What has to be in the outsole brief before it goes to a Chinese OEM? Terrain first, with lug depth and lug spacing hanging off the ground the shoe is built for — until those rows exist, a sample cannot be reviewed at all. A brief that says 'aggressive outsole, trail' and stops there comes back as a sample somebody has to argue about in a meeting, and arguments are rework.
This is written for the sourcing and QC side of a US program: the people who draft the brief, send it to candidate factories, and sign off on the first pairs. The frame is an operating one, and the currency is sample rounds and calendar days. A spec that cannot fail is not a lenient spec; it is a spec that tells you nothing, and it will not stop a weak sample from reaching your desk.
Start the Brief with Terrain, Not Lug Size
Open the matrix with the ground. Terrain is the first column, and lug depth and lug spacing sit underneath the terrain row they belong to rather than in a shared column above the whole table. Two more columns follow: what a sample has to clear before you approve it, and what the candidate has to confirm in writing. Four columns, one row per ground condition, and the rest of the conversation has somewhere to land.
Brand selection guides hand you the logic to borrow. A stored trail-running guide separates unstable or muddy ground from hard-packed, dry trails as two different tread problems, then describes the lug geometry each one calls for. That separation is the useful part. It is an application reference for how terrain maps to tread, not evidence that any factory in China can build either outsole, and treating it as the second thing is how sourcing briefs go wrong.
Writing one lug spec to cover every trail condition is the expensive shortcut. It reads clean on a one-page RFQ and it costs you a sample round, because the first pair that comes back cannot be judged — deep enough for mud or too deep for hardpack, and nobody wrote down which of those you were buying.
So take the decision in front of you and make it explicit. If your program sells mostly into mud, soft ground and wet roots, go with the mud row as the primary approval target and treat hardpack as a secondary row. If most of your volume runs on dry, hard-packed trail, go with the hardpack row first and let the mud version sit as a variant. The trade-off is a longer brief to fill in and more supplier questions landing in your inbox in week one, in exchange for specs that are conditional on named ground instead of vaguely aspirational.
Set Separate Depth Targets for Mud and Hardpack
Depth belongs in two rows, not in one number. The stored guide is direct about direction: deeper lugs give better traction in muddy terrain, while shorter lugs are enough on hard-packed, dry trails. Those two statements pull against each other, which is exactly why a single depth figure cannot serve both.
What the guide does not give you is a number. It talks about deeper and shorter without units, so any millimeter value in your brief is a supplier proposal or your own design intent, not something carried over from an original source. Ask each candidate to propose a depth per terrain row, say how it will be measured, and name the point on the lug where the measurement is taken. Then confirm it against the physical sample rather than the drawing.
The improvement is small and it pays on the first round: instead of one depth line, the brief carries a proposed depth, a measurement method and a measurement point for each terrain row. When the sample arrives, QC compares a recorded number against the row it was built for. A photo of a lug pattern can be argued about; a row number cannot.
Which row leads depends on your sell-through. If mud and soft ground dominate the line, go with the deeper-lug row as the approval target and accept that the same outsole will give up some stability and efficiency on hardpack. If dry hardpack dominates, skip the mud-first depth target, make the shorter row primary, and treat the mud version as a separate SKU. The trade-off runs in one direction only: you buy mud traction with hardpack stability and efficiency, and no single depth buys both.
Write Spacing Rules for Mud Shedding and Hardpack Stability
Spacing is where mud briefs usually fall apart, because it is easy to specify depth and forget the gaps. The stored guide pairs the two. Unstable or muddy terrain calls for a more aggressive tread with deep and widely spaced lugs so mud does not get trapped in the tread, while shorter, more closely spaced lugs are enough on hard-packed, dry trails.
Read those together and the rule for the brief writes itself. Spacing lives in the same terrain row as depth, and the two get judged as a pair. A supplier who proposes deep lugs at tight spacing for a mud row has changed the design intent, and the brief should require that substitution to be flagged instead of quietly absorbed into a quotation.
Add a short note to each row about what the geometry is for — mud shedding on one row, stability and a clean contact patch on the other — and QC gains a sentence it can put in a rejection. A sample that visibly packs mud between the lugs fails the mud row on spacing, without anyone having to argue about whether the lugs looked aggressive enough in the photo.
Use the sample, not the drawing, to decide. If the first pair traps mud between lugs on the soft ground that row was written for, skip approval for the mud row and send the spacing note back. If the lugs come back short and close-set, go with the hardpack row only and stop calling it an all-terrain outsole. The trade-off deserves saying out loud: one outsole geometry is unlikely to be optimal for both rows, so a two-row brief usually resolves into two SKUs or one deliberately compromised design.
Turn the Lug Matrix into a Sample Approval Gate
When a sample lands, review it against the row it was built for, not against the word trail on the shoebox. Depth and spacing get recorded as supplier-confirmed measurements, compared with the agreed row, and marked pass or fail per terrain. Each row gets its own verdict, and the second one is not optional paperwork.
Be clear about what this gate is. It is a review against an agreed specification, not a performance test. Depth and spacing measurements do not tell you how long a compound lasts, how the outsole behaves on wet rock, or how well it sheds mud over a long run. Those questions need physical testing you have not run yet, and no selection guide answers them for you.
Suppose a sample arrives marked for trail use with deep lugs and tight spacing. The mud row fails on spacing even though the depth looks right, while the hardpack row may pass if its short-and-close requirement is met by the same geometry. The same pair fails one row and meets the other, and the disagreement is now specific enough to send back to the factory in a single message.
Then decide what you own. If the sample matches the mud row but not the hardpack row, approve it for mud and commission a separate hardpack sample rather than stretching the approval across both. If it matches neither row, skip the all-terrain claim and rewrite the brief before the next round. The trade-off here is administrative: more rows, more sample rounds, more SKU paperwork, against an approval you can defend line by line when a buyer asks what was actually verified.
Put Supplier Confirmation Beside Every Spec
Every row needs a confirmation column, because a brief is a set of questions until a candidate answers them in writing. The stored guide describes what terrain calls for. It says nothing about any factory's capability, tooling or material stock, so capability questions belong in the document rather than in an assumption you carry into the meeting.
What goes in that column: proposed lug depth and spacing per terrain row, the measurement method and measurement point, the outsole compound and the tooling route, MOQ, lead time, certification status, and the criteria the factory expects the sample to be graded against. All of those are open questions on this program. No supplier quote, MOQ figure, capacity audit, lead-time commitment or certification report is in hand, so none of them can be written into the brief as a fact.
Require a written reply against each row rather than a general quotation. The effect is that a quote cannot quietly assume capability nobody checked: if the factory's number sits next to your mud row with its spacing and measurement method named, you can see what was priced and what was guessed.
Screen on the reply. If a candidate will not put the measurement method and the tooling route in writing, skip it for this program; a quotation you cannot audit is not cheaper, it just moves the cost to the sample round. If it confirms those points but has no sample evidence yet, go with a conditional quote tied to the sample gate instead of a full commitment.
A terrain-first brief buys you samples and supplier replies you can judge against named ground, and it makes each failure point visible at the row where it happened. What you give up is the tidiness of one universal outsole spec: more rows to fill, more questions coming back from candidates, more sample rounds, and every approval narrowed to the terrain you actually verified. If a single all-trail line is all your calendar can carry this season, that is a legitimate choice — just write down which ground you are gambling on.
