The parcel cost under it changed July 12, and nobody has revisited in eighteen months.

Open your store settings and look at your free shipping threshold.

Now try to name who set that number, and what they based it on.

Probably nobody can. Somebody set it eighteen months ago by copying a competitor or rounding up to something clean, and it has sat there while parcel rates and basket sizes moved around it.

That number decides the margin on every order that crosses it.

Recalculate your free shipping threshold this weekend. Here’s the model 🔽

Your fastest growing accounts check their data the way they'd ask a coworker a question.

There's a reason for that, and it has nothing to do with adding another dashboard.

The insights already exist in your data - what's working, what dropped, what needs fixing this week.

The difference is retrieval.

Most operators dig for it. Opening the platform, pulling the same view, translating numbers into a decision. That's where operator hours disappear.

The fastest growing accounts skip that part entirely.

​Omnisend removed the digging. Connect Claude or ChatGPT through MCP and ask directly: what's working this week, what dropped, what to fix. The same AI layer builds a form from a plain description and reads your audience to build segments before the trend cools.

This hands back the retrieval hour. It doesn't write your strategy.

The operator still decides. They just stop digging first.

The gap was never insight. It was the ten minutes standing between you and the insight you already had.

​Ask your own data what it would say:​

Set a free shipping bar and shoppers build their carts up toward it.

Your threshold is a pricing decision hiding in your shipping settings, so nobody audits it the way they'd audit a price.

Three numbers worth knowing:

- 76% of shoppers ranked free shipping their top purchase-decision factor year-round

- 71% of shoppers in that study who pick orders up in store do it to dodge a shipping fee

- 40% of US shoppers abandon carts over unexpected costs like shipping and tax, the top fixable reason

People react hard to the shipping line, so where you set the bar is where you set your margin.

But the same bar can make you money or cost you money.

A shopper below your threshold will often add an item to reach it, and that item is profit you would not have made.

A shopper already above it ships free either way, so you eat the shipping on an order that was already happening.

So the real question: is your bar pulling people to spend more, or covering shipping on orders that would have happened anyway.

The threshold should be tuned to maximize contribution margin per order.

But operators chase AOV instead, which climbs even as profit per order slides.

They set a fixed percentage over AOV and that ignores margin, the one thing that decides whether the offer makes you money.

Walk the arithmetic:

Incremental margin = (threshold minus AOV) times margin percent

Net cost per qualifying order = all-in shipping cost minus incremental margin

Push the bar $15 above AOV while absorbing an $8 parcel.

At 60% margin, that extra $15 earns you $9 and the offer pays for itself. At 20% margin, the same $15 earns you $3 and you lose $5 on every qualifying order.

Same threshold. Opposite outcome.

So the rule takes an afternoon to run: model net CM per order at several thresholds and pick the peak. It usually lands just above AOV; how far above is your call.

A fractional CFO firm published a model worth copying.

Their hypothetical brand: ​$60 AOV

60% contribution margin

$14 all-in parcel

Their curve peaks at a $70 threshold, around $37 of net CM per order, versus about $22 under unconditional free shipping. Their numbers, so do not set your bar at $70.

The warning they attach is the one I would underline twice: use true margin. Plug in markup and every row under it becomes fiction.

Two ways to miss the peak:

- Set below it and you hand free shipping to baskets that were converting anyway.

- Set too far above AOV, the gap stops feeling like one more item, so shoppers stop reaching.

Your peak sits between those, and it will not be $70, because your margin and parcel weight are different.

Copy how the example is built, but run the numbers on your own store.

Rebuild that table with your inputs. Four candidate thresholds, one column for net CM per order.

First you need a floor.

Threshold floor = AOV + (all-in shipping cost divided by margin percent)

That is the lowest the bar can sit before the offer eats margin. Model up until CM per order peaks.

It’s important to consider all-in shipping costs. Not the rate your carrier quotes, the number on the actual invoice:

All-in shipping cost = base rate + fuel and residential surcharges + packaging plus pick and pack

This is the week to rerun your threshold math, and here is why.

Your all-in parcel cost just moved on July 12 when the USPS dimensional-weight formula flipped, which I walked through a couple of weeks back.

Any threshold from before that date runs on a stale input, so the floor is wrong, and the peak with it.

Pull your latest invoice and recompute the real per-parcel number first.

With a candidate threshold, pull last quarter's basket distribution and see what share of orders already clears it.

If a large share did, your threshold is not incentivizing anybody. It is subsidizing orders that were happening anyway.

One note on anchoring: do not use your AOV. The mean gets dragged up by whale orders and can land where your typical shopper never reaches.

Anchor to your modal order value, the bucket where baskets actually cluster.

Chart order values in $10 buckets and find the tallest bar. That's your anchor.

Two more levers, both free. I will not promise a number on either, so treat them as tests and let your own data answer.

Test 1: progress messaging. a live "you're $12 away from free shipping" nudge instead of a static threshold line.

Test 2: placement. Put the message beside the add-to-cart decision instead of a promo bar up top where people scroll past it.

Both are worth testing once your threshold is set, one at a time, or you won't know which moved your numbers.

1. Pull last quarter's basket distribution and the share of orders already qualifying.​​

- - -

2. Compute true all-in parcel cost off your latest carrier invoice, surcharges included.

- - -

3. Calculate your breakeven floor: AOV plus all-in shipping divided by margin percent.

- - -

4. Model net CM per order at your current threshold and two alternatives.

- - -

5. Test in $10 increments and judge on CM per order. AOV movement on its own will lie to you.

- - -

6. Re-run the parcel cost input every quarter. Rates move.

Half a day with a spreadsheet against a number that touches every order you ship.

The bar you're running today was set by feel, and that feel was calibrated to a parcel cost that stopped existing on July 12.

None of that shows up in your dashboard. Contribution margin drains out one order at a time, quietly, and your P&L only tells you about it a quarter later when the numbers stop reconciling.

You can price the threshold this week or keep paying whatever the last person's guess is costing you on every order. Your call.

Nord Media helps DTC brands between $5M – $100M+ in revenue scale by driving profitable growth utilizing Meta, Google and Creative Strategy.

Book a call and let's see how we can help you grow 👉 book a call with me.​

​

​

Twitter

-

LinkedIn

-

Instagram

-

Threads

​