Seven out of ten online shoppers still leave without buying. Baymard Institute’s latest meta-analysis of 50 studies puts the average cart abandonment rate at 70.22%, a number that has barely shifted in a decade despite constant investment in checkout design. For a mid-sized retailer, that gap is not a rounding error; it is the difference between a good quarter and a missed one.

Salesforce has spent the last two years rebuilding its commerce and marketing stack around this exact problem. Commerce Cloud has moved toward a unified, one-click checkout and autonomous shopping agents. Marketing Cloud has been rearchitected into Marketing Cloud Next, running on Data 360 and Agentforce instead of static, rule-based journeys. Used together, the two platforms now address cart abandonment from both directions: removing the friction that causes it, and recovering the revenue when it still happens.

This piece walks through what actually changed, what the data says about why carts get abandoned in the first place, and how the current versions of B2C Commerce Cloud and Marketing Cloud work together to bring shoppers back.

Why the Cart Abandonment Rate Is Still Above 70%

Before looking at the fix, it helps to know what’s actually broken. Baymard’s research points to a handful of causes that repeat across industries and geographies:

  • Unexpected costs at checkout. Shipping fees, taxes, or service charges that appear only at the final step drive roughly 48% of abandonments, the single largest cause on record.
  • Forced account creation. Shoppers who just want to buy one item resent being asked to register before they can pay.
  • Checkout length and complexity. A checkout with too many form fields or steps loses customers who would otherwise complete the order.
  • Payment and security doubts. Shoppers hesitate when they don’t recognize the payment options or trust signals on the page.
  • Device friction. Mobile abandonment consistently runs well above desktop, often by 10 to 15 percentage points, because small screens amplify every one of the issues above.

None of these are new problems. What has changed is the tooling available to fix them, and that’s where Commerce Cloud and Marketing Cloud come in.

How Salesforce B2C Commerce Cloud Reduces Cart Abandonment at Checkout

Salesforce Checkout: A Unified, One-Click Checkout Experience

Salesforce Checkout is now a unified, one-click checkout layer built on the Einstein 1 Platform, connecting payments, tax, and shipping into a single flow rather than separate steps bolted together over time. Because it’s tied to unified customer data, a returning shopper can complete a purchase in far fewer taps than a traditional multi-page checkout requires, directly addressing the “too many steps” complaint that shows up in nearly every abandonment study.

Persistent Cart and Express “Buy Now” Checkout

B2C Commerce Cloud retains cart contents across sessions for both signed-in and guest shoppers, so someone who leaves mid-browse and returns a day later finds their cart intact rather than starting over. A “Buy Now” express flow, paired with the current Payments capability, lets shoppers skip the full cart entirely for a single-item purchase, cutting the checkout down to the minimum number of decisions.

Einstein and Agentforce Commerce Agents for Personalization

Commerce Einstein handles the personalization layer: product recommendations, predictive sort, and search results tuned to what an individual shopper is likely to want. In June 2026, Salesforce moved this further by taking three Agentforce Commerce agents to general availability, a Shopper Agent for conversational B2C buying, a Buyer Agent for B2B purchasing, and a Merchant Agent for storefront and product-page setup. The Shopper Agent in particular is relevant here: it can answer product questions, resolve sizing or availability concerns, and guide a hesitant browser toward checkout in the same conversational flow, rather than losing them to a tab switch.

Transparent Pricing and Real-Time Inventory to Prevent Abandonment

Composable storefront tools and native currency conversion let retailers show landed cost, taxes, and shipping earlier in the journey rather than surfacing them for the first time at payment. Combined with real-time inventory synced through Data Cloud for Commerce, shoppers see accurate stock and pricing from the product page onward, which closes off two of Baymard’s top-cited abandonment triggers before checkout even starts.

How Salesforce Marketing Cloud Recovers the Carts That Still Get Abandoned

No storefront eliminates abandonment entirely, which is why the recovery layer still matters. Marketing Cloud’s approach to this has changed shape considerably since the platform’s Marketing Cloud Engagement era, and it’s worth understanding the current mechanics rather than relying on how the feature used to work.

What Is the Salesforce Cart Abandonment Trigger?

Marketing Cloud’s native Cart Abandonment trigger activates when a shopper adds at least one item to their cart, ends the browsing session, and stays inactive for 30 minutes. It fires once per cart unless the shopper changes the contents, at which point it can activate again. That 30-minute window matters for timing recovery messages: send too early and you interrupt someone still comparing products; send too late and the moment has passed.

Agentforce Marketing and Journey Decisioning for Cart Recovery

Under the older Marketing Cloud Engagement model, abandoned-cart journeys ran on fixed rules: if a shopper didn’t return within X hours, send email A; if they still didn’t convert, send email B. Marketing Cloud Next replaces much of that rigidity with Journey Decisioning and Agentforce Marketing, which read real-time behavioral signals, browsing activity, engagement history, product interest, from Data 360 and adjust the next message accordingly, instead of following a single predefined path for every shopper.

Send-Time Optimization, Engagement Scoring, and Content Recommendations Explained

Three Einstein features do most of the practical work inside an abandoned-cart journey today:

  • Send-Time Optimization delivers each message when that specific subscriber is statistically most likely to open it, rather than a single blast time for the whole segment.
  • Engagement Scoring flags low-engagement contacts so recovery sends don’t damage sender reputation by going to people who reliably ignore them.
  • Content Recommendations populate the email with the actual abandoned products, plus adjacent items pulled from browsing and purchase history, based on live commerce data rather than a static template.

Cart Recovery Beyond Email: SMS, WhatsApp, and Mobile Messaging

Abandoned-cart recovery is no longer an email-only exercise. Marketing Cloud’s business-unit and channel architecture now scopes journeys across Email, SMS, WhatsApp, and mobile app messaging, so a retailer can reach a shopper on whichever channel they actually respond to, rather than assuming inbox is the only option.

How Data 360 Connects Commerce Cloud and Marketing Cloud for Cart Recovery

The real advantage isn’t either platform on its own, it’s what happens when Commerce Cloud and Marketing Cloud read from the same customer profile through Data 360. A shopper’s browsing behavior, cart contents, loyalty status, and past purchases all live in one place, which lets an Agentforce Marketing agent build a genuinely relevant recovery offer instead of a generic “you left something in your cart” reminder. It also means the Merchant and Shopper Agents inside Commerce Cloud and the Agentforce Marketing agents inside Marketing Cloud are working from identical, current data, rather than two systems reconciling overnight through a batch sync.

For retailers still running legacy Marketing Cloud Engagement, Salesforce has been clear that this is a phased convergence rather than a forced migration, existing journeys can keep running while new capabilities get adopted incrementally, which matters for any team weighing the cost of a rebuild against the recovery-rate upside.

Cart Abandonment Checklist for Salesforce B2C Commerce Cloud and Marketing Cloud

  • Show total cost, including shipping and tax, on the product and cart pages, not just at the final checkout step.
  • Enable persistent cart and guest checkout so a shopper never loses progress by closing a tab.
  • Turn on Send-Time Optimization and Engagement Scoring before launching any recovery journey; both are configuration steps, not custom builds.
  • Build a short recovery sequence rather than a single email, an initial reminder inside the first few hours, a stronger incentive if there’s no response, and a final message before any offer expires.
  • Route guest-checkout abandoners into a separate flow, since they don’t sit in a login-based journey by default.
  • Extend recovery beyond email into SMS or WhatsApp for shoppers who have opted in, particularly on mobile where abandonment rates run highest.
  • Review Commerce Intelligence dashboards for orders attributed to Shopper Agent interactions, since Salesforce now reports this directly, giving a clearer read on which AI features are actually driving recovered revenue.

Cloud Odyssey’s Point of View on Salesforce Cart Abandonment Strategy

Most cart abandonment projects we get pulled into don’t fail because a business picked the wrong Salesforce products, they fail because Commerce Cloud and Marketing Cloud were implemented as two separate projects, months apart, with no shared plan for the data layer connecting them. Checkout gets rebuilt, then a marketing team bolts on an abandoned-cart journey afterward using whatever data happens to sync across.

The retailers seeing real recovery-rate gains right now are the ones treating Salesforce Commerce Cloud and Salesforce Marketing Cloud as one program from day one, with Data Cloud as the shared foundation and Agentforce layered in once that foundation is solid, not before. That sequencing is where most of the value in a 2026 Salesforce commerce stack actually sits, and it’s the approach we’ve used on digital commerce work like our Pothys Swarna Mahal implementation.

If your abandoned-cart recovery still runs on a single generic email template, the gap between where you are and what these platforms can now do has widened, not narrowed. Talk to our team about auditing your current commerce and marketing setup.

Frequently Asked Questions

Marketing Cloud’s Cart Abandonment trigger activates once a shopper adds at least one item to their cart, ends the session, and stays inactive for 30 minutes. It fires once per cart unless the shopper updates the contents, at which point it can trigger again.

Much of the core functionality- persistent cart, Send-Time Optimization, Engagement Scoring, and the native Cart Abandonment trigger is configuration rather than custom code. Deeper personalization through Agentforce Commerce and Marketing agents typically needs proper data-layer setup through Data 360, which is where an experienced implementation partner earns its keep.

The strongest-performing emails show the exact abandoned items with pricing, link directly back to a pre-filled cart, and include a time-sensitive incentive. Pairing that with Send-Time Optimization, so the message lands when that specific shopper is most likely to open it, tends to outperform a single fixed send time for an entire list.

Industry benchmarks consistently rank abandoned-cart flows among the highest-yielding automated messages a retailer sends, both by open rate and revenue per recipient. Salesforce’s own Commerce Intelligence dashboards now let retailers attribute a portion of recovered orders directly to Shopper Agent interactions, making the return easier to measure than it was under older, less integrated setups.

A single email leaves revenue on the table. A short sequence, an initial reminder within a few hours, a follow-up with a stronger incentive if there’s no click, and a final message before any offer expires- recovers more shoppers than one message alone. Marketing Cloud’s Journey Decisioning handles this sequencing and exits shoppers automatically once they complete a purchase.