Picture a mid-size toy retailer preparing a building blocks wholesale range for the coming season. The buying team has 40 candidate lines from four Chenghai suppliers, three price tiers, and two conflicting opinions about which sets belong on the shelf. Nothing is wrong with the products — no two people are comparing the same attributes. A spec table fixes that. It is not a spreadsheet for its own sake: it is the artifact that converts a shouting match about SKUs into written, reviewable decision rules, and the same artifact you can hand to a factory when you ask for a sample or a reorder.
The scenario below is hypothetical — no real company, no invented order values, no claimed results. It shows how the table is built, which columns carry the weight, and how the same discipline applies when you evaluate a Chinese building-block factory in Chenghai, Shantou, or Yiwu. The measurable handoff is the point: every row states a SKU's job, where it sells, and what metric will decide whether it stays.
Key Takeaways
- A spec table works when each row carries a role (core, module, local, online-only), a price tier, a coverage rule, and one success metric — not just a description and a cost.
- Sales velocity alone hides substitution behavior: shoppers often buy their second choice. A table that records what the shopper originally sought separates a genuine range gap from a SKU that only looks healthy.
- Cut, keep, add, and test decisions should be written as rules before the meeting, so the debate is about evidence rather than about whose SKU gets protected.
- When you move from assortment to sourcing, the same table becomes a supplier brief: mold/tool number and revision, approved sample identity, and certificate scope mapped to market and lot.
- A photo is not an approved standard. Production release should reference an identified approved sample, especially for color, texture, logo, and assembly.
What Columns Actually Reduce Assortment Ambiguity?
The columns that earn their space are the ones that force a decision. Start with SKU role: national core (broad availability, protected), regional module (defined purpose and approved list), local variant (used sparingly because it multiplies complexity), and online-only. That single field stops the most common argument, because a core SKU and a local variant should never be judged by the same standard.
Next, price tier and coverage rule. A range needs a credible opening price point and a trade-up tier; if every row sits in the middle, the shelf has no story. Then the decision fields: gap statement (what need is unmet), evidence, expected role, trade-off, and success metric. Finally, the operational fields — substitution relationship, pack-fit dimensions, and true landed cost rather than purchase price alone. A table missing landed cost will quietly reward the wrong SKU.
Keep it to one screen per category if you can. A 60-column table gets abandoned by week three.
Decision Rules at a Glance
| Keep — economic | Margin dollars and turns justify the slot; no distinct-need test required |
|---|---|
| Keep — customer | Removing it would visibly disappoint a defined shopper segment; substitution risk is documented |
| Keep — architecture | It anchors a price tier or completes a good-better-best ladder |
| Cut — redundant | Serves nearly the same need as another row; the table shows the overlap explicitly |
| Cut — low rotation | Low turns combined with high inventory value or markdown risk |
| Cut — operationally costly | Special handling, fragile packaging, poor case-pack fit, or repeated supplier service issues |
| Add — gap | A written gap statement plus evidence, expected role, trade-off, and success metric |
| Test — pilot | Hypothesis, control group, success threshold, decision date, and scale rule defined before launch |
| Substitute — map first | Record where demand transfers (trade up, trade down, private label, online, delay, or leave) before delisting |
How Do You Decide Cut, Keep, Add, and Test Without Politics?
Write the rules first, then apply them. A useful five-step check runs: does the SKU fit the category role, how does it perform against comparable rows, does it serve a distinct need, is its complexity justified, and how strong is the evidence. Answer those five and the decision usually makes itself. Where the answer is unclear, the action is test, not cut.
Redundancy is the biggest hidden cost. It shows up as near-identical pack counts, overlapping technical specifications, or several rows at the same price tier. The test is not whether a SKU sells — it is whether the assortment would lose meaningful demand if the row disappeared. That question is answerable only if the table already records substitution relationships.
One caution from published category research: shoppers who cannot find their first choice often buy a substitute, so a healthy-looking SKU may simply be absorbing demand from a missing one. A regional grocery study found 23% of performing SKUs were substitution purchases, and stockouts cost retailers an average of about 4% in sales. Those figures come from that external research, not from the hypothetical retailer here — but they explain why velocity alone is a weak filter.
For additions, apply one-in-one-out only where it fits. Forcing a removal for every new line can strip a tier that still earns its place.
The Information Handoff: Who Fills Which Field
The table only works if the handoff is defined. Merchants own role, price tier, and gap statement. Finance owns true landed cost and margin dollars. Supply chain owns lead time, pack fit, and substitution mapping. The digital team owns online-only eligibility and content readiness. Each field has one owner and a review date.
When a row moves from add to test, the table should already contain the hypothesis and the threshold that will end the test. That is the difference between an assortment review and a monthly argument: the decision criteria were frozen before the data arrived.
The table is a communication device. Its value is not the spreadsheet; it is that four functions wrote down what they mean by good.
Chenghai, Shantou, Yiwu: Which Origin Fits Which Line?
Chenghai, inside Shantou in Guangdong, is the reference point for toy production in China; buyers treat it as the default origin for plastic and building-block lines because tooling, molding, and component suppliers sit within a short drive of each other. That density matters when you need a mold adjusted or a color re-run — the advantage is proximity, not a slogan.
Shantou more broadly covers packaging, printing, and a wider mix of toy categories, which helps when a range includes both plastic sets and plush or novelty items from one consolidation point. Yiwu, in Zhejiang, is the general-merchandise and small-commodity hub: strong for mixed containers, party goods, and low-complexity items, less suited to precision injection work than Chenghai. Choose the origin by the line, not by habit.
None of this replaces verification. A regional cluster tells you where to look; it does not tell you which factory to sign.
How Do You Verify a Chinese Building-Block Factory?
Ask for the tool number and revision, not a product photo. A booth sample at a trade fair is often a hero unit under flattering light, and a fair photo should never be accepted as the production standard. Before release, the reference should be an identified approved sample — a sealed golden sample on the QC shelf — compared on color, fit, flash, and function. Buyers who approve only a photo tend to discover the difference on the second lot.
Map the paperwork to the market and the lot. A UKCA mark and a VOC file are different documents covering different requirements; one certificate does not cover both, and paperwork that does not match the lot is not paperwork. Ask which file covers which test, for which market, and for which production run.
Ask for the inspection plan, not a catalogue AQL number. On the floor, a failed lot should be isolated, the defect class recorded, and release given only after rework. The AQL figure belongs in the agreed inspection plan on the PO. If a supplier quotes you a number from a brochure instead of from your PO, that is a signal about how the rest of the relationship will run.
Freeze the tooling terms before you pay a fee: who owns the tool, how long it is stored, and what happens if it sits idle. And ask how the factory handles process changes — a change in resin, mold, coating, component, or sub-supplier should be notified and risk-assessed before it enters a released product.
Traceability is the last check. A useful record links the SKU, mold cavity, resin lot, component lots, production date, operator or process, and carton or pallet identifier. If a factory cannot describe that chain, the approved sample has nothing holding it in place.
Factory Evaluation Checklist for a Building-Block Range
| Tool identity | Tool/mold number and revision recorded on the sample and the PO |
|---|---|
| Approved standard | Sealed physical sample; photo approval not accepted as production standard |
| Certificate scope | Mapped to destination market, test type, and production lot |
| Inspection plan | Defect classes and AQL agreed in the PO, not quoted from a catalogue |
| Process change notice | Resin, mold, coating, component or sub-supplier changes notified before release |
| Traceability | SKU, mold cavity, resin lot, component lots, production date, operator/process, carton or pallet ID |
| Tooling terms | Fee, ownership, storage period, and idle-tool conditions in writing |
| Molding consistency | Documented process development and monitoring; balanced runners to reduce cavity-to-cavity variation |
FAQ
What MOQ should I expect when sourcing building blocks from Chenghai?
MOQ varies by spec — it depends on whether the item is an existing tool or a new mold, the color and packaging configuration, and the factory's own production planning. Do not treat piece count in a set as the MOQ; those are different numbers. Ask for the MOQ in writing against the specific SKU and packaging version you want.
Can I approve a sample from a trade fair photo instead of a physical sample?
No. A fair booth sample is often a hero unit, and lighting distorts color. Production release should use an identified approved sample, compared on color, fit, flash, and function, especially for color, texture, logo, and assembly.
Which certificate matters for selling building blocks in the EU or UK?
It depends on the market and the product. UKCA and a VOC file are different documents covering different requirements, and one certificate does not cover both. Ask the factory to map each file to the destination market, the test type, and the specific lot, because paperwork that does not match the lot does not protect you.
How do I compare two factories that quote similar prices?
Compare the tooling and release system, not the price sheet. Ask for tool number and revision, the approved sample identity, the inspection plan on the PO, and the traceability record linking mold cavity, resin lot, and carton ID. A factory that can describe these is usually cheaper over a full season than one that only quotes lower.
Should I cut SKUs before negotiating with my building-block supplier?
Decide the assortment first, then negotiate. Cutting duplication before cutting coverage gives you a measurable way to simplify without removing entire needs or price tiers. Supplier terms should follow the assortment strategy, not dictate it.
What happens to my tool if I stop ordering for a season?
That must be written before you pay. A tooling policy should state the fee, ownership, storage period, and what happens if the tool sits idle. A handshake is not a policy — the contract clause is.
Sources
Related Products & Categories
Send a Spec, Not a Wish List
If you are building a building-block range and want a supplier who can answer the table above — tool number, approved sample, inspection plan, and tooling terms — send your spec and target market. BlocksBulk works from Chenghai and can quote against a defined SKU, packaging version, and destination market, or support a factory audit before you commit. A clear spec gets a clear answer.
Turn the note into a buying brief
Send the target market, quantity, packaging preference and destination. BlocksBulk can return a focused shortlist with the questions that still need confirmation.