The Case of the Phantom Free Shipping

A shopper adds a $195 swimsuit to her cart. The product page just promised her, in bold black text above the description, “Free shipping on all orders.” She rounds out the look with a $30 accessory. At checkout, she sees a $12 shipping fee where she expected zero. She closes the tab. She doesn’t come back.

I spent the last few weeks helping a swimwear brand run their first deliberate paid Meta campaign in nine months. $1,000 of validation budget. A tightly-scoped audience. Three creative variants. We watched the funnel hour-by-hour. We diagnosed the bleed at LPV→ATC. We chased an audience-creative-price hypothesis through three rounds of design memos. We were certain we knew what was broken.

We were wrong about half of it.

The discovery

The thing we missed was sitting in plain sight on every product detail page, just below the price. Bold. Black. Underlined accordion arrow next to it. Looked official:

“Free Shipping on all orders!”

It wasn’t true. The shipping rates configured in the Shopify Admin had no free option — Economy $12, Standard $15, Express $25. Every paid clicker who added to cart trusted that line. Every one of them hit checkout, saw a real shipping fee, and a portion of them bounced. The brand had no way to measure exactly how many. But the funnel shape had been telling us something all along.

The funnel math

Direct traffic converted at premium-DTC norm: Direct LPV→ATC at 23%, Organic Search at 50% on a small sample. Paid Social: 3.28%. ATC→IC: 21% versus a 30–50% premium-DTC band. IC→Purchase net (after a buyer-remorse refund) sat at 25% versus a 40–60% norm.

I had filed the underperformance under “audience-creative-price misfit.” That’s what the cumulative metrics looked like. That’s what the next campaign’s design memo set out to fix.

But the misfit explanation had a hole. If audience-creative-price was the dominant problem, you’d expect the bleed to concentrate at LPV→ATC — the moment the customer evaluates whether the product matches their intent. The 2–3× regression at ATC→IC and IC→Purchase points to something else. Something at the cart-summary moment. Something that broke trust after the customer had already committed.

Then I went to verify an unrelated theme upgrade and the PDP text caught my eye. The full funnel shape suddenly made sense.

Why this keeps happening

This isn’t a one-brand story. The mechanism that produces it is structural and I see versions of it on most premium-DTC storefronts.

A founder builds a storefront pre-launch. The PDPs get hand-written copy describing the intended brand experience: “Free shipping. Free returns. 30-day money-back guarantee. Hand-picked materials.” Aspirational. The brand-as-it-should-be.

Or — and this is more common than founders like to admit — the brand operator never writes the line at all. A setup partner adds it during initial buildout. A freelancer drops in “standard PDP copy” from a template they reuse across clients. An early agency leaves a placeholder that everyone forgets to revisit. Months or years later, the founder genuinely doesn’t know the line exists. I’ve sat across from operators who needed me to show them the screenshot before they believed me.

Either way, shipping rates get configured later, when fulfillment costs are known. So do return policies, refund windows, delivery SLAs. The aspirational copy doesn’t survive the cost reality — but the PDP copy lives in a different surface entirely. Product page templates. Section settings. Theme code. Nobody updates it. Nobody knows to update it. It sits there for years.

The customer-side experience is invisible to the brand operator because the gap lives in two unrelated systems. The PDP template says “Free shipping.” The shipping rate config charges $12. No dashboard surfaces both. The misalignment shows up only at the moment the customer rage-quits the checkout flow — which the brand operator never sees.

The variants are everywhere

Once you start looking, the same shape appears across every customer promise that gets made in PDP copy and operationalized somewhere else:

• Returns promise mismatch. “Free returns!” on PDP versus real return-shipping costs at the label step.
• Refund-window mismatch. “30-day money-back guarantee!” on PDP versus a 14-day window on the actual refund policy page.
• Delivery-time mismatch. “Ships within 24 hours!” on PDP versus a 2–3 business day SLA from fulfillment.
• Inventory-promise mismatch. “Always in stock!” on a collection landing page versus sold-out variants on every product.

Each one creates a trust violation at the same fragile conversion moment. None of them surface in analytics. All of them are five-minute fixes once you know they exist.

How to find it

For a Shopify brand, the check is two reads against two surfaces:

• Open the active theme’s product template and any sections it renders. Search the page source (or the theme code editor) for phrases like free shipping, free returns,30-day, guarantee, ships within, and in stock.
• For each promise you find, look up the actual implementation: the shipping rate configuration in Settings, the return policy page, the refund policy page, the fulfillment SLA from your 3PL, the inventory data.

Any mismatch → edit the PDP copy to match what your operations actually do. If the founder wants to keep the aspirational promise, fix the underlying policy to match instead. The unforgivable move is leaving them unaligned.

What we’ll learn next

The fix landed before the brand’s next paid campaign goes live. The shipping rate config now actually honors free shipping above a threshold. The PDP false-promise text is gone. The announcement bar at the top of every page reads Free shipping on orders over $200 — and means it.

If the shipping-promise mismatch was a meaningful contributor to the funnel bleed, the next campaign’s ATC→IC and IC→Purchase rates should normalize toward the premium-DTC band. If they don’t, the audience-creative-price hypothesis was the dominant problem and we still have work to do. Either way, the diagnostic question gets sharper.

Most DTC audits I see go straight to ads, attribution, analytics. The orphan promise on the product detail page is the boring check that nobody runs. It costs five minutes. It’s free to fix. And it might be the single highest-leverage trust signal on the entire storefront.

The customer doesn’t bounce because she doesn’t want the product. She bounces because she thinks you lied.