How to Run a Pilot Composting Program at One Location

Sustainability programs that get rolled out across multi-location operations without piloting first often fail expensively. The hauler relationship breaks down at three of fifteen sites, customer behavior is different than expected at the high-volume locations, signage that worked in the test office doesn’t work in the conference center, and the budget overrun makes the program politically difficult to defend.

Piloting at a single location before chain or campus rollout costs roughly 5-10% of the total program budget and tells you most of what you need to know about whether the full rollout will succeed. The pilot methodology is well-understood — most large foodservice operations use some version of it for any operational change of meaningful scope.

This is the playbook for running a 90-day single-location composting pilot. It’s written for operators contemplating compost program rollouts at chain restaurants, university dining systems, hospital networks, multi-site offices, hotel chains, and similar operations.

What a pilot is and isn’t

A pilot is a structured experiment. It has a hypothesis, a baseline, a controlled intervention, measurable outcomes, and a decision point at the end. Done well, a pilot answers specific questions that inform the scaling decision.

A pilot is not just “trying it at one location and seeing what happens.” That’s a soft launch. Soft launches don’t generate the structured data needed to confidently scale or pull back. They tend to produce ambiguous results that get interpreted to support whatever the operations team already wanted to do.

The discipline of a true pilot is what makes it useful for the scaling decision.

Step 1: Define the hypothesis

Before you start, write down what you’re testing. Specific examples:

• “We can implement a compost program at our flagship location with <10% contamination rate within 60 days, with a per-cover cost premium under $0.40.”
• “Customer compliance with compost bin sorting will reach 75%+ after two weeks of bin-station volunteer support.”
• “Our hauler can reliably collect compost twice weekly without missed pickups across an 8-week test period.”
• “A combined operations and customer-facing compost program will not cause measurable disruption to service times during peak hours.”

The hypothesis should be specific, measurable, and decidable. Vague hypotheses (“we’ll see if it works”) produce vague results.

Multiple hypotheses can be tested in a single pilot. Three to five is a reasonable scope; ten is overreach.

Step 2: Pick the right location

Pilot location selection matters more than people realize. Pick poorly and you get results that don’t predict the rest of the chain.

Pick a location that’s representative. Mid-volume, mid-customer-mix, mid-complexity. Not the highest-volume flagship (problems become catastrophic) and not the lowest-volume sleepy outpost (problems don’t surface).

Pick a location with engaged management. A pilot needs a site manager who’ll engage with the methodology, capture data, and report honestly. Pick a location with leadership that’s interested, not skeptical.

Pick a location with composting infrastructure access. If three of your fifteen sites have working composting infrastructure and twelve don’t, piloting at one of the three doesn’t tell you anything useful about the rollout. Pick a location representative of the broader infrastructure availability.

Don’t pick the obvious sustainability flagship. That location is already running things that other locations don’t. Results there don’t generalize.

Pick a location that can afford to fail. If the pilot reveals serious problems, the location’s normal operations need to be able to absorb the disruption.

Step 3: Establish baseline data

Before any intervention, capture what’s currently happening:

• Total waste volume per week. Pounds, cubic yards, or whatever unit your hauler uses. Across at least 4 weeks.
• Waste composition. Sample audits — sort one day’s trash to estimate what percentage is currently compostable. Most foodservice operations are 60-80% compostable by volume.
• Current waste hauling cost. Per week and projected per year.
• Customer satisfaction baseline. Through whatever measurement you currently use (surveys, reviews, complaints log).
• Service time baseline. Order-to-delivery time during peak hours.
• Staff time on waste handling. Estimated hours per week.

Without baseline data, you can’t tell whether the pilot improved or worsened any metric. You’ll be flying blind.

Step 4: Design the intervention

The compost program itself. Specifics:

• Compostable foodware spec for whatever items the pilot location will switch (cups, plates, utensils, takeout containers).
• Bin placement and signage.
• Staff training plan.
• Hauler service agreement (typically a 90-day pilot rate).
• Volunteer or staff coverage for bin stations during high-traffic periods.
• Customer-facing messaging.
• Data capture protocols (who measures what, when, how).

The intervention should be the version you’re considering scaling. Don’t pilot a watered-down version then scale a more aggressive one.

Step 5: Run the pilot for 90 days

Standard pilot duration is 8-13 weeks. Less than 8 weeks doesn’t give enough time for behavior to stabilize past the novelty period. More than 13 weeks usually doesn’t add new information.

During the pilot:

• Capture data weekly. Volume, contamination rate, cost variance, customer feedback, staff time, hauler reliability.
• Hold a weekly pilot review. 30-minute call with site manager, hauler rep (occasional), sustainability lead, and operations director. Surface issues, adjust if needed.
• Document deviations. When the actual operation deviates from the planned intervention, write down why and what changed. These notes are valuable when interpreting results.
• Don’t change the intervention mid-pilot unless absolutely necessary. A pilot that keeps shifting its parameters produces uninterpretable data.

Things will go wrong. Bin contamination will spike one week. The hauler will miss a pickup. A staff member will complain about the workload. Customer reaction will be mixed. This is normal pilot behavior — capture it as data rather than panicking.

Step 6: Measure results against hypothesis

At the end of 90 days, gather the data and evaluate against the original hypotheses.

Quantitative metrics:
– Did contamination rate stabilize under target?
– Did per-cover cost premium stay within budget?
– Did service times remain within acceptable range?
– Did volume of diverted compost match projections?
– Did hauler reliability match expectations?

Qualitative observations:
– Customer feedback themes (positive, negative, neutral).
– Staff feedback themes (engagement, complaints, observations).
– Operational issues encountered (and resolved or unresolved).
– Surprises — things that worked better or worse than expected.

The decision question: Based on what we observed, do we recommend scaling this program to the full chain/campus? If yes, what adjustments based on pilot learning? If no, what would have to be true to revisit?

Step 7: Make the scale decision

Three possible outcomes from a well-run pilot:

Scale immediately. Pilot exceeded targets, no major surprises, hauler infrastructure verified across the broader footprint. Roll out per planned timeline.

Scale with adjustments. Pilot mostly succeeded but identified specific issues that need addressing before full rollout. Common adjustments: better signage based on observed contamination patterns, additional staff training, hauler renegotiation, foodware spec change, larger budget reserve. Adjust the rollout plan and proceed.

Pause and rework. Pilot revealed fundamental issues that make the current program design unworkable at scale. Common pause triggers: hauler infrastructure gap at most other locations, customer behavior at scale would push contamination above acceptable threshold, cost premium too high to budget across all sites, operational complexity beyond what site managers can absorb.

The third outcome is rare but valuable when it happens. A pilot that produces “no, we shouldn’t scale this” is much cheaper than a chain-wide rollout that fails.

What pilots commonly reveal

Patterns from running compost pilots across many foodservice contexts:

Hauler reliability is more variable than expected. Pilots commonly reveal that the hauler service that looked solid on the contract has 1-2 missed pickups per month at the pilot site, suggesting similar issues at scale.

Staff training requires more time than projected. Most operators project 30 minutes of compost training per staff member; reality is 2-4 hours including follow-up reinforcement.

Customer compliance varies by location type. Office cafeterias hit 75-90% compliance with proper signage and staff support. Casual fast-food locations hit 50-70%. Outdoor venues hit 40-60%. Different rollout approaches make sense for different location types.

Cost premium runs higher than projected. Compostable foodware premiums are usually projected at 15-25% over plastic; actual realized cost premium often runs 25-40% when accounting for cooperative purchasing limitations, supplier inconsistency, and inventory carrying costs.

Contamination requires ongoing management. Initial training reduces contamination to 5-15% within 4-6 weeks; sustained low contamination requires monthly re-training and signage refresh. It’s not a one-time investment.

These patterns are less surprising once they’re observed multiple times. The pilot at your specific operation surfaces your specific operation’s version of these patterns.

Pilot pitfalls

Cherry-picking the location to ensure success. A pilot at the most engaged sustainability-friendly location proves nothing about the rest of the chain. Don’t game the pilot.

Skipping baseline data. Without baseline, results are uninterpretable. Capture before-and-after.

Adjusting the intervention mid-pilot. Tempting when something looks bad, but produces unclean data. Run the planned intervention for the planned duration; adjust afterward.

Over-investing in the pilot location. Some operators put extra staff, extra signage, and extra resources at the pilot site to ensure it succeeds. The pilot then doesn’t predict scale because scale won’t have those extras.

Under-investing in the pilot location. Conversely, treating the pilot as low-priority means it gets less management attention than the eventual scaled program would. Pilot at the same level of investment as scaled rollout.

Overinterpreting one data point. A single bad week doesn’t mean the program failed; a single great week doesn’t mean it succeeded. Trends across the 90-day period matter, not individual incidents.

Skipping the formal decision step. Pilot data needs a structured decision conversation: scale, adjust and scale, or don’t scale. Operations that just let the pilot data drift into general operational learning lose the value of the pilot methodology.

A worked example: a 12-restaurant casual chain

A regional Mexican-American chain with 12 locations across two metro areas wanted to evaluate a compost program before a chain-wide rollout. Total annual waste hauling: about $180,000. Projected compost program addition: $40,000-$70,000 annually.

They piloted at one mid-volume suburban location (350 covers/day average) for 90 days. Hypothesis: contamination under 10%, per-cover cost premium under $0.35, no measurable service time impact.

Results at day 90:
– Contamination averaged 12% across the period; trended down from 22% in week 2 to 8% in weeks 11-12.
– Per-cover cost premium averaged $0.42 — slightly above target due to higher-than-projected compostable foodware costs.
– Service times unchanged from baseline.
– Staff feedback: positive overall but flagged that signage wasn’t translating well for non-English-speaking customers.
– Hauler reliability: one missed pickup over 90 days, recovered next-day.
– Customer feedback: 80% positive, 15% neutral, 5% negative (cost-related).

Decision: scale to all 12 locations with two adjustments — bilingual signage, and renegotiated hauler rate based on the volume commitment. Projected chain-wide rollout cost: $52,000 annually, within original budget envelope.

The pilot informed the rollout plan, identified the bilingual signage gap before it became a chain-wide issue, and gave the operations team confidence in the cost projection. Total pilot cost: about $4,500 — under 3% of the projected annual program spend.

The takeaway

A 90-day single-location pilot is the highest-leverage investment in any chain-wide composting rollout. It costs single-digit percentages of the full program budget and tells you most of what you need to know to scale confidently — or to pause and rework before committing chain-wide resources.

The methodology is structured: hypothesis, baseline, intervention, measurement, decision. The discipline of running a true pilot rather than a soft launch is what makes the data trustworthy. The investment in disciplined pilot design pays off in scaled rollout reliability.

For chain operators sourcing compostable foodware to test in a pilot, the compostable food containers and compostable to-go boxes category pages provide options across the typical SKU range that pilots evaluate.

Skip the pilot, and you’re betting the chain on assumptions. Run the pilot, and you’re betting on data. The latter wins more often.

For B2B sourcing, see our compostable supplies catalog or compostable bags catalog.

Background on the underlying standards: ASTM D6400 defines the U.S. industrial-compost performance bar, EN 13432 harmonises the EU equivalent, and the FTC Green Guides govern how “compostable” can be marketed on packaging in the United States.