A monthly walkthrough is useful, but it is a sample taken after the work has happened. A scorecard gives a facility manager a repeatable way to discuss what the cleaning vendor delivered, what exceptions remained open, how quickly problems were addressed, and whether the invoice matched the contract. The scorecard should make decisions easier, not create a second bureaucracy that rewards a vendor for filling in fields.
ISSA quality concepts commonly begin with a written specification, inspection, evaluation, and corrective action. That sequence matters because a scorecard cannot be fair if the buyer has not defined the result or if the inspection method changes from one month to the next. Metrics should be tied to the service the building purchased and qualified with the limits of the evidence.
StockPoint can supply building-level punch events, photo and PIN proof, live checkpoint status, supervisor-reviewed exceptions, client portal reporting, and audit-logged corrections. It can help a buyer compare evidence to the contract, but a scorecard still needs human judgment. A green status is not a substitute for inspecting a high-risk area, and a red event is not automatically a reason to penalize a vendor.
Start with the specification and the decision
The first scorecard field should be scope: building, areas, frequency, service window, and exclusions. Without that context, an attendance percentage can reward a crew for arriving at a building where the requested task was impossible or punish a vendor for not performing work that was never included. Put change orders and temporary closures into the same record.
Attendance can measure scheduled visits with an arrival or building checkpoint, subject to agreed tolerance and documented exceptions. Task completion can measure the defined work, not the number of taps in an app. Quality can use inspections, samples, complaints, and corrective action. Response time can measure the time from a reported issue to acknowledgment or resolution, with separate definitions where resolution depends on access or a customer decision.
Avoid using a single blended score that hides the reason for failure. A facility manager should be able to say whether the problem was a missed visit, an incomplete task, a quality defect, a late response, a safety concern, or an invoice mismatch. The vendor can then assign the right supervisor and remedy.
Choose evidence the team can interpret
Attendance evidence should be modest and clear. A building-level punch, PIN, QR checkpoint, or supervisor confirmation can show that a visit was associated with the site. GPS can add context, but indoor accuracy varies and should be displayed honestly. Require a review for an outlier rather than treating a weak signal as conclusive proof that a worker was absent.
Task completion needs a defined basis. A checklist can show that assigned areas were reviewed, while a photo can show an agreed condition or specialty result. A supervisor inspection can confirm the process used by the vendor, but it should not be presented as a guarantee that every surface was perfect. The scorecard should say whether the evidence is self-reported, supervisor-reviewed, client-observed, or sample-based.
StockPoint’s client portal can show live checkpoint status and proof-of-work photos selected for the contract. The facility should not require continuous tracking or indiscriminate photography simply to improve a score. Evidence that is intrusive, expensive, or impossible to review may reduce trust without improving quality.
Measure quality with inspection and corrective action
Quality metrics should reflect the building’s risk. A healthcare-adjacent area, food-service space, office lobby, and construction handoff may require different inspections and escalation. Do not invent a universal pass percentage. Define the inspection areas, acceptable condition, sampling method if used, and correction window with the vendor and the building’s responsible stakeholders.
Use complaints as signals, not as an automatic count of failed shifts. A tenant complaint may identify a real missed task, but it may also concern a changed scope, a spill after the cleaning visit, or an area the vendor could not access. The reviewer should record the source, condition, time, investigation, and resolution. A scorecard that counts every complaint as a vendor failure will encourage argument instead of improvement.
A corrective action can be a return visit, a resupply, a method change, a supervisor inspection, an access fix, or a contract change. The record should identify the owner and closeout. Audit logs are useful because they preserve the first report and the later resolution rather than allowing a month-end score to overwrite the operational history.
Separate acknowledgment from resolution
Response-time metrics need a clear starting point. Is the clock the moment the facility reports an issue, the moment the vendor receives an approved work order, or the moment access becomes available? Those are different measures. Define acknowledgment separately from resolution when the vendor cannot complete the work until a tenant, security desk, or building engineer acts.
A scorecard can distinguish routine, urgent, and safety-related exceptions without pretending that every issue has the same remedy. A blocked room should not be treated like a spill on a public path, and a safety stop should not be penalized as a missed task. The buyer should document the urgency rule and the escalation contact.
The client portal can show open exceptions and next actions so the facility does not wait for the monthly meeting to discover a problem. The vendor can use bilingual workforce surfaces to communicate the assignment or correction to the crew, while the supervisor remains responsible for the response. Visibility should accelerate a decision, not create pressure to close an issue inaccurately.
Include safety without rewarding silence
Safety belongs on the scorecard as a process, not as a vanity number. Track whether required site rules were communicated, whether unsafe access or chemical conditions were reported, whether a worker stopped and escalated a task, and whether corrective action was completed. OSHA’s Hazard Communication materials and workplace-safety guidance emphasize hazard communication, training, and prevention; a facility should not encourage a vendor to hide a safety stop in order to preserve a perfect service score.
A buyer can require a vendor to maintain applicable training, safety, insurance, and incident records and to report contract-relevant events. It should not demand personal medical details or use the scorecard to discipline an individual cleaner. Route a concern to the vendor’s supervisor and record the site facts.
Safety metrics should be reviewed alongside access and scope. If the facility repeatedly blocks the safe route or changes a chemical-storage rule without telling the vendor, the remedy may be a building control rather than a vendor penalty. A joint review is more useful than a score that assigns blame to the party least able to fix the condition.
Make invoice accuracy traceable to the contract
Invoice accuracy is a buyer metric and a vendor metric. Compare the invoice to the contracted unit, fixed fee, approved time, or cost-plus rule. The facility should know whether a line comes from a verified building punch, a change order, a scheduled periodic task, or a manual adjustment. The vendor should be able to explain a correction without re-keying the entire month.
Consider a worked example. A vendor serves four buildings under a cost-plus hourly clause. The month includes 120 approved building hours at an agreed bill rate of $32, plus a separately approved floor-care project. The scorecard should reconcile the 120 hours to building-level events and show the floor-care authorization, not compare the total invoice to a blended average and call every variance an error. The figures are illustrative; the contract controls.
StockPoint can use the same approved punch for worker-pay preparation and cost-plus support, with per-worker payroll locking so a correction is not paid twice. It does not decide whether a task is billable under the contract, and bank-feed reconciliation is on the roadmap, not shipped. Accounting should reconcile invoices and deposits in its own process.
Tie service credits to reviewed evidence
Service credits should follow evidence and a defined remedy. If the contract says an unperformed visit earns a credit, the parties should define the evidence, the exception review, the calculation, and the timing. Do not apply a credit automatically from a failed GPS signal, a missing photo, or a complaint that has not been investigated.
A fair scorecard can report “open,” “under review,” “accepted,” “corrected,” and “credit issued” as separate states. The vendor can explain that access was unavailable, a safety issue required a stop, or a scope changed. The facility can still decide that the contract remedy applies, but its decision should be visible and consistent.
Use the proof-of-work guide to define the evidence and the RFP audit-rights guide to define access. A scorecard should not create a new legal or employment control by accident. Facility staff should report results to the vendor’s manager rather than direct a worker’s discipline or time.
Use the scorecard to improve the operating relationship
Review the scorecard with the vendor on a regular cadence and ask which metric changed behavior. If a field does not inform a staffing, scope, safety, access, or contract decision, remove it. If a metric improves while complaints or unresolved exceptions increase, investigate whether the measurement is rewarding the wrong behavior.
Show trends with the underlying count and denominator. “Three unresolved exceptions out of 84 scheduled visits” is more useful than “96 percent compliant” when the reader needs to understand what remains open. Use qualified language; a monthly score is evidence about that period, not proof that every future shift will match it.
Give the vendor a chance to correct data errors and give facility staff a channel to explain building-caused exceptions. A bilingual worker-facing correction process helps the vendor verify time and task facts without making the employee responsible for the buyer’s metric.
Build the scorecard the contract can support
A useful scorecard has a small set of contract-linked fields: scheduled service and attendance context, task or inspection result, exception and response, safety process, invoice reconciliation, and corrective status. Each field has a definition, evidence source, owner, review cadence, and dispute path. That is enough to move a monthly walkthrough from opinion toward an accountable operating review.
StockPoint’s features connect per-building punches, photo and PIN proof, live client status, bilingual workforce surfaces, payroll preparation, payroll locking, and audit history. The platform does not replace ISSA-informed specifications, the facility manager’s inspection, the vendor’s supervision, or the contract’s remedy language.
Sign up at getstockpoint.com to give your facility and cleaning vendor a shared scorecard foundation based on work that can be reviewed. You get clearer service conversations, more defensible invoice decisions, and a practical record of exceptions and corrections without turning every operational question into a surveillance contest.