Free Shipping Threshold
by
Threshold setting for ecommerce teams: read your real basket distribution and shipping cost, then recommend the threshold that lifts order value without giving margin away.
Free shipping thresholds get set at a round number and left there through years of changing basket sizes and rising carrier rates. Too low and you pay shipping on orders that were always going to happen. Too high and the customers who would have added an item just leave. The right number is specific to your basket distribution and your shipping cost, and it moves. This skill reads both, models what each candidate threshold does to order value and to margin, and recommends the one that actually improves contribution.
What It Does
- Reads the full basket distribution rather than the average order value.
- Prices your real shipping cost per order, including the orders just above the line.
- Models the share of baskets that would move up to each candidate threshold.
- Accounts for the customers who abandon rather than add an item.
- Separates the lift in order value from the lift in contribution.
- Recommends the threshold, with what it costs and what it earns.
Use Cases
- Replace a round number threshold with one fitted to your baskets.
- Stop paying shipping on orders that cluster just above the line.
- Find the threshold where order value rises and margin does not fall.
- Test a change with a number attached before committing to it.
Example Session
Is our free shipping threshold set right, and what should it be?
Move it from $75 to $85. Above that you start losing more than you gain.
- Threshold | Baskets qualifying | Median basket | Shipping cost | Contribution
- $65 | 71% | $69 | $61,400 | $412,000
- $75, current | 58% | $71 | $49,800 | $428,600
- $85 | 44% | $84 | $38,200 | $451,000
- $95 | 29% | $87 | $26,100 | $437,400
At $85 the median basket rises $13 and shipping cost falls $11,600, which is worth $22,400 a quarter. At $95 the basket barely moves further and the abandonment starts to bite: the customers who would stretch to $85 do not stretch to $95, they leave. The curve has a peak and you are currently $10 below it.
Why This Beats Prompting It Yourself
Average Order Value vs The Shape Of The Distribution: The Hidden Risk
A generic prompt sets the threshold slightly above average order value, which is the standard advice and ignores that the average tells you nothing about where baskets actually cluster. A distribution with a dense cluster at $70 and a long tail has a completely different right answer from a flat one with the same average.
When You Still Need The Abandonment On The Other Side
Every threshold model shows order value rising as the threshold rises, because the qualifying baskets get larger by definition. What that hides is the customers who leave instead of adding an item. This skill prices both sides, which is why the recommendation stops at $85 rather than continuing up the apparent trend.
Known Limitations
The share of baskets that move up is estimated from your own history of threshold changes, so a store that has never changed its threshold gets a wider band the output labels. It assumes your shipping cost per order stays roughly linear across basket sizes, which breaks on heavy or oversized products. It recommends the threshold and does not change your storefront settings.
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/free-shipping-threshold/
SKILL.md
references/basket-model.md
recommendations/
# one file per review, with the curve
# 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 setting the threshold above my average order value?
That rule assumes the distribution is symmetric around the average, which it never is. This reads where your baskets actually cluster and models each candidate threshold against shipping cost and abandonment. The answer is usually several dollars away from the rule of thumb, and occasionally in the other direction.
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?
Order lines with basket values, your actual shipping cost per order, and any history of previous threshold changes. Without that history it still models the distribution and widens the estimate of how many baskets would move.
Does a higher threshold always raise order value?
Apparent order value, yes, because qualifying baskets are larger by definition. Contribution, no. Past a point the customers who would have stretched simply leave, and the skill prices that side explicitly rather than reading the rising line as progress.
Should I test the change or just make it?
Test it. The recommendation comes with a duration and what to watch, because a threshold change affects browsing behavior as well as checkout and the effect takes a couple of weeks to settle. The numbers are a model until your own customers confirm them.
What is included with the skill?
The skill itself, the basket model reference, the recommendation format, and the abandonment estimates. It is free and security scanned. We re-publish the ZIP when the model changes, so download it again if your curve starts shifting.
The threshold lives where AOV meets shipping cost.
Polar joins order value, shipping spend and margin in one semantic layer, so the number you set still earns at the new basket size.
<script type="application/ld+json">{"@context":"https://schema.org","@type":"FAQPage","mainEntity":[{"@type":"Question","name":"How is this different from setting the threshold above my average order value?","acceptedAnswer":{"@type":"Answer","text":"That rule assumes the distribution is symmetric around the average, which it never is. This reads where your baskets actually cluster and models each candidate threshold against shipping cost and abandonment. The answer is usually several dollars away from the rule of thumb, and occasionally in the other direction."}},{"@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":"Order lines with basket values, your actual shipping cost per order, and any history of previous threshold changes. Without that history it still models the distribution and widens the estimate of how many baskets would move."}},{"@type":"Question","name":"Does a higher threshold always raise order value?","acceptedAnswer":{"@type":"Answer","text":"Apparent order value, yes, because qualifying baskets are larger by definition. Contribution, no. Past a point the customers who would have stretched simply leave, and the skill prices that side explicitly rather than reading the rising line as progress."}},{"@type":"Question","name":"Should I test the change or just make it?","acceptedAnswer":{"@type":"Answer","text":"Test it. The recommendation comes with a duration and what to watch, because a threshold change affects browsing behavior as well as checkout and the effect takes a couple of weeks to settle. The numbers are a model until your own customers confirm them."}},{"@type":"Question","name":"What is included with the skill?","acceptedAnswer":{"@type":"Answer","text":"The skill itself, the basket model reference, the recommendation format, and the abandonment estimates. It is free and security scanned. We re-publish the ZIP when the model changes, so download it again if your curve starts shifting."}}]}</script>