Formulation Brief Template With a Bench Trial Plan for Chemicals

When should you write a formulation brief before bench work?

You should write a formulation brief before bench work when catalog and commercial candidates have already failed a customer requirement and a new formulation must be designed, because the brief is what tells the laboratory what to make, how to test it and when a run passes or stops. Without it, the bench round tends to drift toward beating the last sample rather than meeting a stated threshold.

The brief starts only from a requirement that existing candidates failed in evaluation. Record which candidates were tested, which target each one missed, its measured value against the threshold, and why it fell short, named at component level where you can, such as a surfactant that foams at the customer's wash temperature or a chelant that is overwhelmed by hard dilution water. If the mechanism is unknown, write "mechanism unknown" rather than guessing, and note the risk that the target turns into a score to beat.

Write the brief before any candidate is drafted, and freeze it with a date. The candidates, the run matrix and the criteria all refer back to it, so a brief that changes during the round makes the results hard to read. A brief also helps when the request comes from sales: the customer requirement, the operating window and what will count as success are written once, in the customer's terms and units, and the laboratory works from that record instead of from a conversation. The editable Word template is at the end of this page.

What does each section of the formulation brief template cover?

The formulation brief template has 7 sections that run from the customer's unmet requirement to the open items left after the plan is written, and each section takes its inputs from the one before. The table below previews every section so the lab knows what to make, how to test it and when a run passes or stops; all example values are illustrative.

The key row is section 2: a target without a unit, a method and a quantitation limit cannot be judged at the bench. Fill the template in four passes. First, sections 1 to 3 with the customer contact and the technical manager, then date and freeze them. Second, section 4, with every candidate summing to 100 wt%. Third, sections 5 and 6, signed before batch 1. Fourth, section 7, updated as each open item is settled, without touching the frozen sections.

How do you turn an unmet customer requirement into measurable formulation targets?

You turn an unmet customer requirement into measurable formulation targets by writing each one as a threshold with a unit and a direction, tied to one test method, its quantitation limit and the customer's matrix conditions, such as dilution water, substrate and temperature. "Low foam" is a wish; "foam height at most 50 mL after 5 min at 1 % dilution and 50 °C, by the customer's cylinder test" is a target (illustrative values).

Take each threshold verbatim from its source, such as the customer specification or an agreed test report, and record the source next to it. If the customer gives no number, the target goes to the open items in section 7 as something to confirm, not into the run matrix.

Check that each target is decidable with the method you will use. Eurachem's method validation guide describes the limit of quantification as "the lowest level at which the performance is acceptable for a typical application" (Eurachem, 2025). A threshold close to that limit cannot separate a pass from a fail, so the template asks for the ratio of threshold to quantitation limit; this template flags any target where the threshold is less than 3 times the quantitation limit, a working rule of the template rather than a value from a standard. In the illustrative example, a residual soil limit of 5 mg per coupon measured with a quantitation limit of 0.5 mg gives a ratio of 10, so the target is decidable.

Which constraints belong in a formulation brief for an industrial chemical?

The constraints that belong in a formulation brief for an industrial chemical are a numeric range or a prohibition for each of three states, the in-process intermediate, storage over shelf life and in-use dilution, plus regulatory, safety and process-capability limits and the boundary of this bench round. An industrial chemical can fail in the drum or in the customer's dilution tank even when the finished-product test passes, so each state gets its own line.

For the in-process state, record limits such as the maximum batch temperature during an exothermic addition, the pH window in which a component stays dissolved, or the viscosity the plant's pumps can handle. For storage, give the temperature range and duration the product must survive without separation, crystallization or loss of active. For in-use dilution, give the dilution range and the water quality, such as hardness and pH, at which the product must still work.

Regulatory constraints name the market and the rule, such as a substance a customer or jurisdiction prohibits, and the classification the finished mixture may carry. Under the US Hazard Communication Standard, mixture makers "may rely on the information provided on the current safety data sheets of the individual ingredients", unless they know or should know the SDS misstates or omits required information (OSHA, 29 CFR 1910.1200, accessed 2026), so the ingredient SDS files belong in the brief. Process capability covers what the plant can mix, heat and fill. Write the bench boundary too: production formulas and scale-up are outside this round.

How do you lay out the bench trial run matrix in the formulation brief?

You lay out the bench trial run matrix in the formulation brief as 2 to 4 factors at 2 or 3 levels, with replicated center points, a randomized run order and the total run count written down before any batch is made. Factors are the component levels or process settings expected to move the targets.

The NIST/SEMATECH e-Handbook sets out the run counts: a two-level full factorial design in k factors has 2 to the power k runs, 4 runs for 2 factors and 8 for 3, and "when the number of factors is 5 or greater, a full factorial design requires a large number of runs and is not very efficient" (NIST/SEMATECH e-Handbook, section 5.3.3.3, accessed 2026). For center points, its worked example says the design "would be improved by adding at least 3 centerpoint runs placed at the beginning, middle and end of the experiment" (NIST/SEMATECH e-Handbook, section 5.3.3.3.2, accessed 2026). Center points also show curvature: "While a two-level design with center points cannot estimate individual pure quadratic effects, it can detect them effectively." (NIST/SEMATECH e-Handbook, section 5.3.3.6, accessed 2026). In another worked example, centerpoint runs "need to be made at regular intervals during the experiment to monitor for process drift" (NIST/SEMATECH e-Handbook, section 5.3.3.4.1, accessed 2026).

Add the procedure parameters beside the matrix: vessel scale, stirrer type and speed, addition rates, temperature control and hold times, plus a caution field for each addition filled from that raw material's SDS and your site rules. A bench result transfers only to plant conditions of comparable mixing and heat history; that is reasoning, not a sourced value. Add one run at the stressed edge of the operating window.

How do you set bench trial success and stop criteria before results?

You set bench trial success and stop criteria before results by writing, for each response variable, one method, its quantitation limit, the time point of measurement and the acceptance criterion, and for each stop criterion, an observable failure with the time point and condition at which it is checked. The person who approves the plan signs and dates the criteria before batch 1, and the template has a field for that signature.

Success criteria come straight from the targets in section 2. Each one reads as pass or fail on its own: "foam at most 50 mL at 5 min" can be read off a cylinder, while "better foam than the current product" cannot (illustrative values). Add the criterion that moves the trial from the preferred candidate to the alternative, so the next step after a failure is already agreed.

Stop criteria end a run or a candidate early. Write them as things a bench chemist can see or measure: phase separation on standing, viscosity drift on temperature cycling, loss of active over a hold time, or precipitation in the hardest dilution water. Each needs the time point and condition, such as "any visible layer after 24 h at 5 °C" (illustrative).

The rule that holds the plan together is simple: a criterion changed after results are seen opens a new design and a new dated revision of the brief. The open items in section 7 record what each outcome leads to: a second bench round, samples to the customer, or stop.

Illustrative example: a completed formulation brief and bench trial plan

The illustrative example below shows a completed formulation brief and bench trial plan for a low-foam alkaline cleaner concentrate for spray washers; every value in it is illustrative, not data from a real product or customer. In section 1, three commercial cleaners failed the customer's foam limit of 50 mL at 50 °C, with the closest at 120 mL, attributed to its surfactant foaming at that temperature (illustrative).

Section 2 holds two targets: foam at most 50 mL after 5 min at 1 % dilution by the cylinder test with a quantitation limit of 5 mL, and residual soil at most 5 mg per coupon by a gravimetric test with a quantitation limit of 0.5 mg (illustrative values). Section 3 limits batch temperature during the potassium hydroxide addition to 50 °C, requires no separation after 6 months at 5 to 40 °C, and requires performance in dilution water up to 300 mg/L hardness as calcium carbonate (illustrative values).

The center-point candidate, Candidate 1, sums to 100 wt%, with 2.0 wt% copolymer and 4.0 wt% gluconate. Candidate 2, the alternative, keeps the same composition but replaces the EO/PO block copolymer with a C12-14 alcohol ethoxylate propoxylate (CAS 68439-51-0) at 2.0 wt%, so the only difference between the two is the surfactant; the trial moves to Candidate 2 if foam exceeds 50 mL at every corner run (illustrative). The run matrix varies copolymer from 1.5 to 2.5 wt% and gluconate from 3.0 to 5.0 wt%: 4 corner runs plus 3 center points make 7 runs in random order, plus one run in 300 mg/L water at 60 °C at the stressed edge, and water takes up the difference so every run sums to 100 wt% (illustrative). The caution field for each addition points to that raw material's SDS.

When does a formulation brief and bench trial plan template not fit?

A formulation brief and bench trial plan template does not fit when a catalog product can still meet the requirement, when the work is a production formula or a scale-up, or when the change is a single raw material substitution in an existing formula. Choosing among catalog products is a selection decision, not a design, and it needs a comparison against the customer's conditions rather than a run matrix.

Production formulas, manufacturability and scale-up need plant trials, equipment data and approvals that a bench plan does not give. The template stops at bench candidates and the plan that decides among them. It also does not give a safety clearance for the whole mixture: the caution field records where each hazard is described, and the finished mixture's classification and SDS follow the rules of each market it is sold in.

A single raw material substitution starts from the function the outgoing material served and why it leaves, so it needs a substitution checklist rather than a new brief. More than 4 factors usually means the question is too broad for one round; split it, or screen first with a fractional design. Lubinpla's formulation design work fixes a measurable formulation brief first, then designs bench candidates and a bench trial plan with success and stop criteria set before results, and it does not produce a production formula or a scale-up approval. Whoever signs the brief owns the decision on what is made next.