Most DTC brands using Klaviyo are using about 40% of it. The flows are running, the campaigns go out weekly, and the basic segments exist. What is not being used - in almost every Klaviyo account Tribe audits - is the data layer underneath all of it: the predictive analytics, the customer data platform, and the segmentation logic that determines whether flows reach the right people at the right time or get sent to everyone indiscriminately.
This matters because the data layer is where the performance gap between a good Klaviyo account and a great one actually lives. Two brands with identical flows and identical creative can produce very different revenue per recipient figures depending on how well their segmentation uses the behavioural and predictive data Klaviyo has collected. For DTC brands looking to improve Klaviyo performance without adding more campaigns, this is the right place to look first.
Klaviyo as a customer data platform
Klaviyo is not just an email tool. Since launching its Klaviyo Data Platform (KDP) in 2024 and extending it through 2025, Klaviyo has become a genuine customer data platform - a system that unifies every customer interaction into a single, real-time profile. Purchase history, browsing behaviour, email engagement, SMS interactions, subscription status from Recharge or Skio, loyalty data, support ticket history - all of it can sit in one place and be used to drive automation and personalisation across every channel.
For DTC brands, the practical implication is significant. The customer profile in Klaviyo is no longer just an email address with a purchase history attached. It is a behavioural record that updates in real time as the customer interacts with the store, the email programme, and the subscription. Every flow and campaign can be built against that full picture rather than against a subset of it.
Most brands are not using this. The default Klaviyo setup - Shopify integration active, standard events firing, a handful of manual segments built at setup - captures a fraction of what is available. The brands generating 30-40% of total revenue from email are almost always the ones using the full data picture, not just the purchase events.
Klaviyo predictive analytics
Klaviyo's predictive analytics suite generates forward-looking data points at the individual customer level - not just what a customer has done, but what they are likely to do next. These predictions update continuously as new data arrives and are available as properties on every customer profile, which means they can be used directly in segment definitions and flow splits.
Predicted lifetime value
Predicted LTV gives a monetary estimate of the total revenue a customer is likely to generate over the next 12 months. Most brands treat all customers identically in their marketing - the same campaign, the same offer, the same frequency. Predicted LTV allows meaningful differentiation: high-predicted-LTV customers receive early access, VIP communications, and loyalty-reinforcing content; low-predicted-LTV customers with high recent engagement receive conversion-focused sequences designed to increase their value before it is established.
For subscription brands specifically, predicted LTV is one of the most useful inputs into the VIP flow architecture covered in the Klaviyo flows for DTC brands guide. Defining VIP thresholds by predicted LTV rather than historical spend alone catches high-potential customers earlier in their lifecycle.
Predicted next order date
Klaviyo's model estimates when each customer is likely to make their next purchase based on their individual purchase cadence. This is the trigger for replenishment flows - the most reliable use case for predictive analytics on consumable DTC products. Instead of sending a replenishment email on a fixed schedule (30 days after purchase, regardless of how quickly each customer actually consumes the product), a replenishment flow triggered by predicted next order date reaches each customer at their individual reorder moment. The conversion rate difference between fixed-schedule and prediction-triggered replenishment is consistently material - typically 20-40% higher RPR for the prediction-driven approach.
For non-subscription brands, predicted next order date is also the most reliable signal for identifying customers who are about to lapse. A customer whose predicted next order date has passed without a purchase is a win-back candidate - which is the correct trigger for a win-back flow, far more precise than a fixed time-since-last-order approach.
Churn risk
Klaviyo's churn risk score predicts the probability that a customer will not make another purchase. High churn risk customers are the audience for proactive retention sequences - targeted content, offers, or simply a re-engagement flow that reminds them why they bought in the first place. The churn risk property integrates directly with the retention and win-back flow architecture covered in the subscription retention strategy guide - and for one-time buyer brands it is equally applicable for identifying lapsing customers before they are gone entirely.
The brands not using churn risk are running win-back flows triggered by time since last purchase - a blunt instrument that sends the same re-engagement sequence to customers who bought six months ago regardless of whether their predicted behaviour suggests they are genuinely at risk or simply buying at a longer natural cadence. Churn risk makes that distinction. The first group needs intervention; the second does not.
The accuracy of Klaviyo's predictive models depends on the quality and completeness of Shopify data flowing into the platform. See our guide to configuring the Klaviyo-Shopify integration correctly for how to ensure the data foundations are right.
Klaviyo segmentation best practices
Segmentation is the mechanism that connects the data Klaviyo holds to the messages customers receive. A Klaviyo account with excellent data and poor segmentation produces the same result as one with poor data - everyone gets the same message regardless of what Klaviyo knows about them. The segmentation practices that produce the biggest performance improvements are not complex to implement, but they require a deliberate decision to move beyond the default segments most accounts are built on.
RFM segmentation
RFM - Recency, Frequency, Monetary - is the most commercially useful segmentation framework available in Klaviyo and the most consistently underused. It divides the customer base into meaningful groups based on how recently they purchased, how often they purchase, and how much they spend. The resulting segments map directly to the right marketing action for each group: recent high-frequency buyers are VIP candidates; high-spend infrequent buyers need cross-sell content; low-recency, low-frequency buyers need win-back sequences or list suppression.
Klaviyo's segment builder supports RFM construction natively using date filters, order count properties, and total spend figures. A basic four-quadrant RFM model built in Klaviyo takes an hour to configure and immediately produces a cleaner, more targeted campaign programme than one built on default segments.
Behavioural segmentation
Klaviyo captures detailed behavioural data from the Shopify integration - product views, collection views, add-to-cart events, search terms - that most brands do not use in segmentation or flow logic. A customer who has viewed a specific product category multiple times without purchasing is a browse abandonment candidate with a known product affinity. A customer who consistently opens emails about one product range and ignores communications about another is telling Klaviyo something about their preferences that manual segmentation does not capture.
Product affinity segments - built on purchase and browse data to identify which product families each customer engages with - allow campaign personalisation at a level that the standard "all subscribers" approach cannot reach. For DTC brands with a product range wider than two or three SKUs, product affinity segmentation is one of the highest-leverage improvements available without touching flow architecture.
Subscription status segmentation
For brands on Recharge or Skio, subscription status is one of the most important segmentation dimensions in Klaviyo and the one most commonly absent from campaign logic. Active subscribers, lapsed subscribers, one-time buyers who have never subscribed, and cancelled subscribers are four distinct audiences with four distinct commercial relationships to the brand. Sending the same campaign to all four is not personalisation - it is a missed opportunity at every send.
Active subscribers should receive content that reinforces the subscription value and deepens the brand relationship - not promotional offers to buy again, because they are already buying. One-time buyers who have never subscribed should receive subscription conversion content - the case for recurring ordering, the saving, the convenience - not the same campaign the subscriber receives. Cancelled subscribers are a distinct reactivation audience. The Klaviyo flows for subscription brands guide covers the flow architecture; subscription status segmentation is the campaign-level complement to it.
Klaviyo AI features in 2026
Klaviyo has shipped AI features steadily since 2023 and by 2026 has a meaningful suite that DTC brands are underusing. The most commercially relevant for DTC:
Smart Send Time
Smart Send Time uses machine learning to determine the optimal delivery time for each individual subscriber based on their historical engagement patterns. Rather than sending a campaign at 10am to everyone on the list, Smart Send Time delivers to each subscriber at the time they are most likely to open. The open rate uplift varies by list quality and engagement baseline, but consistently produces 5-15% improvement on lists with sufficient engagement history to train the model. It is most valuable for broadcast campaigns and least valuable for flows with intrinsic timing logic (abandoned checkout at one hour, replenishment at predicted order date).
Subject line and content AI
Klaviyo's AI subject line assistant generates alternatives based on the email content and historical performance data from the account. It is a starting point for iteration, not a replacement for editorial judgement - the best subject lines for a specific brand with a specific voice require human refinement. Used as a prompt for A/B test variants rather than a final output, it accelerates the testing cadence that compounds open rate improvements over time.
Channel affinity
Launched in 2025, Channel Affinity identifies whether each customer responds better to email, SMS, or push notifications based on their engagement history and automatically prioritises delivery via their preferred channel. For brands running both email and SMS, this removes the guesswork from channel allocation and reduces the risk of over-communicating via a channel a customer ignores while under-communicating via the one they consistently engage with.
Connecting the data layer to flow performance
None of the data, predictive, and segmentation features above are useful in isolation. Their value is in the decisions they power - specifically, the conditional splits and audience definitions inside flows that determine which message each customer receives.
A post-purchase flow with no conditional splits sends the same onboarding sequence to a first-time buyer and a customer who has bought six times before. Adding a split based on order count changes the first-time buyer experience (education, brand story, subscription conversion) and the returning customer experience (loyalty acknowledgement, cross-sell, VIP entry) simultaneously. Adding a second split based on predicted LTV further differentiates between the high-value customer who should receive premium treatment and the lower-value customer whose sequence should focus on increasing their purchase frequency before investing heavily in retention.
The Klaviyo flows guide covers the flow architecture in detail. The data features in this post are the inputs that make those flows perform materially better than they would with default segmentation.
If your Klaviyo account has flows running but is not generating the RPR benchmarks it should, the data layer is almost always where the gap is. A Tribe Klaviyo audit covers predictive analytics activation, segmentation logic, flow split architecture, and subscription event mapping - and produces a prioritised set of improvements with estimated revenue impact. Get in touch if you want to understand where yours is falling short.
Frequently asked questions
What is Klaviyo's customer data platform?
Klaviyo's customer data platform - the Klaviyo Data Platform (KDP) - unifies customer interactions from every channel into a single real-time profile. Purchase history, email and SMS engagement, browsing behaviour, subscription status, loyalty data, and support history all sit in one place and are available for segmentation, flow logic, and personalisation. For DTC brands, this means every flow and campaign can be built against a complete customer picture rather than a limited subset of it. It is meaningfully different from Klaviyo's earlier list-based model and makes segmentation accuracy significantly higher when properly configured.
What is Klaviyo predictive analytics?
Klaviyo predictive analytics generates forward-looking data points at the individual customer level - predicted lifetime value, predicted next order date, and churn risk score. These update in real time and are available as customer profile properties, meaning they can be used directly in segment definitions and flow conditional splits. Predicted next order date is most commonly used to trigger replenishment flows. Churn risk identifies customers at risk of lapsing before they have actually left. Predicted LTV defines VIP thresholds and priority audiences for retention investment.
How do I improve Klaviyo segmentation for a DTC brand?
The highest-impact segmentation improvements for most DTC Klaviyo accounts are: building RFM segments (Recency, Frequency, Monetary) to replace default engagement-only segments; adding subscription status splits to campaign logic so active subscribers and one-time buyers receive different messaging; using product affinity data from Shopify browse and purchase events to personalise content by product category; and activating predictive analytics properties (predicted LTV, churn risk) in flow conditional splits rather than using fixed time-based logic. None of these require advanced technical setup - they use Klaviyo's native segment builder applied to the data already in the account.
What is Klaviyo Smart Send Time?
Smart Send Time is Klaviyo's machine learning feature that delivers each campaign to each subscriber at the time they are individually most likely to open, based on their historical engagement patterns. Rather than sending a campaign at a fixed time to the whole list, Smart Send Time staggers delivery across a 24-hour window to optimise for each recipient. It typically produces 5-15% open rate improvement on lists with sufficient engagement history to train the model. It is most valuable for broadcast campaigns and less relevant for flows where the trigger timing is intrinsic - abandoned checkout flows, for example, should fire on behavioural logic rather than send time optimisation.