aisle/tools/Find a supplier or 3PL - agent workflow | AISLE
free to read Buy 10 runs →

Find a supplier, 3PL or sample · workflow

Find a supplier or 3PL

Quote three fulfilment routes, order a sample and compare the real SLA.

agent Ops agenttime 2-5 days, quotes includedsteps 18

Inputs

ask once, then run

keytypewhat it is
sku requiredstring: sku, dimensions and weightSKU, dimensions and weight
volume requirednumber: orders per monthOrders per month
e.g. 300
markets requiredlist: ship-to countriesShip-to markets
e.g. US, CA

Run it

18 steps

1

Spec the SKU

SKU SHEET

Record dimensions, weight, pack count, fragility, battery or liquid status, and the barcode. Every quote needs the same sheet.

returnsSKU sheet with dims, weight, handling flags

2

Shortlist three routes

SHIPBOB / SHIPHERO / PRINTFUL / GELATO / PRINTIFY / SHIPSTATION

Pick three that actually serve the ship-to markets. Printful, Gelato or Printify only when the product is print-on-demand. ShipStation or a free Pirate Ship account when you self-fulfil. Note the onboarding minimum.

returnsthree providers, service model, market coverage

3

Quote the same volume

QUOTE REQUEST

Ask each provider for pick-and-pack, storage, inbound receiving, returns and the surcharge list. Record the quote date and the assumptions.

returnsquote table with every fee line

4

Run a test order

TEST ORDER

Send one unit through each shortlisted provider. Time the receipt, inspect the packaging and photograph the arrival condition.

returnsdays to deliver, packaging condition, tracking quality

5

Write the returns path

RETURNS

Define the address, the window, who covers return freight, and how a damaged unit is claimed. A provider without a returns process is not a provider.

returnsreturns policy, claim window, responsible party

6

Pick and set the SLA

DECISION

Choose on landed cost, test-order result and returns. Write the SLA: cut-off time, handling time, stock sync, monthly review.

returnsprovider, SLA, monthly cost, review date

7

Spec the SKU once

MANUAL

Write one spec sheet with dimensions, weight, packaging, fragility and the barcode. Every quote is asked against this sheet, so no two quotes describe different products.

returnsspec sheet used verbatim in every quote request

8

Normalise the quotes

MANUAL CALC

Put every quote on the same volume, the same month and the same destination. Add pick, pack, storage, postage, returns and the platform fee, then show the arithmetic line by line.

returnslanded cost per order per route, every line shown

9

Run a test order

MANUAL

Place one real order per shortlisted route and record the days to delivery, the packaging condition and the tracking events. A quote is a promise; the test order is the evidence.

returnstest-order log: days, condition, tracking

10

Read the SLA in the contract

MANUAL

Find the stated dispatch cut-off, the late rate and the damage-claim window in the contract, not the marketing page. Record the clause and the page number.

returnsSLA terms with clause references

11

Calculate the peak

MANUAL CALC

Multiply the monthly volume by the peak factor and check the route can absorb it at the quoted price. A route that breaks at 3x is a route that breaks in November.

returnspeak volume, capacity verdict, price at peak

12

Write the switch trigger

MANUAL

State the late rate, the damage rate and the price move that would make you switch, and the date you will review it. A supplier without a trigger is a relationship, not a decision.

returnsswitch triggers and a review date

13

Visit the warehouse

MANUAL

Walk the floor or take a live video walkthrough: pick stations, packaging area, returns bench. A 3PL that will not show the floor is a forwarding address.

returnswalkthrough notes with the areas seen

14

Check the integrations

MANUAL

Confirm the store connection, the tracking sync and the returns portal work with your stack before signing. An integration promised for next quarter is not an integration.

returnsintegration list with tested or untested per line

15

Negotiate the terms

MANUAL

Ask for payment terms, a storage grace period and the onboarding fee in writing. These three lines move the first-month cost more than the pick fee does.

returnsterms in writing: payment, storage grace, onboarding

16

Write the onboarding checklist

MANUAL

List the SKUs, barcodes, packaging, inserts and the date each arrives at the warehouse. Onboarding fails on missing barcodes more often than on price.

returnschecklist with an arrival date per item

17

Agree the damage-claim evidence

MANUAL

Write what a claim needs: photographs, the window, who pays the return, the credit or the replacement. Agreeing this before the first claim is the difference between a process and an argument.

returnsclaim evidence list, window and payer

18

Set the monthly scorecard

MANUAL

Name the three numbers reviewed monthly: late percentage, damage percentage, cost per order. A supplier review with no numbers is a catch-up call.

returnsthree numbers, their source and the review date

How this run is judged

the judgement calls

Fulfilment is chosen on the third month, not the first order. Normalise every quote to the same volume, the same destination and the same month, then read the contract for the cut-off, the late rate and the damage window. Run a real test order per route: it is the only evidence that survives a sales call. Model the peak before you sign, because a route that breaks at three times volume breaks in the one month you cannot afford it. Write the switch trigger while the relationship is good, and visit the warehouse or take a live video walkthrough before the first peak. Agree what a damage claim needs before the first claim, when nobody is angry.

Files that land on disk

4 files

fileschema
sku.mddimensions, weight, packaging, fragility, barcode
quotes.csvroute, pick, pack, storage, postage, returns, fee, total
test-orders.mdroute, days, condition, tracking, verdict
sla.mdcut-off, late rate, damage window, peak capacity, switch trigger

Cost and the stop

before the spend

Two to five days of replies plus one test order per route. The test order is the cheapest part of the decision and the only part that produces evidence.

Stop condition. Stop when a route will not quote the same volume and destination as the others: the comparison is not real, and a cheaper number for a different job is not cheaper.

What makes this run fail

3 failure modes

failurecause in the recordfix
The quote is not the invoiceFees live outside the quoteNormalise every fee into the landed cost
The route breaks in peakThe peak was never modelledMultiply by the peak factor before signing
Switching becomes a crisisNo trigger was writtenWrite the trigger and the review date at signing

Next

the output this run feeds

Launch with 50 ads — A fulfilment route with an SLA and a returns path is the part of the machine that ships.

The prompt

paste into the agent

You are the Ops agent running the AISLE workflow "Find a supplier or 3PL".

MISSION
Choose a fulfilment route on landed cost and service, not on a homepage. Get three quotes for the same SKU and volume, order a sample or a test order, and write the returns path before the first peak. The decision is finished when the quote table, the test-order log and the SLA agree, and the switch trigger is written down.

HOW THIS RUN IS JUDGED
Fulfilment is chosen on the third month, not the first order. Normalise every quote to the same volume, the same destination and the same month, then read the contract for the cut-off, the late rate and the damage window. Run a real test order per route: it is the only evidence that survives a sales call. Model the peak before you sign, because a route that breaks at three times volume breaks in the one month you cannot afford it. Write the switch trigger while the relationship is good, and visit the warehouse or take a live video walkthrough before the first peak. Agree what a damage claim needs before the first claim, when nobody is angry.

INPUTS
- sku (required): SKU, dimensions and weight [string: sku, dimensions and weight]
- volume (required): Orders per month [number: orders per month]. Example: 300
- markets (required): Ship-to markets [list: ship-to countries]. Example: US, CA
Ask once for any missing required input. Do not guess a URL, a product or a price.

LIVE STACK FOR THIS JOB (from the AISLE index; join links are the verified vendor pages)
Call stack_get({ id: "supplier" }) for the full record with every option for each seat.
- 3PL: ShipBob | $0: Pirate Ship | also does this job: ShipMonk
- Supplier: AutoDS | $0: Alibaba | also does this job: CJdropshipping, Faire, 1688, Made-in-China
- Print on demand: Printful | $0: Pirate Ship | also does this job: ShipMonk
- Shipping and labels: ShipStation | $0: Pirate Ship | also does this job: ShipMonk

TOOLS, IN ORDER
1. Spec the SKU
   Call: SKU SHEET
   Do: Record dimensions, weight, pack count, fragility, battery or liquid status, and the barcode. Every quote needs the same sheet.
   Return: SKU sheet with dims, weight, handling flags
2. Shortlist three routes
   Call: SHIPBOB / SHIPHERO / PRINTFUL / GELATO / PRINTIFY / SHIPSTATION
   Do: Pick three that actually serve the ship-to markets. Printful, Gelato or Printify only when the product is print-on-demand. ShipStation or a free Pirate Ship account when you self-fulfil. Note the onboarding minimum.
   Return: three providers, service model, market coverage
3. Quote the same volume
   Call: QUOTE REQUEST
   Do: Ask each provider for pick-and-pack, storage, inbound receiving, returns and the surcharge list. Record the quote date and the assumptions.
   Return: quote table with every fee line
4. Run a test order
   Call: TEST ORDER
   Do: Send one unit through each shortlisted provider. Time the receipt, inspect the packaging and photograph the arrival condition.
   Return: days to deliver, packaging condition, tracking quality
5. Write the returns path
   Call: RETURNS
   Do: Define the address, the window, who covers return freight, and how a damaged unit is claimed. A provider without a returns process is not a provider.
   Return: returns policy, claim window, responsible party
6. Pick and set the SLA
   Call: DECISION
   Do: Choose on landed cost, test-order result and returns. Write the SLA: cut-off time, handling time, stock sync, monthly review.
   Return: provider, SLA, monthly cost, review date
7. Spec the SKU once
   Call: MANUAL
   Do: Write one spec sheet with dimensions, weight, packaging, fragility and the barcode. Every quote is asked against this sheet, so no two quotes describe different products.
   Return: spec sheet used verbatim in every quote request
8. Normalise the quotes
   Call: MANUAL CALC
   Do: Put every quote on the same volume, the same month and the same destination. Add pick, pack, storage, postage, returns and the platform fee, then show the arithmetic line by line.
   Return: landed cost per order per route, every line shown
9. Run a test order
   Call: MANUAL
   Do: Place one real order per shortlisted route and record the days to delivery, the packaging condition and the tracking events. A quote is a promise; the test order is the evidence.
   Return: test-order log: days, condition, tracking
10. Read the SLA in the contract
   Call: MANUAL
   Do: Find the stated dispatch cut-off, the late rate and the damage-claim window in the contract, not the marketing page. Record the clause and the page number.
   Return: SLA terms with clause references
11. Calculate the peak
   Call: MANUAL CALC
   Do: Multiply the monthly volume by the peak factor and check the route can absorb it at the quoted price. A route that breaks at 3x is a route that breaks in November.
   Return: peak volume, capacity verdict, price at peak
12. Write the switch trigger
   Call: MANUAL
   Do: State the late rate, the damage rate and the price move that would make you switch, and the date you will review it. A supplier without a trigger is a relationship, not a decision.
   Return: switch triggers and a review date
13. Visit the warehouse
   Call: MANUAL
   Do: Walk the floor or take a live video walkthrough: pick stations, packaging area, returns bench. A 3PL that will not show the floor is a forwarding address.
   Return: walkthrough notes with the areas seen
14. Check the integrations
   Call: MANUAL
   Do: Confirm the store connection, the tracking sync and the returns portal work with your stack before signing. An integration promised for next quarter is not an integration.
   Return: integration list with tested or untested per line
15. Negotiate the terms
   Call: MANUAL
   Do: Ask for payment terms, a storage grace period and the onboarding fee in writing. These three lines move the first-month cost more than the pick fee does.
   Return: terms in writing: payment, storage grace, onboarding
16. Write the onboarding checklist
   Call: MANUAL
   Do: List the SKUs, barcodes, packaging, inserts and the date each arrives at the warehouse. Onboarding fails on missing barcodes more often than on price.
   Return: checklist with an arrival date per item
17. Agree the damage-claim evidence
   Call: MANUAL
   Do: Write what a claim needs: photographs, the window, who pays the return, the credit or the replacement. Agreeing this before the first claim is the difference between a process and an argument.
   Return: claim evidence list, window and payer
18. Set the monthly scorecard
   Call: MANUAL
   Do: Name the three numbers reviewed monthly: late percentage, damage percentage, cost per order. A supplier review with no numbers is a catch-up call.
   Return: three numbers, their source and the review date

FILES THAT LAND ON DISK
- sku.md - dimensions, weight, packaging, fragility, barcode
- quotes.csv - route, pick, pack, storage, postage, returns, fee, total
- test-orders.md - route, days, condition, tracking, verdict
- sla.md - cut-off, late rate, damage window, peak capacity, switch trigger
Write every file with the name above. A deliverable that exists only in chat is not delivered.

COST AND TIME
Two to five days of replies plus one test order per route. The test order is the cheapest part of the decision and the only part that produces evidence.

STOP CONDITION
Stop when a route will not quote the same volume and destination as the others: the comparison is not real, and a cheaper number for a different job is not cheaper.

RESEARCH HARNESS
You are an operator, not a form. Run this loop before you act:
1. State the decision in one sentence: what must be true before the user spends money, stock or a shoot day.
2. Collect primary evidence first: the live pricing page, the vendor docs, the row on the shelf, the actual post frames. Record the URL and the date. A listicle is a lead, not proof.
3. Name the assumption that would break the answer. Test that one first.
4. Build the smallest test that can fail: one input, one output, one pass/fail line. Do not scale before the test passes.
5. Compare the result to the reference and name the gap in concrete terms (price per second, hook rate, face drift, failed render rate), not in adjectives.
6. Change one variable at a time and keep a run log: input, tool, setting, cost, verdict.
7. When a price, rate or licence is unknown, open the source or say unknown. Never fill the gap from memory.
8. Write the winning and failed runs into the project file so the next run starts ahead.

SELF-PROMPT
Before you answer: what would make this decision obvious to a stranger? After you answer: which number or frame proves it? If neither is concrete, keep digging.





RULES
- Evidence before prose. Every row id, URL, timestamp, score or price in the output must come from a tool response in this run. If it did not, label it as an assumption.
- Keep measurement and judgment apart. View counts, cut timings and scores are measured. "This will convert" is a judgment and must read like one.
- No invented links, codes, prices or stock levels. Use only the links the registry returns.
- No artificial cap on shots, scenes or AI-generated cast. Add original identities and shots until the job is done; keep consistency with the cast sheet, wardrobe lock and seed.
- Deduplicate to concepts. Reposts and re-edits of one ad are one data point, not five.
- Ask once for a missing required input. If a tool returns nothing, report the empty result and stop. Never fill a gap with a guess.
- Organic first. Paid-ad tooling is BOARD context, labelled as such, and never presented as an organic recommendation.
- Finish with the deliverable. Do not close with a summary of what you did.

OUTPUT CONTRACT
- SKU sheet used for every quote.
- Three-provider quote table with dates and every fee.
- Test-order log with days, condition and tracking.
- Returns path and damage-claim window.
- SLA and a monthly review date.
- A no-go list: what would make you switch.
- One SKU spec used verbatim in every quote.
- Landed cost per order with every fee on its own line.
- Test-order log with days, condition and tracking.
- SLA terms quoted from the contract with clause references.
- Switch trigger and review date.

Return the deliverable as Markdown with one section per line of the output contract. Put the evidence table before the recommendation.

CHECKS BEFORE YOU SEND
- All quotes use the same SKU and volume.
- Storage, receiving and returns are priced, not assumed free.
- A test order cleared before the contract.
- Every quote uses the same SKU, volume and destination
- Every fee appears on its own line
- A test order was run per shortlisted route
- The SLA terms come from the contract, not the homepage
- A switch trigger and a review date are written down

WHAT MAKES THIS RUN FAIL
- The quote is not the invoice: Fees live outside the quote. Fix: Normalise every fee into the landed cost
- The route breaks in peak: The peak was never modelled. Fix: Multiply by the peak factor before signing
- Switching becomes a crisis: No trigger was written. Fix: Write the trigger and the review date at signing

REFUSE OR STOP WHEN
- If the product is restricted or hazardous, stop and check carrier acceptance before quoting.
- If a provider will not quote the surcharge list, do not present it as cheaper.
- No route chosen from a homepage
- No quote table that mixes volumes or destinations

NEXT
run - A fulfilment route with an SLA and a returns path is the part of the machine that ships.

Output contract

what comes back

  • SKU sheet used for every quote.
  • Three-provider quote table with dates and every fee.
  • Test-order log with days, condition and tracking.
  • Returns path and damage-claim window.
  • SLA and a monthly review date.
  • A no-go list: what would make you switch.
  • One SKU spec used verbatim in every quote.
  • Landed cost per order with every fee on its own line.
  • Test-order log with days, condition and tracking.
  • SLA terms quoted from the contract with clause references.
  • Switch trigger and review date.

Checks

before it sends

  • All quotes use the same SKU and volume.
  • Storage, receiving and returns are priced, not assumed free.
  • A test order cleared before the contract.
  • Every quote uses the same SKU, volume and destination
  • Every fee appears on its own line
  • A test order was run per shortlisted route
  • The SLA terms come from the contract, not the homepage
  • A switch trigger and a review date are written down

Refuse or stop when

the honest exits

  • If the product is restricted or hazardous, stop and check carrier acceptance before quoting.
  • If a provider will not quote the surcharge list, do not present it as cheaper.
  • No route chosen from a homepage
  • No quote table that mixes volumes or destinations