.

Assign decision rights clearly:

• FM lead: Owns the composite score and operational interpretation.
• Procurement or analytics: Validates data integrity and version control.
• Vendor manager: Confirms action owners and due dates.
• Steering committee: Decides on remedies, scope changes, incentives, or exit.

Keep the monthly agenda practical: review open actions, confirm the current score, examine Yellow and Red KPIs, test root causes, agree due dates, and record contract or invoice decisions. For landscaping scopes, teams may also use practical yard care contract help when defining seasonal deliverables and visit evidence.

For multi-site contracts, compare like with like. Apply the same KPI definitions, but record site-level exceptions such as access restrictions, weather, occupancy, or approved scope changes. The score should inform renewals, incentives, invoice approvals, and off-boarding. A scorecard that never affects a decision is only a report.

Common Pitfalls That Break a Scorecard in Practice

The most common failure isn't a bad formula. It's a scorecard that asks for more information than the team can maintain. I've seen templates grow past the point where site supervisors can explain the result, then lose credibility because nobody knows which measures change a decision.

Too many metrics dilute accountability

Keep the core scorecard to 8 to 10 KPIs. Retire any measure that has no reliable source, no assigned owner, or no corresponding management action. A long list can create the appearance of control while hiding the metrics that reveal missed cleans, repeat HVAC failures, or unsafe grounds conditions.

Mixed data sources create a second problem. A CMMS may show 95% SLA compliance while an occupant survey shows 60% satisfaction, but those figures aren't necessarily contradictory. They may measure different populations, definitions, or time windows. Declare one source of truth for each KPI, then retain other inputs as diagnostic context. Audit the source and the rule on a regular schedule.

Soft measures need discipline. Responsiveness and relationship quality matter, but they drift when reviewers use different interpretations. Cap soft KPIs at 20% of the total weight, use a defined rubric, and require evidence such as response logs, action closure, or documented escalation handling supplier scorecard implementation guidance.

Protect multi-site comparisons

One loud tenant shouldn't define a vendor's performance across an entire portfolio. Use site-count weighting, a minimum response sample, or a separate site-level view before calculating the portfolio result. A campus with intensive event turnover shouldn't be compared directly with a quiet office without documenting the scope modifier.

Before launch, confirm this readiness checklist:

• Definitions: Every KPI has a written numerator, denominator, target, and exclusion rule.
• Evidence: Each result points to a work order, inspection, invoice, survey, or compliance record.
• Ownership: One person validates each data stream.
• Cadence: The review date and escalation route are written into the governance calendar.
• Actions: Yellow and Red results automatically enter an action log.
• Retirement: The team has permission to remove KPIs that don't support decisions.

The final pitfall is treating the scorecard as a filing cabinet. Require the Yellow and Red action log to appear in every QBR, with a named owner and due date. If the meeting doesn't produce a decision, the template isn't doing its job.

FAQ on Piloting, Multi-Site Fairness, and Underperformers

How long should a pilot run

A cautious rollout can use 30 days to calibrate scoring rules with one vendor, 60 days to validate data feeds, and 90 days to expand to a second site. Those stages are useful because they separate a bad definition from a bad supplier result. During calibration, compare the score produced by the template with the evidence a site supervisor would normally use, then resolve disagreements before expanding.

Exit criteria should be practical. The team should know where every KPI comes from, reproduce the composite without manual guesswork, explain exceptions, and complete a review with the vendor using the same data. If the second site needs a different rule, document whether that difference reflects legitimate scope or an inconsistent process.

How do you score different sites fairly

Normalize for square footage, occupancy, service scope, operating hours, and event intensity before comparing results. A 50,000-square-foot office and a 500,000-square-foot campus can use the same core KPI definitions, but the workload, traffic, and asset complexity may require site-specific modifiers.

Simple averages are easy to calculate but can let a small site carry the same influence as a major campus. Site-count weighting can correct that in some portfolios. A z-score approach can help compare performance relative to a peer group, but only when the underlying data is consistent enough to support meaningful distribution analysis. Teams should begin with transparent target-based normalization and add statistical methods only when the data governance is mature.

Fairness doesn't mean every site gets the same formula. It means every difference is documented before the score is used.

Keep core KPIs consistent, then add modifiers for service scope, building type, region, or criticality. A recreation center may require equipment sanitization checks and event turnover evidence, while a corporate office may need greater emphasis on scheduled restroom audits.

What should happen when a vendor stays below target

Use a decision tree tied to the score and the breached KPI:

1. Yellow result: Open a corrective action, confirm root cause, and set a review date.
2. Repeated Yellow or Red result: Start a formal cure period under the contract and require a recovery plan.
3. Critical KPI breach: Escalate immediately, even if the composite remains acceptable.
4. No sustained improvement: Consider scope reduction, added oversight, alternate sourcing, or contract exit.
5. Exit decision: Preserve the evidence trail, transition requirements, asset information, and service continuity plan.

For janitorial teams, corrective action may involve retraining, chemical-use verification, inspection calibration, or dispenser replenishment controls. In fitness facilities, operators can standardize wipes to disinfect gym equipment, including yoga mat wipes where the product label supports that use. OSHA states that when an employer determines an emergency could develop, affected housekeeping personnel must receive at least first-responder awareness-level training under 29 CFR 1910.120(q)(6)(i) OSHA interpretation for housekeeping personnel.

For bulk purchasing, compare commercial disinfecting wipes and other facility supplies by label suitability, dispenser compatibility, replenishment controls, and contract cost rather than unit price alone. Facilities teams can review options such as commercial cleaning and sanitizing supplies as part of that documented procurement decision.

A scorecard should make the next action obvious. If it only produces a color, it hasn't finished the job.

---

Download or build your own vendor performance scorecard template today, then pilot it with one janitorial, HVAC, or landscaping contract. Define the evidence, assign the data owners, schedule the first review, and connect every Yellow or Red result to a corrective action before the next renewal discussion.