Magento Conversion Rate Optimization: A Checkout-First Plan

Magento conversion rate optimization is the work of removing the reasons shoppers leave before they pay. Baymard’s research puts documented cart abandonment at roughly 70%, and the fixable reasons it lists are mostly about checkout: surprise costs, forced accounts, long forms and errors. Start there, before you touch banners or button colours.

Key takeaways

• Baymard averages 50 studies to a 70.22% cart abandonment rate, but its own research found 42% of US shoppers had abandoned a cart simply because they were “just browsing.” Not all abandonment is a problem to solve.
• Among the rest, the biggest reason is extra costs being too high at 40%, ahead of slow delivery (20%), distrust of the site with card details (19%) and being forced to create an account (18%).
• Baymard says an ideal checkout can be as short as 12 to 14 form elements. Its benchmark of US checkouts averages 23.48.
• Adobe Commerce lets you change guest checkout, one-page checkout, terms acceptance, address search, cart and mini cart behaviour from Stores > Settings > Configuration, under Sales > Checkout. No code is needed.
• Google’s Core Web Vitals targets are LCP within 2.5 seconds, INP of 200 milliseconds or less and CLS of 0.1 or less, measured at the 75th percentile.

Most conversion advice reads like a list of tricks. Add urgency. Change the button. Show a badge. Some of it works on some stores, and none of it works until the basics are right. A shopper who hits an unexpected shipping charge on the last step isn’t going to be rescued by a clever banner.

This guide is built the other way round. It starts from what shoppers say stops them, maps each reason to something you can actually change in Magento (Adobe Commerce and Open Source), and puts the changes in an order that makes sense. Research figures come from Baymard’s published statistics, platform behaviour comes from Adobe’s configuration reference, and performance targets come from Google’s web.dev. We checked all three on the date at the bottom. We don’t quote uplift percentages, because we don’t have your store’s data and nobody else does either.

Why should Magento conversion rate optimization start at checkout?

Because that’s where the shoppers who already wanted to buy are lost. Someone who has added an item to a cart has told you they’re interested. If they leave at that point, something went wrong that you can usually see and fix.

Baymard’s cart abandonment page averages 50 separate studies into a headline figure of 70.22%. That number sounds terrible, but Baymard is careful to say a lot of it is unavoidable. Its own quantitative study found that 42% of US online shoppers had abandoned a cart because they were “just browsing” or not ready to buy. People use carts to compare prices, save items for later and explore gift options. You can’t convert them, and you shouldn’t try to with pop-ups.

What’s left is the part that matters. Once you set the just-browsing group aside, Baymard lists the remaining reasons and notes that many of them can be fixed through design changes alone. That’s a more useful framing than “lower abandonment.” The goal isn’t zero abandonment. It’s removing the avoidable reasons.

A note on scope: those percentages come from surveys of US online shoppers, and they describe all online retail, not Magento stores specifically. Your own analytics are the final word on your own store.

What are the biggest reasons shoppers abandon a cart?

Baymard’s distribution, with the “just browsing” group removed, is the best shortlist to work from. The table pairs each reason with the Magento lever that addresses it. The middle column is Baymard’s data; the right-hand column is our mapping, not research.

Two things stand out. First, costs and delivery together dominate. A store with a flawless checkout can still lose shoppers on the shipping line. Second, several of these reasons are the same problem seen from different angles. “Extra costs too high” and “couldn’t see the total up front” are both about surprise. Fix the surprise and you help both groups.

How do you measure before you change anything?

Set up a funnel you can read. At a minimum, that means counting sessions that reach each of these steps: product page, add to cart, cart, checkout start, shipping step, payment step and order confirmation. The drop between two adjacent steps tells you where to look first.

Segment it. Mobile and desktop shoppers behave differently, new and returning visitors behave differently, and each country may have its own payment and delivery expectations. An average across all of them hides the pattern. Google’s Web Vitals guidance makes the same point about performance data: it recommends assessing the 75th percentile of page loads, segmented across mobile and desktop.

Alongside the funnel, collect qualitative evidence. Session recordings and short exit surveys answer the “why” that the funnel can’t. Baymard’s own list includes a 7% “I don’t know” category, which is a reminder that surveys don’t capture everything either.

Then write down a baseline before touching anything. A change you can’t compare against a baseline is just a change. Record conversion rate by device, funnel drop-off by step, and your Core Web Vitals field data. You’ll want them all for the 30-day plan later in this guide.

What can you change in Magento’s checkout configuration?

More than most store owners realise, and without code. Adobe’s configuration reference for Sales > Checkout lists the settings. Reach them from the Admin sidebar via Stores > Settings > Configuration, then Sales > Checkout. The reference was last updated on 8 October 2026.

Under Checkout Options, the settings that bear directly on conversion are these.

Allow Guest Checkout. Determines whether guests can go through checkout without registering for an account. Baymard’s data says 18% of shoppers abandoned because a site wanted them to create an account. Unless you have a business reason to require accounts, leave this on and offer account creation after the order instead. This is a store view scope setting, so you can differ by store view.

Enable Onepage Checkout. Determines whether one-page checkout is the default checkout format. Whether one page or several steps is better for your store is an empirical question. We have a separate comparison of one-step versus multi-step checkout, and one on Hyvä Checkout versus one-step checkout if you’re on or considering Hyvä.

Enable Terms and Conditions. Determines whether customers must agree to the terms of sale before purchasing. Required in some jurisdictions and optional in others. If you need it, keep the wording short, because every extra required checkbox is one more thing to do.

Display Billing Address On. Lets you choose between showing the billing address on the payment method or on the payment page. Try both and measure.

Maximum Number of Items to Display in Order Summary. Defaults to 10. It controls how many items show in the checkout order summary.

Enable Address Search and Number of Customer Addresses Limit. These are Adobe Commerce only and set at website scope. When address search is enabled and a customer’s number of saved addresses meets the limit (default 10), only the default address is rendered on the shipping and review-and-payments steps, and the customer can search to switch. It mostly matters for B2B buyers with many saved addresses.

The Shopping Cart group has settings worth reviewing too, including whether the cart page appears immediately after adding a product (After Adding a Product Redirect to Shopping Cart), whether cross-sell items show in the cart (default Yes), and whether customers see an option to clear the cart in one action. Cross-sells can add revenue, but each one is also a chance to distract someone who was ready to pay, so test it.

None of these settings is a silver bullet. They’re just the cheapest experiments you can run, and they’re reversible.

Is guest checkout login worth turning on?

It’s a more subtle setting than it looks, and Adobe’s description is candid about the trade-off. The “Enable Guest Checkout Login” setting lets unauthenticated users, through the storefront and APIs, check whether an email address is already associated with a customer account. Adobe says this can improve the checkout flow for guests by showing a sign-in prompt when the entered email is already registered.

But Adobe adds that it comes at the cost of exposing information to unauthenticated users. In plain terms: anyone can probe whether a given email has an account with you. That’s a privacy consideration.

Our recommendation is to treat it as a deliberate decision rather than a default. If your analytics show many returning customers checking out as guests with an existing account, the prompt could help them. If your customer base is privacy-sensitive, or you have no evidence of the problem, leave it off. Document the decision either way.

How does checkout speed and stability affect conversion?

Baymard’s list includes “website had errors or crashed” at 17%, tying with the long, complicated checkout. So reliability isn’t a developer nicety. It’s a reason shoppers leave. Slow and failing checkouts also feel untrustworthy, which feeds the 19% who didn’t trust the site with card details. That second link is our inference, not Baymard’s.

Adobe’s checkout performance documentation, last updated on 8 October 2026, describes several options for high-throughput order processing. They’re aimed at stores that need to handle heavy order volumes, so read the caveats before enabling any of them.

AsyncOrder. Processes orders asynchronously using a queue. It’s disabled by default. The order is marked as received, queued, and processed first in, first out. Until the order is processed, product quantity isn’t adjusted and no sales email goes out. Once the consumer processes it, the order becomes Pending if the product is available, or Rejected if it’s out of stock or low on supply, and the customer is emailed either way. Adobe lists supported checkout types (Onepage Checkout, Standard Checkout and B2B Negotiable Quote) and payment methods (Check/Money Order, Cash on Delivery, Braintree and PayPal PayFlow Pro), and says features outside the list, such as multi-address checkout, admin order creation and other payment methods, continue to work synchronously. A conversion note: because a shopper can be shown an order as received and later receive a rejection email, think through the customer experience before enabling it.

Deferred Total Calculation. Defers calculating order totals until they’re requested for the cart or during the final checkout steps. Only the subtotal is calculated as items are added. Disabled by default. Adobe notes that with it on, Fixed Product Tax isn’t included in the mini cart’s price and subtotal after adding a product, and shows correctly in the shopping cart on the way to final checkout. That’s a pricing display difference you should check if you use that tax.

Inventory Check on Cart Load. The “Enable Inventory On Cart Load” setting is on by default. Turning it off improves performance across checkout steps, especially with bulk products in the cart. Adobe warns that skipping the check means some out-of-stock scenarios could throw other types of errors, though an inventory check always occurs at order placement. That’s in Stores > Configuration > Catalog > Inventory > Stock Options.

Load balancing. On cloud infrastructure, you can enable secondary read-only connections for MySQL and Redis through two environment variables, so that only one node handles read-write traffic.

Two cautions. First, these features target scale, and Adobe describes each as independent: you can use any combination. Second, Adobe’s page includes a note against using custom PHP code to customise the built-in cart and checkout capabilities, citing performance issues and complex upgrades that increase total cost of ownership. If you have to customise, Adobe says to use Adobe Commerce Marketplace-approved extensions only. Hold that line in conversion work: every “quick tweak” in checkout code is future upgrade risk.

For the wider performance picture beyond checkout, see our Magento performance optimization playbook and how page speed affects bounce rate.

What Core Web Vitals targets should you aim for?

Google’s web.dev defines three Core Web Vitals and gives target thresholds. Largest Contentful Paint (LCP) measures loading and should occur within 2.5 seconds of when the page first starts loading. Interaction to Next Paint (INP) measures interactivity and should be 200 milliseconds or less. Cumulative Layout Shift (CLS) measures visual stability and should be 0.1 or less.

web.dev recommends measuring at the 75th percentile of page loads, segmented by mobile and desktop, and says a page passes if it meets all three targets at that percentile. INP became a stable Core Web Vital in 2024, replacing First Input Delay.

Two practical points follow. Field data is what counts: Google’s own tools, including PageSpeed Insights and the Search Console Core Web Vitals report, draw on real-user data from the Chrome User Experience Report. Lab tools such as Lighthouse can’t measure INP because there’s no user input, though Total Blocking Time is a lab proxy.

Be careful about promises here. web.dev describes these as thresholds for a good user experience. It doesn’t claim a conversion lift, and neither do we. What you can say is that a checkout that stalls or shifts while a shopper is tapping is more likely to produce errors and abandonment, and a store that meets the targets removes one class of excuse.

If your theme is the bottleneck, see our comparison of Hyvä versus Luma performance and cost, and our Hyvä performance optimization service if you want help with the work.

How do you remove cost and delivery surprises?

Costs and delivery are the two biggest reasons on Baymard’s list, so they deserve the most attention. The principle is simple: tell shoppers the full price as early as you can.

Baymard found 40% of shoppers who abandoned for a fixable reason cited extra costs such as shipping, tax and fees, and 12% said they couldn’t see or calculate the total order cost up front. Put together, a lot of lost orders come from shoppers who got a number at the end that was bigger than the number they had in their heads.

What you can do about it in practice (this is our advice, mapped to Baymard’s findings):

• Show shipping information before the shopper starts checkout. A line on the product page or in the cart saying what shipping costs, or when it’s free, prevents the end-of-funnel shock.
• Make free-shipping thresholds visible. If you offer free shipping over a threshold, show how far the cart is from it. Our guide to a free shipping bar module for Magento 2 covers this approach.
• Use plain estimates for delivery. Twenty percent of abandoners cited slow delivery. A clear estimate lets shoppers decide before they invest effort in a form.
• Don’t hide fees. If a handling or service fee exists, show it with the other costs rather than adding it on the last step.
• Make returns information easy to find. Thirteen percent were unhappy with the returns policy. A short, visible summary near the buying decision helps.

How do you reduce checkout form friction?

Baymard’s benchmark gives you a number to aim at. Its large-scale checkout usability testing found that an ideal checkout flow can be as short as 12 to 14 form elements, or 7 to 8 counting only form fields. Its checkout benchmark database found the average US checkout contains 23.48 form elements shown by default (14.88 counting only fields). Baymard concludes that most checkouts could cut the number of form elements shown by default by 20 to 60%.

So audit yours. Count the elements a first-time guest must complete, then ask of each one: do we need this to fulfil the order? Common candidates are second address lines, company fields on a consumer store, phone numbers that carriers don’t require, and duplicate email fields. Some of these may be legitimately required in your market, and Baymard’s figures are averages, so test removals rather than assuming.

Account creation belongs here too. Eighteen percent of abandoners cited being asked to create an account. Offering guest checkout and inviting account creation on the confirmation page keeps the form short while still growing your customer base.

If you manage many saved addresses for B2B buyers, the Adobe Commerce address search feature above addresses a different friction: a long list of saved addresses.

Should you change your checkout theme or extension?

Sometimes, but not first. Replacing the checkout is the biggest and riskiest change in this guide, and it’s easy to do too early. Before you do it, run the cheap experiments: configuration settings, form-field reductions and cost transparency. If conversion is still stuck at the checkout step after those, a different checkout may be justified.

When it is, the choice is usually between staying on the default Magento checkout, using a third-party one-step extension, or moving to a Hyvä-based checkout. Our post on Hyvä Checkout versus one-step checkout is the place to start. We won’t declare a winner here, because the right answer depends on your payment methods, your customisations and your theme.

Whichever you pick, Adobe’s warning applies: customisation of the built-in checkout with custom PHP increases upgrade complexity. Prefer extensions that are reviewed and maintained.

How do you test changes without fooling yourself?

Testing conversion changes is mostly about discipline, not tools.

Change one thing at a time where you can. If you change guest checkout, form fields and shipping messaging in the same week, you won’t know which one mattered.

Run tests for full weekly cycles. Shopper behaviour varies by day of week. Stopping a test mid-week can mislead.

Decide your success measure before you start. Pick completed orders per session, or checkout completion rate, in advance. Choosing it afterwards, once you’ve seen the data, invites self-deception.

Watch for novelty and seasonality. A sale period, a holiday or a marketing campaign can change who arrives at your checkout. Note anything unusual alongside your results. If a seasonal peak is approaching, see our Magento Black Friday readiness checklist.

Keep a changelog. Record what changed, when, and why. Six months from now you’ll be glad of it.

We’re deliberately not giving a sample size or a number of weeks, because the right figures depend on your traffic and your baseline conversion rate. A store with few daily orders needs longer to see a real difference than a high-volume one.

A 30-day Magento conversion rate optimization plan

This is an order of operations, not a promise of results.

Week 1: measure. Build the funnel by step and device. Record baselines for conversion rate, funnel drop-off and Core Web Vitals field data. Review recordings of abandoned checkouts. List every step, field and fee a guest meets.

Week 2: fix the free, reversible things. Review Allow Guest Checkout, Enable Onepage Checkout, Enable Terms and Conditions, Display Billing Address On and the cart settings. Make costs and delivery estimates visible earlier. Cut form fields you can’t justify.

Week 3: fix speed and stability. Check checkout for errors in your logs. Compare your Core Web Vitals field data against the 2.5 second, 200 millisecond and 0.1 targets. Review whether any of Adobe’s checkout performance options suit your order volume, and read their caveats first.

Week 4: review and decide. Compare against baselines. Keep what helped, revert what didn’t, and decide whether a larger change such as a different checkout or theme is justified by what you’ve learned.

If you want help with the technical parts, our Magento development services team can take on the configuration, performance and checkout work, and the Hyvä performance optimization service is there if your theme is the constraint.

Common Magento conversion rate optimization mistakes

Optimising the homepage first. Visitors who are browsing aren’t the shoppers you’re losing at the cart. Start where intent is highest.

Treating all abandonment as a problem. Forty-two percent of US shoppers in Baymard’s study were just browsing. Chasing them with discounts can train customers to wait for offers.

Forcing an account. It’s one of the best-documented reasons for abandonment, and fixing it is one setting.

Adding trust badges instead of removing friction. Trust matters (19% cited it), but a badge can’t compensate for a confusing form or a surprise fee.

Customising checkout code for small wins. Adobe’s own documentation warns about the upgrade and performance costs. Weigh a one-percent idea against permanent maintenance.

Declaring victory too early. A good week proves little. Wait for enough orders and a full cycle before drawing conclusions.

Ignoring mobile. Your funnel report may look healthy overall while mobile is failing. Always segment.

For help judging where your store’s UX is leaking, our mobile UX audit guide is a useful companion.

Frequently asked questions

What is Magento conversion rate optimization?

It’s the process of increasing the share of store visitors who complete a purchase by finding and removing the causes of drop-off, especially in the cart and checkout. It combines measurement, configuration, performance work and testing.

What is a good conversion rate for a Magento store?

We don’t give a figure, because it varies widely by product, price, traffic source and market, and we won’t invent one. Compare your own funnel against your own baseline, segmented by device.

Why do shoppers abandon carts?

Baymard’s data shows that many are just browsing (42% in its US study). Of the remaining reasons, the top are extra costs (40%), slow delivery (20%), distrust with card details (19%) and forced account creation (18%).

Should I enable guest checkout in Magento?

Baymard found 18% of abandoners cited being asked to create an account, so for most stores guest checkout is the safer default. It’s the Allow Guest Checkout setting under Sales > Checkout, set per store view.

How many fields should a checkout have?

Baymard says an ideal flow can be as short as 12 to 14 form elements, versus an average of 23.48 in its US benchmark. Treat that as a target to test toward, not a rule, since some fields are legally or operationally required.

What are the Core Web Vitals targets?

According to Google’s web.dev: LCP within 2.5 seconds, INP of 200 milliseconds or less, and CLS of 0.1 or less, measured at the 75th percentile of page loads across mobile and desktop.

Does Adobe Commerce have built-in checkout speed options?

Yes. Adobe documents AsyncOrder, Deferred Total Calculation, an inventory check on cart load setting and load balancing for MySQL and Redis. Each has caveats, so read the documentation before enabling it.

Will a new theme improve my conversion rate?

It can remove performance problems, but it isn’t a guaranteed fix. Run the cheaper checks first and measure before and after.

Where to go from here

Open your Sales > Checkout configuration today and check the five settings that shape the guest experience: guest checkout, onepage checkout, terms, billing address placement and the cart options. Then count the form elements a guest has to fill in and write down the number. Those two checks cost nothing and tell you whether your biggest problems are the ones Baymard’s data predicts.

After that, measure your funnel by device, compare your Core Web Vitals against Google’s targets, and work through the 30-day plan. If you’d like a developer to do the technical parts, talk to our Magento team.