Home

/

Claude Skills

/

Carrier Routing

Carrier Routing

by

  • LOGISTICS
  • CARRIERS
  • COST
  • ECOMMERCE
Works with the AI tools you already use
+19 agents

Carrier allocation for ecommerce teams: price each carrier on delivered cost including failures and support, then shift volume to where it actually performs.

Carrier volume gets allocated on rate cards, which describe what a carrier charges and not what it costs you. A carrier that is forty cents cheaper per parcel and loses one in two hundred is more expensive once you count the replacement, the support ticket and the customer who does not order again. Those costs land in three different budgets, so nobody adds them up. This skill prices each carrier on delivered cost including failures and support, then shifts volume to where it actually performs.

What It Does

  • Computes delivered cost per parcel including failures, claims and support time.
  • Reads on-time performance by lane and by service level, not as one average.
  • Prices the customer cost of a late or lost delivery, not just the replacement.
  • Finds the lane where your cheapest carrier is actually your most expensive.
  • Checks capacity before recommending a shift that a carrier cannot absorb.
  • Shifts volume between carriers, with the delivered cost behind each move.

Use Cases

  • Find the carrier whose rate is cheapest and whose delivered cost is not.
  • Route a problem lane to a carrier that actually performs on it.
  • Price the support cost each carrier generates before renewing.
  • Move volume without exceeding what a carrier can absorb.

Example Session

Are we using the right carriers, and where should volume move?

Your cheapest carrier costs the most per delivered parcel on two lanes.

  • Carrier, lane | Rate | On time | Failures | Support cost | Delivered cost
  • Carrier A, metro | $4.20 | 96% | 0.2% | $0.11 | $4.38
  • Carrier B, metro | $3.80 | 91% | 0.9% | $0.47 | $4.61
  • Carrier A, rural | $7.90 | 88% | 1.1% | $0.62 | $8.91
  • Carrier B, rural | $6.40 | 94% | 0.3% | $0.19 | $6.71

Neither carrier is better; each is better on one lane. Carrier B is forty cents cheaper into metro and loses nearly five times as many parcels there, which makes it twenty-three cents more expensive delivered. On rural the position reverses completely and B wins by two dollars. Split the volume by lane rather than picking a primary carrier.

Why This Beats Prompting It Yourself

A Rate Card vs Delivered Cost: The Hidden Risk

A generic prompt compares rates, which is the comparison the rate cards are designed for. The forty cent saving is real and so is the 0.9% failure rate, and only one of them appears in the shipping budget. The replacement cost lands in inventory, the support time lands in headcount, and the lost customer lands nowhere at all.

When You Still Need The Answer To Be Per Lane

The question is usually posed as which carrier is better, and there is rarely one answer. Carrier B is the wrong choice into metro and clearly the right one into rural, by two dollars a parcel. Picking a single primary carrier gives up whichever lane it loses, and that concession never appears in a review.

Known Limitations

Support cost per carrier requires tickets to be attributable to shipments, and where they are not the skill estimates from failure rates and says so. The customer cost of a failed delivery uses your own repeat-purchase data where available and a conservative default otherwise. It recommends the routing and does not change carrier selection in your shipping platform.

Install It

  • Download the ZIP.
    It is free and there is no account to create.
  • Unzip it into your agent's skills folder.
    Claude Code reads ~/.claude/skills/, which is hidden by default: the command in the folder block opens it. Other agents scan their own directory, so drop the same folder there instead.
  • Ask your agent to use it.
    Restart the agent if it was already running, then it picks the skill up with no config.

skills folder

Copy

~/.claude/skills/carrier-routing/
  SKILL.md
  references/lane-performance.md
  routing/
  # one file per review, with delivered cost per lane

# macOS: create the folder and open it in Finder
mkdir -p ~/.claude/skills && open ~/.claude/skills

# Windows: paste in the Explorer address bar
%USERPROFILE%\.claude\skills\

Frequently asked questions

How is this different from comparing rate cards?

A rate card is what a carrier charges. This adds what its failures cost in replacements, support time and lost customers, per lane. A carrier that is cheaper on paper and worse in practice shows up immediately, and it is a common pattern.

Which agents does the skill run in?

Any agent that supports the open SKILL.md format: Claude Code, Cursor, Codex CLI, GitHub Copilot, Gemini CLI, Manus, Grok and others load it unmodified. The format is portable, the location is not. Each agent scans its own skills directory, so you drop the same folder into whichever one yours uses. There is no config file to edit and no API key to provision.

What does it need connected to work?

Shipment records with carrier, lane, cost and delivery outcome, plus support tickets linked to shipments where possible. Repeat purchase data makes the customer cost estimate specific rather than a default.

How do you price a lost customer?

From your own repeat rate after a failed delivery against your normal repeat rate, applied to the margin those orders would have carried. Where that history is thin the skill uses a conservative default and labels it, because overstating this number is the easy way to justify any carrier change.

Why split by lane rather than pick one carrier?

Because carrier performance is not uniform and the differences by lane are large. Consolidating onto one carrier buys simpler operations and gives up real money on whichever lanes that carrier is weak on, which the delivered cost column makes visible.

What is included with the skill?

The skill itself, the lane performance reference, the routing format, and the delivered cost model. It is free and security scanned. We re-publish the ZIP when the model changes, so download it again if your routing starts shifting.

Carrier choice is a margin line, not a logistics preference.

Polar joins shipping spend to order margin across 45+ sources, so moving volume between carriers is priced on what it leaves you.

Book a demo

Popular in E-commerce

<script type="application/ld+json">{"@context":"https://schema.org","@type":"FAQPage","mainEntity":[{"@type":"Question","name":"How is this different from comparing rate cards?","acceptedAnswer":{"@type":"Answer","text":"A rate card is what a carrier charges. This adds what its failures cost in replacements, support time and lost customers, per lane. A carrier that is cheaper on paper and worse in practice shows up immediately, and it is a common pattern."}},{"@type":"Question","name":"Which agents does the skill run in?","acceptedAnswer":{"@type":"Answer","text":"Any agent that supports the open SKILL.md format: Claude Code, Cursor, Codex CLI, GitHub Copilot, Gemini CLI, Manus, Grok and others load it unmodified. The format is portable, the location is not. Each agent scans its own skills directory, so you drop the same folder into whichever one yours uses. There is no config file to edit and no API key to provision."}},{"@type":"Question","name":"What does it need connected to work?","acceptedAnswer":{"@type":"Answer","text":"Shipment records with carrier, lane, cost and delivery outcome, plus support tickets linked to shipments where possible. Repeat purchase data makes the customer cost estimate specific rather than a default."}},{"@type":"Question","name":"How do you price a lost customer?","acceptedAnswer":{"@type":"Answer","text":"From your own repeat rate after a failed delivery against your normal repeat rate, applied to the margin those orders would have carried. Where that history is thin the skill uses a conservative default and labels it, because overstating this number is the easy way to justify any carrier change."}},{"@type":"Question","name":"Why split by lane rather than pick one carrier?","acceptedAnswer":{"@type":"Answer","text":"Because carrier performance is not uniform and the differences by lane are large. Consolidating onto one carrier buys simpler operations and gives up real money on whichever lanes that carrier is weak on, which the delivered cost column makes visible."}},{"@type":"Question","name":"What is included with the skill?","acceptedAnswer":{"@type":"Answer","text":"The skill itself, the lane performance reference, the routing format, and the delivered cost model. It is free and security scanned. We re-publish the ZIP when the model changes, so download it again if your routing starts shifting."}}]}</script>