Klaviyo Email and SMS ROI: What Lorna Jane’s 84x Result Really Teaches Marketers

Klaviyo email and SMS ROI is often presented as a channel question: should a brand invest more in email, SMS, or WhatsApp? Lorna Jane’s reported 84x ROI over 12 months suggests the more useful question is whether a business can coordinate those channels around one reliable view of the customer.

In an original Klaviyo video featuring Lorna Jane CRM Manager Luke Styles, the Australian activewear brand said it generated 84x ROI from email and SMS while managing one million active profiles. The number is attention-grabbing, but the strategic lesson is more durable than the headline: Lorna Jane moved away from fragmented customer data and separate channel operations toward an omnichannel lifecycle program built around shared signals, consent, and customer preferences. Klaviyo’s published case study similarly frames the transformation as one of connecting online, in-store, app, email, and SMS activity rather than simply sending more messages. (youtube.com)

For founders, retention teams, and ecommerce marketers, this is a useful case study because it challenges a common assumption. The path to better CRM performance is not necessarily a bigger promotional calendar. It is usually better orchestration: knowing who a customer is, where they are in their journey, what they have done recently, what they consented to receive, and which channel is least likely to become noise.

The headline: Lorna Jane reports 84x ROI from email and SMS

Lorna Jane is an Australian activewear brand with more than 100 retail stores and an audience of more than one million active Klaviyo profiles, according to the vendor’s case study. In the video, Styles says the company achieved 84x ROI from email and SMS over the preceding 12 months. That figure should be understood as a company- and vendor-reported result, not a universal performance benchmark for every retailer. (klaviyo.com)

An 84x ROI claim commonly means that attributed revenue was many multiples of the platform and program investment. But ROI calculations vary substantially. One team may include software, agency, creative, and SMS delivery fees; another may compare attributed revenue only with platform cost. Attribution windows also matter: a short post-click window will produce a different answer from a model that credits purchases after opens, clicks, or multiple touches.

That does not make the result unhelpful. It means marketers should separate two questions:

1. Is the reported result credible as a directional signal? Yes, because the mechanics behind it are recognizable: consolidated data, lifecycle automation, segmentation, and coordinated channels.
2. Can another brand reproduce 84x? Not automatically. Product margins, repeat-purchase cycles, list quality, average order value, discount strategy, data completeness, and attribution policy all affect the result.

The more transferable lesson is that high CRM returns tend to emerge when a team makes each message more relevant, rather than treating email and SMS as independent broadcast machines.

Why one million profiles changes the operating model

A database of one million active profiles is not simply a larger mailing list. At that scale, manual segmentation rules become brittle, campaign calendars can create overlap, and small data-quality failures multiply fast. A customer who browses on mobile, purchases in a store, receives an SMS, returns an item, and later opens an email should not appear as five disconnected records or be pushed into five competing automations.

That is why the Lorna Jane example is fundamentally about operational design. The brand needed a system that could translate large volumes of customer behavior into decisions that were timely enough to be useful and controlled enough to protect the customer experience.

The real problem was fragmented customer data

Styles describes Lorna Jane’s pre-Klaviyo environment as one where data existed in many places and the customer journey was not seamless. Email and SMS operated through different channels and systems. That scenario is common even at sophisticated retail businesses: ecommerce data lives in the store platform, point-of-sale data sits elsewhere, customer service has another system, and loyalty or app data may be loosely connected—or not connected at all. (youtube.com)

The consequences go beyond inconvenient reporting. Fragmented systems make it difficult to answer basic lifecycle questions accurately:

• Has this shopper already bought the item they are being reminded about?
• Did they opt in to SMS, email, both, or neither?
• Did an in-store customer later browse online without purchasing?
• Has a support problem made a promotional message inappropriate today?
• Which customers are at risk of lapsing, and which are simply between normal purchases?

When those answers are split across tools, marketers often compensate with broad segments and more campaigns. That creates a dangerous feedback loop: less confidence in data produces more blanket messaging, which can reduce engagement and lead to more opt-outs.

A unified profile is an execution advantage, not a dashboard feature

A unified customer profile matters because it enables a shared decision layer. A purchase, product view, subscription change, store visit, or support event can inform the same journey rather than triggering separate, contradictory communications.

For example, imagine a customer who purchases leggings in-store. A disconnected ecommerce email platform may still send a browse-abandonment email for that product later that day. A unified profile can suppress that prompt, switch the customer to post-purchase education, and reserve SMS for a future replenishment or loyalty moment—provided the customer has opted in and the data arrives quickly enough.

That is the distinction between multichannel marketing and omnichannel marketing. Multichannel means a brand uses several channels. Omnichannel means those channels recognize one another and behave as parts of the same customer relationship.

Klaviyo’s current product materials position its platform around an AI-powered B2C CRM that connects email, SMS, WhatsApp, ecommerce, and customer data. Its documentation also describes cross-channel flows that can coordinate email, SMS, and push messages in one automation. Whether a business uses Klaviyo or a competing stack, the underlying requirement is the same: the identity, event, consent, and suppression logic must be dependable. (klaviyo.com)

Why email and SMS perform better together than apart

Email and SMS are often incorrectly treated as substitutes. They have different strengths, costs, attention patterns, and permission expectations.

Email is generally better for richer merchandising, editorial content, product education, receipts, long-form launch stories, and complex visual storytelling. It can carry several product modules, explain sizing or material details, and support a customer who is still researching.

SMS is better suited to concise, high-intent, time-sensitive communication: a back-in-stock alert, early-access reminder, delivery update, expiring loyalty benefit, or a limited-time prompt that has clear value. It is also more intrusive. That makes restraint an important performance tactic, not merely a compliance exercise.

A mature lifecycle program uses those differences deliberately. It does not send every email as a text message and every text message as a shortened email.

A practical channel-role framework

A simple way to design a cross-channel program is to assign each channel a primary job:

This approach reduces duplication. It also lets a marketer build escalation rules. Send an informative email first; if the person does not engage and has explicit mobile consent, send a shorter mobile prompt later. Or do the reverse for a customer whose historic behavior shows they reliably engage by SMS.

The point is not to force every customer through every channel. It is to maximize relevance while controlling frequency.

Channel affinity is the AI feature worth understanding

In the original video, Styles highlights Klaviyo’s channel affinity feature, describing its role in identifying whether a customer prefers SMS, WhatsApp, or email. More recently, Klaviyo announced that its channel-affinity modeling includes WhatsApp in addition to email, SMS, and push. The company says the signal can be used in segments, flows, and conditional splits to route messages toward the channel on which a profile is expected to engage. (youtube.com)

This is a meaningful use of AI when it is treated as a routing aid rather than an autonomous marketing strategy. Channel affinity does not create customer desire, fix weak offers, or replace consent. It can, however, help answer an expensive operational question: if a person is eligible to receive a message through several channels, which one should carry the next message?

What channel affinity should do

Used well, a preference model can help a team:

• Route an alert through email, SMS, WhatsApp, or push based on expected engagement.
• Avoid sending identical messages everywhere by default.
• Build conditional paths into lifecycle flows instead of maintaining duplicate automations.
• Identify customers who respond to a channel differently at different stages.
• Limit expensive or intrusive mobile messages to moments where they are more likely to matter.

Klaviyo’s help documentation says channel affinity requires at least two active channels and at least 1,000 profiles with at least one delivery through each channel. It also notes that the feature considers engagement measures such as campaigns, rates, and site activity over different periods. Those eligibility requirements are a reminder that prediction quality depends on having enough relevant data. (help.klaviyo.com)

What channel affinity should not do

A model’s recommendation should not override customer choice. If a person unsubscribes from SMS, no engagement prediction makes a text permissible. If email deliverability is deteriorating, a model cannot solve it merely by selecting another channel. If a customer received three marketing messages yesterday, sending a fourth through a different channel is not necessarily a better experience.

There is also a subtle measurement issue. A model trained on historical engagement can reinforce current behavior: customers may open email more often because email is what they receive most. Teams should test the logic against holdout groups and monitor whether channel allocation improves incremental conversions, not just opens or clicks.

The practical standard is simple: let AI help choose among permitted, useful communications, then validate whether it creates incrementality and lowers message fatigue.

Lifecycle marketing is where the return compounds

The Lorna Jane story emphasizes nurturing customers through different lifecycles. This is important because lifecycle programs compound in a way campaign-only marketing does not. A campaign starts and stops. A well-designed flow becomes a repeatable system for responding to customer intent.

For apparel and activewear brands, lifecycle stages can include subscriber onboarding, first purchase, second-purchase encouragement, category expansion, replenishment, VIP development, inactivity, and reactivation. The exact labels are less important than the behavioral logic behind them.

Five flows to prioritize before adding more campaigns

Most teams will gain more by improving a handful of core flows than by launching another weekly promotion.

1. Welcome flow: Teach new subscribers what makes the brand valuable. Capture preferences where appropriate, introduce bestsellers, and set communication expectations.
2. Browse and cart recovery: Use viewed-product, cart, inventory, and purchase events to distinguish casual browsing from high intent. Avoid messaging people who already completed checkout elsewhere.
3. Post-purchase flow: Help customers use, care for, and enjoy the product. This is where a brand earns the right to ask for a review, recommend complementary products, or invite loyalty participation.
4. Second-purchase flow: Treat the first repeat order as a strategic milestone. Use product category, time since purchase, and customer value to make the next recommendation relevant.
5. Win-back flow: Separate genuinely lapsed customers from those with naturally long buying cycles. Lead with new value, not an automatic discount that trains customers to wait.

A unified data platform improves every one of these because it can combine profile attributes with event history. For Lorna Jane, that means the same system can theoretically account for customer behavior across its retail footprint, digital storefront, app, and direct messaging touchpoints. The vendor’s case study explicitly describes customers moving between in-store and online interactions within short periods, which is exactly the condition where disconnected automation creates mistakes. (klaviyo.com)

How to measure CRM ROI without fooling yourself

The biggest risk with impressive ROI figures is not that they are false; it is that teams copy the headline without copying the measurement discipline. Better measurement starts with definitions.

At a minimum, track four layers:

• Attributed revenue: Revenue the platform credits to messages under its selected attribution settings.
• Incremental revenue: Revenue that would not likely have occurred without the message or program.
• Contribution margin: Revenue after product cost, shipping, discounts, delivery fees, agency time, and software costs.
• Customer impact: Opt-outs, complaint rates, engagement decay, repeat purchase rate, and lifetime value.

Attributed revenue is useful for daily optimization. Incrementality is essential for deciding whether a program genuinely creates value. A cart-abandonment message may receive revenue credit for purchases that would have happened anyway. Conversely, a post-purchase education series can improve retention even if its immediate attribution appears modest.

A more rigorous ROI calculation

A practical formula is:

CRM ROI = (incremental gross profit − total program cost) / total program cost

Total program cost should include more than software. Add creative production, copywriting, data engineering, agency retainers, SMS or WhatsApp delivery fees, offer cost, and internal labor when those costs are material.

To estimate incremental lift, use randomized holdouts where possible. For example, hold back 5% of eligible cart abandoners from an SMS reminder while keeping all other experiences comparable. Compare conversion rate, average order value, refund rate, and subsequent engagement. This is not perfect experimentation, but it is far more informative than assuming every attributed order was caused by the message.

At Lorna Jane’s scale, even a modest improvement in conversion or repeat rate can justify investment in data quality and lifecycle design. That is another reason the 84x figure should not distract from the deeper operating model. High volume makes orchestration gains economically meaningful.

Consent, deliverability, and frequency are part of the strategy

The fastest way to damage an omnichannel program is to treat every reachable profile as a marketing opportunity. Marketing permission is channel-specific, customer trust is cumulative, and regulations matter.

For Australian audiences, the Australian Communications and Media Authority says the Spam Act 2003 and its regulations establish responsibilities for marketing emails and messages. ACMA’s guidance emphasizes consent and covers businesses of all sizes using email, SMS, or telephone marketing, while its anti-spam guidance points marketers toward the legal obligations around commercial electronic messages. (acma.gov.au)

The Office of the Australian Information Commissioner also identifies direct marketing as the use or disclosure of personal information to promote goods or services directly, including via email and text message. That makes governance of data collection, purpose, preferences, and opt-outs central to marketing operations—not an afterthought delegated entirely to legal teams. (oaic.gov.au)

Build the control layer before scaling channels

Before deploying sophisticated routing or AI, establish these controls:

• Store consent status separately for email, SMS, WhatsApp, and other channels.
• Preserve the source, date, and method of consent collection.
• Apply global and channel-level suppression immediately after opt-out.
• Cap frequency across channels, not just within each individual channel.
• Exclude customers with active service issues, returns, or recent complaints when appropriate.
• Monitor spam complaints, unsubscribes, bounce rates, and engagement by segment.
• Sunset persistently unengaged profiles rather than sending indefinitely.

Klaviyo’s own guidance advises targeting engaged profiles or subscribers to protect deliverability and reduce complaints, and its active-profile documentation notes that suppressing long-term unengaged profiles can reduce the number of billable active profiles. That aligns business incentives with a healthier list: better targeting can protect both inbox placement and spend. (help.klaviyo.com)

The community reaction is limited, but the industry signal is clear

The supplied video had no top comments to analyze, so there is no meaningful public viewer consensus to report from that source. That absence matters: marketers should not confuse a vendor-produced success story with independently audited industry evidence.

Still, related Klaviyo case studies show a recurring pattern among ecommerce brands: teams that had email and SMS on separate platforms describe coordination problems, difficulty managing timing, and underuse of mobile messaging. For example, bootmaker Lucchese reported a 65x 12-month ROI after consolidating channels and said the prior split setup made omnichannel automation harder for a lean team. That does not validate Lorna Jane’s exact result, but it supports the broader operational premise that a unified system can reduce friction in lifecycle execution. (klaviyo.com)

The useful takeaway is not “every brand needs Klaviyo.” It is “every brand needs a clear orchestration model.” A business may achieve that with Klaviyo, another CRM platform, a composable customer-data architecture, or a carefully integrated stack. The right choice depends on data maturity, budget, channel mix, technical resources, geographic requirements, and the degree to which retail and ecommerce events must be connected.

A 90-day plan to apply the Lorna Jane lesson

Brands do not need one million profiles to benefit from this approach. They do need to sequence the work correctly.

Days 1-30: Fix the customer record

Audit identities, event names, data freshness, consent fields, and source-of-truth systems. Map every point where a customer can enter or change the relationship: website sign-up, checkout, store purchase, loyalty enrollment, customer service, app, lead form, and unsubscribe.

Then answer a blunt question: can your system recognize that the person who purchased in-store yesterday is the same person browsing online today? If not, do not rush into AI-driven personalization. Solve matching, event ingestion, and consent first.

Days 31-60: Rebuild the highest-value journeys

Prioritize welcome, cart recovery, post-purchase, and win-back flows. Add decision logic that suppresses duplicate sends, distinguishes subscribers from non-subscribers, and escalates across channels only when there is a legitimate reason.

Set a shared frequency cap. A customer should not receive an email campaign, an SMS campaign, a cart reminder, and a product-review request within hours merely because four separate automations fired independently.

Days 61-90: Test routing and measure incrementality

Once the basics are stable, create tests around channel allocation. Compare email-first versus SMS-first for a small eligible segment. Test whether a second channel adds lift after non-engagement or simply creates more opt-outs. Use a control group for high-value flows.

Review results by customer type, not only by aggregate revenue. New subscribers, repeat buyers, VIPs, discount-driven customers, and long-lapsed customers may respond very differently. That is where lifecycle marketing becomes a learning system rather than a scheduled-message machine.

The bottom line: orchestration is the real growth lever

Lorna Jane’s reported 84x result is compelling, but the headline should not be interpreted as proof that SMS, email, or AI alone generates outsized returns. The case is stronger as evidence for a more practical idea: when a retailer centralizes customer data, coordinates channel roles, respects preferences, and designs flows around lifecycle moments, the economics of retention marketing can improve dramatically.

For marketers, the next move is not to copy Lorna Jane’s claimed ROI into a board deck. It is to assess whether your current stack can answer the essentials: who is this customer, what have they done, what have they consented to, what do they need next, and what is the least intrusive useful way to reach them?

If the answer is unclear, solve that first. A better customer record is the foundation for better automation. Better automation makes AI routing more useful. And better routing can turn email, SMS, WhatsApp, and push from competing send channels into one coordinated customer experience.

FAQ

What does 84x ROI mean in email and SMS marketing?

An 84x ROI means a program reported returns equal to 84 times its investment over a stated period. The exact interpretation depends on what costs and revenue attribution rules are included, so it should be treated as a case-specific metric rather than a guaranteed benchmark.

Why should email and SMS be managed together?

Managing them together lets marketers suppress duplicates, coordinate timing, use shared customer behavior, and assign each channel a different job. The objective is not to contact customers through more channels; it is to deliver the most relevant message through the most appropriate permitted channel.

What is channel affinity in Klaviyo?

Channel affinity is a Klaviyo AI signal designed to estimate which channel a profile is most likely to engage with. Klaviyo says it can inform segmentation and flow routing across channels including email, SMS, push, and WhatsApp, subject to eligibility and available data. (klaviyo.com)

Can a small ecommerce brand use the same approach?

Yes. Smaller brands should start with clean consent records, reliable purchase and browsing events, and a few high-impact lifecycle flows. They do not need a million profiles; they need enough trustworthy data to avoid duplicate sends and make messages relevant.

Does AI replace segmentation and testing?

No. AI can improve routing and prioritization, but it cannot replace clear customer permissions, thoughtful offers, strong creative, frequency limits, or controlled experiments. Treat it as a decision-support layer and measure whether it creates incremental value.