Pilot Order vs Full Production Run: How to Test a Candy Product
15 min readShort Answer
A pilot order is useful when an approved sample proves that a candy concept can work, but the buyer still lacks evidence that the factory can reproduce it on the intended commercial route. The pilot should test the parts of the project that change when production moves from a few samples to an industrial run: the actual line, batch size, ingredient grades, forming method, coating or filling process, conditioning, packing machine, commercial packaging, in-process quality controls, coding, carton configuration and documentation.
The critical rule is representativeness. A 50 kg batch made by hand in the R&D room may be a good development trial, but it is weak evidence for an 8,000 kg launch on a continuous line. Conversely, a smaller run made on the same production and packing route can be highly valuable even when it is not sold commercially.
A pilot is not automatically required for every first order. If the factory already manufactures the exact stock SKUOne distinct sellable product unit. routinely and the buyer changes only a label or outer carton, an enhanced first-production inspection may provide more value than a separate pilot. The decision should be based on what remains unproven after sample approval.
Buyer Decision Summary
| Decision situation | Recommended route | Main risk if skipped | Evidence required before scaling |
|---|---|---|---|
| New formula, texture, coating, filling or shape | Representative line trial or pilot order | Sample cannot be repeated at scale | Same site/line/materials, defined acceptance criteria, retained samples |
| Standard factory SKU with only label/carton changes | Full run with enhanced first-batch control may be sufficient | Packaging or coding errors | Final artwork, packing-line test, first-off approval, pre-shipment inspection |
| Small, fragile, sticky or difficult pack format | Pilot including final packing equipment/material | Product passes but packing fails | Fill-weight, seal, breakage, orientation and transport checks |
| New factory or weak history with the exact SKU | Pilot plus enhanced first commercial run | Unknown execution discipline | Batch records, QCChecks confirming product meets specification. evidence, deviation handling, document readiness |
| Large launch with expensive custom printed packaging | Pilot before irreversible packaging commitment where feasible | Large unusable inventory if product or pack is wrong | Product route proven; final packaging validated before full print commitment |
Decision: Is the next step a laboratory sample, a production-line trial, a pilot order or a full commercial run?
Main variables: novelty of the formula/process, sample origin, factory experience, packaging complexity, full-run exposure, shelf-life sensitivity and the cost of irreversible materials.
Primary risk: treating a successful sample as proof of commercial repeatability.
Verification required: identify every element that remains unvalidated and design the smallest representative test that produces evidence for those elements.
1. A Pilot Order Is an Evidence Tool, Not Just a Small Order
Buyers often use "pilot," "trial batch," "sample run" and "first order" as if they were the same thing. They are not. The useful distinction is what the run is intended to prove.
| Stage | Main purpose | What it can prove | What it normally cannot prove |
|---|---|---|---|
| Laboratory / bench sample | Develop the concept | Direction of flavor, texture, color, shape | Industrial repeatability, packing speed, long-run drift |
| Production-line trial | Test selected production steps | Line compatibility, forming, coating, filling, first-off settings | Full packing/logistics chain unless included |
| Pilot order | Test the integrated commercial route at controlled exposure | Production + packing + QC + documents + selected logistics | Full-scale duration effects and future batch consistency |
| First full commercial run | Produce saleable launch quantity | Real commercial execution | Long-term supplier performance across multiple runs |
| Routine reorder | Repeat an established controlled process | Ongoing consistency against frozen baseline | New changes unless separately validated |
A pilot is therefore defined by its purpose and representativeness, not by an arbitrary weight. A "pilot" that uses different ingredients, different equipment and neutral hand packing may prove much less than a larger line trial that uses the future commercial route.
Buyer takeaway: Ask the factory to describe exactly which commercial steps the pilot will reproduce and which will remain different. Any difference should be treated as an open validation item.
2. When a Pilot Order Adds Real Value
A pilot is most valuable when one or more critical variables remain uncertain after sample approval.
Product-Side Uncertainty
- New gelatin/pectin/starch system or unusual sweetener architecture.
- New sourness level, coating load or anti-stick treatment.
- Filled, layered, peelable or multi-component gummy.
- New piece geometry, logo mold, thin belt, rope or fragile structure.
- Tight piece-weight or dimension tolerance.
- Color/flavor combinations that may behave differently after heating or drying.
Process-Side Uncertainty
- R&D sample was handmade or made on a small depositor.
- Commercial line has not previously run the exact configuration.
- Product requires new extrusion/depositing, rolling, cutting, filling, coating, drying or conditioning settings.
- Factory must move equipment, use manual operations or combine steps that are normally on different lines.
Packaging-Side Uncertainty
- Very small retail packs or low target net weight.
- New film width, sealant layer, zipper, individual wrapper or flowpack.
- Fragile pieces that can break or deform during filling.
- Sour powder, oil or fine particles that may contaminate seals.
- Custom presentation requiring a fixed orientation or count.
- High-barrier pack required for moisture-sensitive products.
Commercial Exposure
A pilot can also be justified when the technical product is not especially novel but the consequences of a first-run failure are large: expensive printed packaging, a high launch quantity, strict retailer onboarding, multiple SKUs or a narrow launch window.
3. When a Separate Pilot May Be Unnecessary
A pilot is not a ritual. It costs money, uses capacity and can delay launch. It should not be inserted merely because the project is new to the buyer.
A buyer may reasonably move directly to a full run when:
- the product is an active stock SKU produced routinely on the same site and line;
- the buyer approves the existing formula, shape and finish without modification;
- the selected pack format is already standard on the factory's line;
- the change is limited to a low-risk label or master-carton artwork;
- the first full order is commercially manageable;
- strong batch records and recent comparable production evidence are available;
- enhanced first-batch QC and pre-shipment inspection can close the remaining risks.
The mistake is not "skipping the pilot." The mistake is scaling before identifying what has and has not been proven.
4. The Pilot Must Be Representative of the Full Production Route
| Control element | Pilot should match full run where material | Why it matters |
|---|---|---|
| Manufacturing site | Same site | Different site can mean different licence, equipment, controls and environment |
| Production line | Same line or documented equivalent | Small R&D equipment can hide scale-up problems |
| Formula revision | Same controlled specification | Prevents "pilot passed, formula later changed" |
| Ingredient grade/source | Same approved commercial materials | Gelatin, pectin, colors, flavors and acids can change process behavior |
| Mold/die/forming | Same production tooling | Geometry affects weight, drying, coating and deformation |
| Drying/conditioning | Same route and target | Texture and surface stability often develop after forming |
| Filling/layers/coating | Same commercial equipment and target | Distribution and adhesion can change at line speed |
| Packing machine | Same machine or technically equivalent | Fill accuracy, seal contamination and breakage are machine-dependent |
| Packaging material | Final material where the pack itself is a risk | Barrier, friction, sealing and machinability differ |
| QC plan | Same critical checks | Pilot data is only useful if it measures the future release criteria |
| Coding/cartons | Representative configuration | Reveals artwork, traceability, count and handling errors |
A representative pilot does not mean every commercial detail must be final. It means each deliberate difference is documented and the buyer knows what evidence is still missing.
5. Product Validation: What to Measure During the Pilot
The pilot should begin with a frozen product specification and an approved sample reference. "Same as sample" without a sample ID and written criteria is not a release standard.
Sensory and Physical Checks
Evaluate at least:
- appearance and color;
- shape and dimensions;
- unit weight and variation;
- aroma and off-odors;
- flavor identity and intensity;
- sweetness and acidity profile;
- texture: firmness, chew, rebound, fracture and tooth stick;
- coating coverage and loose powder;
- filling ratio, leakage and layer alignment where relevant;
- surface oil, tack, sweating or crystallization;
- deformation and breakage after conditioning and packing.
In-Process Evidence
The factory should record the parameters that materially control the product. These differ by process but may include cooking/depositing temperatures, solids/Brix, pH, cooling, drying or conditioning time, piece-weight checks, coating settings and packing checks. The buyer does not need confidential recipe percentages to require evidence that the approved process was controlled.
Start / Stable Run / End Sampling
Do not evaluate only the best pieces taken after the line is tuned. Retain samples from the start, stable middle and end of the run where practical. This helps reveal drift in weight, coating, color, filling, texture or seal performance.
6. Packaging Must Be Validated as a Separate Workstream
A candy can pass the product trial and still fail commercially because the pack does not run well or does not protect the product.
| Pilot packaging route | What it proves | What remains open |
|---|---|---|
| Stock pouch + compliant label | Product sell-through and basic filling/handling | Final printed film machinability, exact barrier and artwork print quality |
| Representative unprinted film on final machine | Film dimensions, friction, forming and seals | Final ink/lamination/print registration if structure changes |
| Short-run digital print | Artwork, dimensions and near-commercial presentation | Performance may differ if full run uses another print/conversion process |
| Final commercial printed packaging | Strongest evidence for line and pack | Future material/batch variation still requires incoming control |
For small packs, flowpacks, powder-coated products and fragile candy, include real line speed, fill-weight control, seal integrity, code legibility, product-in-seal checks, pouch dimensions and carton loading. A hand-packed pilot cannot validate these points.
AXTIMES Operational Insight — Amanda XUN, Head of Sourcing, AXTIMES: In candy development projects, we repeatedly see a gap between "the sample is approved" and "the line is ready to reproduce it." For products with strong sour coating, custom geometry, filling or sensitive packaging, the highest-value control is often not another courier sample. It is a short, representative production event with agreed acceptance criteria, retained samples and a written decision on what still remains unvalidated.
7. How Large Should a Pilot Order Be?
There is no universal kilogram or carton number. The correct pilot quantity is the smallest amount that allows the critical process to reach a normal operating state and produces enough units to evaluate the intended controls.
Ask the factory for three different numbers:
- Minimum representative production batch — the smallest batch that can run through the intended commercial process without becoming a laboratory simulation.
- Minimum packaging/printing commitment — the smallest film, pouch, label, wrapper or carton purchase required for the chosen packaging route.
- Minimum saleable finished quantity — the smallest quantity the factory is willing to finish, inspect, carton and release commercially.
These are often different. A factory may need to cook a larger batch than the buyer wants to sell; packaging may create another minimum; unused candy or packaging needs an agreed disposition.
What changes the answer: Pilot size changes with line throughput, cooking-vessel size, mold capacity, dryer load, coating equipment, conditioning time, pack speed, number of SKUs, changeovers, destructive testing, shipping-test needs and whether the pilot must be commercially saleable.
8. Define Pass/Fail Criteria Before the Run
A pilot without pre-agreed acceptance criteria becomes a negotiation after the fact. The buyer and factory should approve the scorecard before materials are committed.
Typical criteria include:
- approved sample and specification version;
- piece-weight and dimensional tolerance;
- sensory acceptance method and panel responsibility;
- critical, major and minor defect definitions;
- coating/filling/layer acceptance;
- pack net weight and seal criteria;
- coding and artwork checks;
- carton count and configuration;
- laboratory or food-safety evidence where required;
- retained-sample quantity and storage;
- disposition of nonconforming pilot goods;
- decision rule: pass, conditional pass, rework/retest or fail.
A good pilot report does not merely say "approved." It states what passed, what deviated, what was corrected and what remains unvalidated.
9. Pilot Evidence Pack
At the end of the run, the buyer should receive a compact evidence file.
| Evidence item | Why the Buyer needs it |
|---|---|
| Pilot run date, site and line | Identifies the production event |
| Product/SKU and spec revision | Links results to the controlled product |
| Approved sample reference | Creates a sensory benchmark |
| Batch/lot identifier | Supports traceability |
| Commercial raw-material confirmation | Confirms the intended material set was used |
| Key in-process QC records | Shows the process was controlled, not merely visually acceptable |
| Finished-product test/inspection results | Supports pass/fail decision |
| Start/middle/end retained samples | Provides evidence for later comparison |
| Packing-line data and seal/weight checks | Proves pack execution |
| Photos/video of representative line stages | Documents route and configuration |
| Deviation/CAPA record | Prevents hidden changes from becoming the new normal |
| Final decision and open items | Defines what may proceed to full run |
10. Worked Buyer Examples
Example 1 — Extra-Sour Belts with Custom Presentation
Illustrative numbers only; not industry benchmarks.
A buyer plans an 8,000 kg commercial launch. The approved courier sample confirms flavor and base texture, but the final acid coating and rolling method require production-line setup. The factory states that 600 kg is the minimum representative run on the intended extrusion/coating route. Custom printed film, however, has a separate 60,000-pouch minimum per design.
- Inputs: 8,000 kg planned launch; 600 kg representative line run; high-sour coating; custom presentation; high printed-film commitment.
- Pilot design: run 600 kg on the intended extrusion, coating, rolling/cutting and conditioning route; take start/middle/end samples; pack most pilot goods in neutral compliant pouches; run a smaller representative film test on the final packing machine if technically possible.
- Constraint: neutral pouches do not prove final printed rollstock performance.
- Buyer decision: approve the product process only if sensory/coating/geometry criteria pass. Keep final commercial packaging as a separate release gate before committing the full print run.
Example 2 — Standard Gummy with Only a Label Change
Illustrative example.
The factory produces the same gelatin gummy every week using an established formula, mold and stock pouch. The buyer changes only the label and master-carton graphics.
- Inputs: established SKU; same site/line; same product and primary pack; low technical novelty.
- Comparison: a dedicated pilot would mainly reproduce a process the factory already controls.
- Buyer decision: skip a separate production pilot. Approve physical samples and artwork, require a first-off label/carton check, enhanced first-batch QC and pre-shipment inspection.
Example 3 — Freeze-Dried Fragile Candy in a 25 g High-Barrier Pouch
Illustrative example.
The product exits the dryer correctly, but the commercial concern is breakage and moisture pickup during filling and sealing.
- Inputs: fragile pieces; moisture-sensitive product; new 25 g high-barrier pouch.
- Pilot design: use the intended dryer route, transfer method, packing machine and final or technically equivalent high-barrier material; measure breakage before and after packing, net weight, seal integrity and retained-sample stability.
- Buyer decision: a product-only pilot is insufficient. Packaging is a critical part of the validation.
11. Failure Scenarios
| Initial assumption | What was overlooked | Consequence | Correct control |
|---|---|---|---|
| "The handmade sample is perfect, so mass production will match." | Commercial line shear, heat, residence time and throughput | Texture/color/shape changes at scale | Representative line run before scaling |
| "A 100 kg pilot passed, so the 10 t run is fully proven." | Long-run drift and multi-shift variation | Later cartons differ from first-off samples | Enhanced start/middle/end controls in first full run |
| "Neutral bags are fine for the pilot, therefore packaging is approved." | Final film friction, sealing, barrier and registration | Jams, weak seals or shelf-life risk after printing | Separate final packaging validation |
| "Factory says the pilot is same formula." | Different ingredient grade/source used for R&D | Commercial texture or claim differs | Commercial material-set confirmation |
| "Pilot looked good on production day." | Product had not completed conditioning/equilibration | Stickiness, sweating or deformation appears later | Recheck after defined conditioning and retained-sample interval |
| "Pilot quantity is small, so inspection can be informal." | No predefined criteria or evidence | Buyer and factory disagree on whether it passed | Written scorecard and decision rule before run |
12. What a Successful Pilot Does Not Prove
A pilot reduces uncertainty; it does not eliminate it. It may not fully prove:
- performance during a very long uninterrupted production campaign;
- variation across shifts, operators or raw-material lots;
- seasonal heat/humidity effects;
- every pallet's transport exposure;
- full shelf lifePeriod product remains within specification. unless an appropriate stability program is completed;
- future supplier discipline;
- destination-market regulatory compliance unless that work is separately completed.
This is why the first full production run should still receive enhanced monitoring even after a successful pilot.
13. Pilot Economics: Test the Cost of Failure, Not Only Unit Price
Pilot unit cost is often higher. Setup, cleaning, QC, short packaging runs, manual handling and inspection are spread over fewer saleable units. That does not automatically make the pilot uneconomic.
Compare: Pilot cost + remaining open risk versus expected loss if the full run fails.
The full-run failure exposure may include product, unused printed packaging, rework, inspection, delayed launch, freight, retailer penalties, write-offs and reputational cost. The calculation should also separate one-time costs from recurring unit economics.
A pilot becomes a false economy when it is so small or so unrepresentative that it does not generate useful evidence.
14. Buyer Control Table
| Control item | Owner | Timing | Required evidence |
|---|---|---|---|
| Pilot objective and open risks | Buyer + Export Company | Before quotation | Written pilot brief |
| Factory/site/line identity | Export Company + Factory | Before deposit | Site/line confirmation |
| Approved sample/spec revision | Buyer + Factory | Before materials | Coded sample + controlled spec |
| Commercial materials | Factory | Before run | Material/ingredient confirmation |
| Production settings and in-process QC | Factory | During run | Batch/QC records |
| Sensory approval | Buyer / trained representative | Stable run + post-conditioning | Scorecard + retained samples |
| Packaging validation | Factory + Export Company | During packing | Weight/seal/code/line data |
| Deviations and CAPA | Factory + Buyer | Before release | Written deviation/CAPA record |
| Pilot disposition | Buyer + Factory | After review | Pass/conditional pass/fail decision |
| Full-run release gate | Buyer | Before scale-up | Closed open-item list |
15. Questions to Ask the Factory
- Was the approved sample made on the same site and line proposed for mass production?
- Which attributes of the sample are already proven on the commercial line?
- Which attributes require a line trial or pilot?
- What is the minimum representative production batch for this exact process?
- Which ingredients or grades will differ between sample, pilot and commercial run?
- Will the same mold, die, depositor, extruder, coater, dryer and conditioner be used?
- Which steps are manual in the pilot and which will be automated in full production?
- What must be changed or scaled when the order becomes larger?
- Which in-process parameters will be recorded?
- Can start, stable-run and end samples be retained?
- Can the final packing machine be used during the pilot?
- What packaging minimums are separate from the candy minimum?
- What happens to unused candy, ingredients and packaging?
- What is the CAPA/retest route if one criterion fails?
- What evidence will be provided before the buyer authorizes the full run?
16. Questions the Buyer Should Answer Internally
- Which failure would create the largest commercial loss?
- What exactly did the approved sample prove, and what did it not prove?
- Is the primary uncertainty product, process, packaging, supplier execution or market demand?
- Which packaging/tooling costs become irreversible before the full run?
- Does the pilot need to be saleable, or is technical evidence the only purpose?
- Which defects are unacceptable regardless of price or schedule?
- Who has authority to approve sensory criteria and deviations?
- What evidence is required before paying for full production?
- Which changes after the pilot would trigger a new validation?
- How much first-full-run monitoring is needed even after a pass?
17. Implementation Workflow
- Define the commercial full-run configuration.
- List every assumption that is already proven by evidence.
- List the remaining technical, packaging and execution risks.
- Decide whether a lab sample, line trial, pilot or direct full run closes those risks most efficiently.
- Freeze the sample reference and product specification.
- Map pilot vs full-run differences.
- Agree quantity, materials, packaging route, cost and unused-material policy.
- Define pass/fail criteria and evidence before production.
- Run the pilot with start/middle/end sampling and representative packing.
- Review results after required conditioning, testing and retained-sample checks.
- Close deviations and update the controlled specification/process where justified.
- Authorize the full run only after the open-item list is explicitly closed or accepted.
18. Pilot-to-Full-Run Buyer Checklist
- ☐ Full commercial SKU and pack configuration are defined.
- ☐ Approved sample has an ID and retained reference.
- ☐ Product specification has a controlled revision.
- ☐ Pilot objective is written.
- ☐ Open risks are listed before the run.
- ☐ Same manufacturing site is confirmed.
- ☐ Same commercial production line/equipment is confirmed where material.
- ☐ Commercial ingredient grades/sources are confirmed.
- ☐ Mold/die/forming route is representative.
- ☐ Drying/conditioning/coating/filling route is representative.
- ☐ Pilot quantity is large enough to reach normal operating conditions.
- ☐ Final packing equipment is included where packaging is a risk.
- ☐ Packaging differences are documented as open items.
- ☐ In-process QC and finished-product checks are predefined.
- ☐ Sensory evaluator and decision rule are defined.
- ☐ Start/middle/end retained samples are planned where useful.
- ☐ Deviations/CAPA are documented.
- ☐ Unused product/material/packaging ownership is agreed.
- ☐ Pilot decision is recorded as pass, conditional pass, retest or fail.
- ☐ First full run has enhanced monitoring planned.
19. Important Limitations
A pilot does not replace food-safety certification, site due diligenceStructured checks before choosing a supplier., mandatory testing or label/legal checks for the destination market, shelf-life validation or pre-shipment inspection. These tools answer different questions.
For established stock candy, the best "pilot" may be a smaller commercial order from a routine production run. For a very sensitive new formula, a pilot may be premature before laboratory development is stable.
If the full commercial configuration cannot be genuinely reproduced at small scale, do not present a small run as proof. Use staged validation: product route first, then packaging/line interface, then enhanced controls on the first full run.
20. Sources and Verification Notes
- IFS Food Version 8, Section 4.3: controlled product/process development, including factory trials, product testing, process monitoring and recorded results.
- Codex Alimentarius, General Principles of Food Hygiene (CXC 1-1969, 2022).
- Codex Alimentarius, Guidelines for the Validation of Food Safety Control Measures (CXG 69-2008).
- AXTIMES operational observations on stock/lab samples not automatically proving line coating, geometry, rolling, filling and packing behaviour; packaging capability validated separately; product MOQMinimum quantity accepted per order. and packaging/printing MOQ as practical constraints on pilot design.