How to Run a Cleaning
Vendor Pilot Program
A 90-day pilot is the most reliable way to evaluate a cleaning vendor before committing to a full contract. Here is the structure that makes pilots useful rather than just a formality.
Facilities that run a structured pilot program before full-term contract award report 73% lower vendor replacement rates in the first 18 months compared to facilities that skip the pilot phase. (IFMA Procurement Best Practices, 2024)
The Short Answer
A cleaning vendor pilot program should run 90 days, cover a zone that reflects your facility full complexity, and include three measurement tracks: inspection scores on a defined rubric, response time adherence, and incident documentation. The pilot agreement should include an explicit exit right at the end of the period with no automatic transition to a full contract. A pilot without measurement is not a pilot. It is a trial period with no accountability.
Vendor Selection
The most common pilot program failure is running the pilot with no defined evaluation criteria. The pilot ends, everyone agrees it went 'pretty well,' and the full contract is signed. Then the new-account effect wears off and service declines. The pilot told you nothing because you were not measuring anything specific.
Standard pilot program duration. Long enough for the new-account effect to wear off and two full inspection cycles to complete. Short enough to protect both parties from an extended bad-fit engagement.
Millennium Facility Services Pilot Framework, 2026
Why Most Pilots Fail to Produce Useful Information
A pilot program without defined evaluation criteria is not a pilot. It is an informal trial period that ends with subjective impressions rather than data. The new-account effect is real: cleaning vendors typically over-staff, over-communicate, and over-respond to requests during the first 30 days of any account. A 30-day pilot captures the vendor at their best, not at their normal operating level.
The 90-day timeline is designed to get past that effect. By week six or seven, the vendor operations team has shifted to their normal staffing rhythm. By week ten, you have seen how they handle a crew absence, a service complaint, and a schedule conflict across at least one full inspection cycle. That is the vendor you are actually hiring.
The second reason pilots fail is that the evaluation criteria are not established before the pilot starts. When both parties agree at the beginning on what success looks like, the decision at day 90 is a data-driven conclusion, not a political negotiation.
The Four-Phase Pilot Structure
Phase 1: Pilot Design (Weeks 1-2 before start)
- Define the pilot zone: areas, square footage, cleaning frequency
- Draft the pilot agreement with clear evaluation criteria and exit rights
- Establish the inspection rubric and scoring threshold
- Set the measurement schedule (inspection dates, reporting cadence)
- Confirm vendor mobilization requirements and timeline
Phase 2: Mobilization (Days 1-14 of pilot)
- Verify all personnel have completed site access requirements
- Collect current certificate of insurance and background check documentation
- Conduct joint walkthrough with vendor account supervisor
- Review and approve vendor task schedule
- Complete baseline inspection of pilot zone (pre-pilot benchmark)
Phase 3: Active Pilot (Days 15-75)
- Conduct inspections on defined schedule using the agreed rubric
- Track all service requests and response times
- Document any failures, missed service, or quality issues in writing
- Share documented findings with vendor within 24 hours
- Hold a 30-day mid-pilot review with vendor account supervisor
Phase 4: Evaluation (Days 76-90)
- Compile inspection scores across all completed inspections
- Calculate average response time for all service requests
- Review incident log and resolution outcomes
- Score vendor against the defined pilot evaluation rubric
- Make and document the go/no-go decision with supporting data
What the Pilot Agreement Must Include
The pilot agreement is distinct from a full service contract. It covers the pilot period only, specifies evaluation criteria and documentation requirements, and includes an explicit exit right at the end of the period with no automatic transition to a full agreement.
FM Intelligence Series
Pilot program research for facility managers
Download our guides on vendor evaluation, pilot structures, and the criteria that predict long-term program success.
Making the Go/No-Go Decision
At day 90, you have three data sets: inspection scores across all completed inspections, a response time record for all service requests, and an incident log with resolution documentation. Those three data sets should drive the decision, not the relationship that has developed over the pilot period.
A pilot that produced average inspection scores below the defined threshold, two or more unresolved service failures, or consistent response time misses has told you something accurate and valuable about this vendor. The decision to decline a full contract based on that data is not a judgment call. It is a data-supported conclusion.
A pilot that produced strong inspection scores, prompt response times, and documented resolution of every service issue is equally informative. Those vendors have earned a full contract offer with the confidence of real performance data behind it.
We Welcome Pilot Programs
We propose pilots for complex accounts because we know what our performance data looks like at day 90. If you are evaluating vendors, a structured pilot is the right way to do it.
(800) 956-8745Get facility insights that save money
Join facility managers who get our operational intelligence. No spam. Unsubscribe anytime.
Or download our free research reports without signing up.
Frequently Asked Questions
Ninety days is the standard pilot duration for commercial cleaning programs. It is long enough for the new-account effect to wear off, for the vendor operations team to hit their normal operating rhythm, and for at least two full inspection cycles to occur. A 30-day pilot is too short: vendors typically over-staff new accounts in the first month. A 180-day pilot is unnecessarily long for most facilities. If you cannot evaluate a cleaning vendor core operational capacity in 90 days, the problem is not the timeline, it is the measurement framework.
For multi-building campuses, select a pilot zone that represents your facility full complexity: a mix of high-traffic and low-traffic areas, at least one full restroom bank, at least one floor type that requires specialized care, and any areas with regulatory or sensitivity requirements (server rooms, medical areas, production floors). Avoid selecting only easy areas for the pilot. A vendor who performs well on easy areas and struggles in complex ones will not reveal that capability in a cherry-picked pilot zone. The pilot area should reflect your actual operation.
The pilot agreement should explicitly state that it is a pilot evaluation with no automatic transition to a full-term agreement. At the end of the pilot period, both parties can evaluate the results and negotiate a full contract based on actual performance data. Avoid language that treats the pilot as the first 90 days of a longer agreement, because that removes the exit right that makes a pilot valuable. A vendor who resists a true pilot structure and insists on a full-term agreement from day one is a signal worth paying attention to.
Three measurement tracks: inspection scores (weekly or biweekly, using a defined rubric), response time adherence (for any service requests during the pilot period), and incident tracking (any missed service, quality failure, or complaint and its resolution). Document every measurement from day one. By day 90, you should have a quantitative record of performance that either supports a full contract offer or documents why the pilot did not meet your standard. A pilot without documentation is an experience, not an evaluation.
Yes, and for large or complex facilities, it is often the best approach. Assign each vendor a zone and evaluate on the same rubric simultaneously. This eliminates the sequential comparison problem where the second vendor performance cannot be directly compared to the first because the facility, the season, or the evaluation team attention has changed. Parallel pilots produce the cleanest data. The logistical complexity of running two or three vendor zones simultaneously is manageable and well worth the investment for a major contract.
During mobilization, which typically spans the first one to two weeks of the pilot, the vendor should: complete site access procedures for all assigned personnel, provide an up-to-date certificate of insurance and personnel background check documentation, conduct a full site walkthrough with the facility manager, produce a task schedule mapping their cleaning cycle to your facility operational hours, and introduce the named account supervisor and confirm their contact information and availability hours. A vendor who treats mobilization as informal is showing you how they will manage the account long term.
A well-structured pilot agreement gives you a clean exit. If the vendor does not meet the documented performance standard during the pilot period, the pilot ends without obligation and you repeat the selection process with another candidate. The cost of a 90-day pilot that does not result in a contract is infinitely lower than the cost of an 18-month contract with a vendor who was never going to meet your standard. Document every failure during the pilot period in writing, share it with the vendor in real time, and give them the opportunity to respond. If the pattern does not improve, the data supports the decision to exit.
Pilot pricing can reasonably be slightly higher than full-contract pricing because the vendor is absorbing higher mobilization costs and cannot guarantee labor continuity at scale for a 90-day engagement. A 5 to 10% premium over the expected annual contract rate is reasonable. Be cautious of vendors who quote very low pilot prices and significantly higher full-contract prices. That pricing structure is designed to get you into the relationship, not to reflect their actual cost model. Ask the vendor to provide a line-item breakdown of what drives the difference between pilot and full-contract pricing.
Days. Four phases. One decision based on actual performance data, not a sales presentation.
A structured pilot is the right way to evaluate any cleaning vendor. We propose them for complex accounts because we are confident in what the data shows at day 90. Request an assessment and let us discuss the structure.