653: This Checkout Trick Can Increase Sales 91% Overnight. Should You Use It?

Podcast: Download (Duration: 43:14 — 49.8MB)

43% of shoppers will abandon your store entirely when buy now pay later is not offered at checkout. Another 42% who stay will downgrade to a cheaper item or trim their cart.

That stat is what made me reconsider a payment method I publicly criticized in a YouTube video two years ago. The numbers that changed my mind: showing BNPL on product pages rather than only at checkout drove a 13% lift in sales and pushed average order value up 91%.

Toni Anderson joins me on this episode to work through whether the seller upside justifies the roughly 5.5% processing rate and the consumer downsides. We also cover the other checkout changes we have each tested on our own stores.

This post covers how buy now pay later works for sellers, what it costs, why higher income shoppers use it, and seven checkout changes that moved conversion rate or average order value for us.

Key takeaways

• 43% of shoppers abandon a store that does not offer buy now pay later.
• Buy now pay later placed on product pages lifted sales 13% and average order value 91%.
• 74% of BNPL users have accounts with two or more providers, so offering one may not be enough.
• Higher income shoppers use BNPL for cash flow flexibility rather than affordability.
• Sellers pay roughly 5.5%, more than double a typical payment processing rate.
• A visible phone number is one of the strongest trust signals at checkout.
• Lowering a free shipping threshold raises average order value when you stock cheap add-on items.
• Walk your own checkout on mobile, since desktop testing hides breakage.

How does buy now pay later work for ecommerce sellers?

The provider pays you the full order amount immediately and collects from the buyer in four installments, and you pay a processing fee of roughly 5.5% for that service. The customer receives the product right away.

Toni describes it as reverse layaway. Layaway held the merchandise until you finished paying, and BNPL ships first and bills after.

The buyer pays no interest across those four payments. That is what makes it attractive enough to change purchase behavior.

The provider absorbs the default risk. If a customer stops paying, you keep the money you were already paid.

Returns work differently. On a return, the merchant generally eats the cost of that transaction.

Affirm, Klarna, and PayPal are the major providers in this space. Several credit card issuers now offer their own pay-over-time features on existing card purchases.

How much does buy now pay later increase sales?

The largest single lift comes from where you place it rather than whether you offer it. Buy now pay later displayed on product pages drove a 13% sales lift and a 91% increase in average order value.

Checkout-only placement leaves that on the table. A shopper who sees the installment option before adding to cart shops at a different price point.

The abandonment number is the other half. 43% of shoppers leave a store that does not offer it at all.

The trade-down effect is quieter and still costly. 42% of shoppers who stay will pick a cheaper item or reduce their cart value.

How much does buy now pay later cost a seller?

Roughly 5.5% of the order, which is more than double what most sellers pay in standard credit card processing. That premium buys you the default protection and the conversion lift.

The math has to clear your margin. A 5.5% cut on a thin-margin product can erase the benefit of the higher order value.

Weigh it against the abandonment rate. Losing 43% of shoppers costs more than 5.5% on the orders you keep.

Test it as a placement change and not a permanent commitment. Running it on product pages for a defined period tells you your own numbers.

Should you offer more than one buy now pay later provider?

Yes, because 74% of BNPL users hold accounts with two or more providers and will jump to a second one when the first declines them or they are maxed out. Offering only Affirm loses the shopper whose Affirm limit is spent.

Spending limits are the reason. These are approval-based products, so a shopper can be approved on one provider and declined on another.

The visual cost is real. Multiple provider buttons crowd a checkout page that you have otherwise worked to simplify.

Weigh the crowding against the conversion math. Four buttons that recover a meaningful share of a 43% abandonment rate are worth some visual clutter.

New account signups add friction. A shopper without an existing account has to complete an application mid-checkout, which is where some of them drop.

Who actually uses buy now pay later?

Higher income shoppers use it more than sellers assume, which means presenting it as a budget tool misses a large part of the audience. The appeal is cash flow flexibility on larger purchases.

Toni describes buying a Sleep Number bed with the cash sitting in her account. She took 18 months interest free so the money kept earning while the bed came home that day.

She did the same on an air conditioning unit. Interest free financing on a large purchase is an easy yes when you already have the money.

That framing changes your merchandising copy. Position BNPL around flexibility and keeping your cash, rather than around affordability.

Toni’s caution stands on the consumer side. Missing a payment triggers steep penalty fees, and consumer advocates recommend against it for that reason.

The seller conclusion and the consumer conclusion differ. Consumers are buying this way regardless, which makes it worth testing as a merchant.

Does a countdown timer at checkout increase conversions?

A cart-hold countdown timer creates urgency and pushes shoppers to complete checkout. I run one on Bumblebee Linens that tells shoppers we will hold their cart for 15 minutes.

It is a genuine urgency mechanic. The visible clock changes how quickly people move through the form.

Expect some customer service on it. Older customers occasionally email asking for more time, and we extend it.

Isolate the test if you can. I implemented several checkout changes at once, so I never measured the timer’s individual lift.

Why does a phone number at checkout build trust?

A visible phone number tells shoppers a real business stands behind the site, and most of them will never call it. That presence alone converts.

Big-box brands do not need it. Shoppers already trust Amazon, Walmart, and Target because they can walk into a store.

An unfamiliar store has to earn that. A phone number, your photo, and your story all signal there is a real person behind the site.

Many large ecommerce stores still omit it. That is a free trust signal sitting unused.

Solo operators can still offer one. Post a number with instructions to text for fast service or leave a voicemail.

Callers skew older or are planners working against a deadline. Younger shoppers text or use live chat instead.

Can an AI agent handle ecommerce customer service calls?

For routine order questions, yes, and I just finished implementing one on my own store. It handles roughly the 70% of our calls that ask where an order is, when it ships, or what to do about damage.

It searches the site agentically. Ask it for a handkerchief with a script font and it assembles a cart and emails it to you.

Latency is good enough to feel conversational. It is not pretending to be human.

Toni’s objection is the phone tree experience. She recently spent an hour with a Department of Veterans Affairs system that failed to understand a spoken number.

The distinction is capability rather than category. A rigid menu tree frustrates people, and an agent that answers the actual question does not.

My rollout starts with off hours. Our phone hours are 10 to 5 with no weekend coverage, so the agent covers the gap.

Should you show a free shipping threshold in the cart?

Yes, and make it visually prominent, because shoppers will add items specifically to cross it. I display mine in large yellow text and have watched people add products in real time to clear the bar.

Toni tested the threshold level directly. Lowering it from $100 to $75 during a promotion raised average order value significantly.

The cluster sits just above the line. A large share of orders landed between $78 and $80.

The tactic requires cheap add-on inventory. Shoppers need $5 to $13 items they can add without a real commitment.

Toni deliberately built those into her catalog. Low-commitment impulse items exist so buyers can top up a cart.

Without them the threshold backfires. A shopper facing a $12 minimum add-on will pay the shipping instead.

How do cart page product recommendations affect average order value?

Showing frequently-bought-with products on the cart page adds a few percentage points to average order value without hurting conversion rate. I display recommendations alongside the items already in the cart.

Toni sees the same pattern. The lift comes without a conversion penalty.

Automated pairing needs a real pattern to find. Her customers buy multiple unrelated Bible studies, so frequently-bought-together data produces recommendations she would never suggest.

Match the recommendation to the goal. Cheap impulse items work when you want threshold-clearing adds, and true complements work when you want basket depth.

How should you word shipping protection at checkout?

Frame it as something the customer gains rather than a risk they are absorbing, and keep it to one line. Toni’s out-of-the-box shipping protection copy ran three lines and read as off-putting.

Rewriting it produced a lift. Shorter and positive outperformed the default language.

My version leads with the benefit. “Protection from damage, loss and theft. Get peace of mind with a delivery guarantee in the event your delivery is stolen or lost during transit.”

Default app copy is rarely optimized. Every third-party checkout app ships with generic language you should rewrite.

Why should you test your own checkout on mobile?

Because desktop testing hides breakage that only appears on phones, and most of your traffic is mobile. Toni’s site redesign went live with a rewards button covering the mobile checkout button.

Sales dropped immediately. The team troubleshot for 24 hours before someone opened the site on a phone.

The trigger was a feature enabled after testing. Rewards was turned back on after the dev site checks, so nobody had tested that combination.

Walk the full purchase yourself. Place a real order on mobile and refund it afterward.

Most sellers skip this entirely. Going through your own checkout as a customer is the cheapest conversion audit available.

When is a Shopify site redesign worth the cost?

When your current theme blocks a specific revenue mechanic you cannot build any other way. Toni’s redesign happened because her theme could not support a bundle builder through any out-of-the-box plugin.

The goal was a guided flow. She wanted the Domino’s pizza experience, where the site prompts you through add-ons rather than leaving you to assemble the order yourself.

The customer pattern justified it. Most buyers have multiple children and need a study for each plus a parent guide, and the old flow made that painful.

Custom-building the bundle builder alone would have cost nearly as much as the full redesign. That is what tipped the decision.

App consolidation offset part of the cost. Coding landing pages, bundle discounts, and a free shipping bar into the theme let her drop Zipify and several subscription apps.

Usage-based app pricing was the driver. A $19 app in the Shopify store ran $400 a month at her traffic volume.

Set realistic conversion expectations. Her site already converted above 3% against a typical 1%, so the agency told her upfront not to expect a large lift.

Two months in the gains are small and concentrated in mobile. Slight percentage point improvements, with more expected as the checkout changes compound.

Can AI replace an ecommerce design agency?

For straightforward layout and design work, largely yes, and complex custom functionality still favors an agency. I rebuilt a site in roughly four hours by directing Claude through the changes I wanted.

Hobby-level designers face real pressure. Simple redesign work is the most exposed.

Specialist agencies bring pattern knowledge. Toni points to small details her agency added from working across hundreds of stores, like graphic designators showing grade level.

Involved builds still need people. A custom bundle builder sits well beyond a prompt.

You can close part of the knowledge gap with research. I start every project by asking for a deep research guide on the topic, which often returns something genuinely comprehensive.

Frequently asked questions

What is buy now pay later in ecommerce?

A payment method that splits an order into four interest-free installments while shipping the product immediately. The provider pays the merchant upfront and collects from the buyer over the following weeks.

How much does buy now pay later cost merchants?

Roughly 5.5% of the order value, more than double a standard credit card processing rate. That fee covers the provider taking on the buyer’s default risk.

Does buy now pay later increase average order value?

Yes. Displaying it on product pages rather than only at checkout drove a 13% sales lift and a 91% increase in average order value.

What percentage of shoppers abandon carts without buy now pay later?

43% will leave the store entirely. Another 42% stay and downgrade to a cheaper item or reduce their cart value.

Which buy now pay later providers should you offer?

Offer more than one, since 74% of users hold accounts with two or more providers and switch when one declines them. Affirm, Klarna, and PayPal are the major options.

Do you need a phone number on your ecommerce site?

Yes, primarily as a trust signal for shoppers who do not recognize your brand. Most visitors never call, and its presence alone increases confidence enough to matter.

Does lowering your free shipping threshold increase sales?

It raised average order value significantly in Toni’s test, dropping from $100 to $75, with orders clustering just above the new line. It only works when your catalog has cheap add-on items shoppers can use to cross it.

How do you test your ecommerce checkout?

Place a real order on a mobile phone and refund it afterward. Desktop-only testing misses layout breakage, and elements enabled after your dev-site testing never get checked at all.