Loading tutorials…
Loading tutorials…
Most stores send to 'everyone who bought.' That's a list, not a segment. RFM + Predictive CLV segmentation is what separates email-as-newsletter from email-as-channel — typically 15-30% revenue lift on the same list.
Who this is forStores with 500+ orders in Klaviyo, mature welcome/cart/post-purchase flows, and a sense that 'email is plateauing.' Below 500 orders, predictive analytics doesn't have enough data and these segments don't fire reliably.
What you'll need
Step 1
Klaviyo Predictive needs 180+ days of orders AND 500+ orders. Check status before building segments.
Klaviyo → Analytics → Customer Behavior → Predictive Analytics.
If "Insufficient data" shows, you need more order history. Common reason: historical sync was never enabled, so Klaviyo doesn't see your pre-Klaviyo orders.
Once active, you have access to: Predicted Lifetime Value (CLV), Expected Date of Next Order, Churn Risk, Expected Order Number, Average Days Between Orders.
These properties update nightly. Build segments after the first nightly refresh post-activation.
Step 2
VIPs (high R + high F + high M), Champions, At-Risk High-Value, Lapsed, One-and-Done.
Klaviyo → Lists & Segments → + Create Segment.
Segment 1 — VIPs: 'Has placed at least 4 orders in the last 365 days' AND 'Has Placed Order with $value greater than $300 in last 90 days.' Calibrate the order count + dollar threshold to your AOV.
Segment 2 — Champions: 'Has placed at least 2 orders in the last 180 days' AND 'Predicted CLV greater than [top 25% of CLV distribution].'
Segment 3 — At-Risk High-Value: 'Predicted CLV greater than [top 25%]' AND 'Days since last order greater than [2x your median repurchase window].' These are the most important segment for win-back.
Segment 4 — Lapsed: 'Has placed at least 1 order' AND 'Days since last order greater than 180.'
Segment 5 — One-and-Done: 'Has placed exactly 1 order' AND 'Days since last order greater than 60.' These are the segment that converts well to a 2nd-order incentive.
Step 3
Top 10% CLV, Top 25% CLV, Bottom 50% CLV. Used to differentiate campaign targeting + offer strategy.
Create 3 segments by Predicted CLV:
CLV Tier 1: 'Predicted CLV greater than $[top 10% threshold].' These get premium service treatment — early access, no discounts (they don't need them), VIP cadence.
CLV Tier 2: 'Predicted CLV greater than $[top 25% threshold] AND less than $[top 10%].' Standard cadence + occasional special offers.
CLV Tier 3 (Bottom 50%): 'Predicted CLV less than $[median].' These benefit most from discount-driven campaigns and re-engagement.
To find the right thresholds: Klaviyo → Profiles → sort by Predicted CLV descending → eyeball the values at the 10th and 25th percentile of your active list.
Step 4
Churn Risk = High (top win-back target). Expected Date of Next Order approaching = replenishment target.
Segment — Imminent Churn: 'Churn Risk equals High' AND 'Has placed at least 2 orders' (excludes one-and-done from churn segment to keep it actionable).
Segment — Replenishment Window: 'Expected Date of Next Order' is between 'now' and '14 days from now.' This segment auto-updates daily and is the target for replenishment campaigns.
Replenishment-window segment is most useful for consumable categories (skincare, supplements, food). For non-consumables, churn-risk segment is more valuable.
Step 5
Klaviyo → Flows → + Create Flow → Trigger: Segment → "Imminent Churn." 3-email win-back with escalating offer.
Klaviyo → Flows → + Create Flow → From Scratch.
Trigger type: Segment → 'Imminent Churn' segment (the one we just built).
Email 1 (Day 0): soft 'we miss you' — no discount yet. Hero product they last bought.
Email 2 (Day 7): 10% off + 3 best-sellers since their last purchase.
Email 3 (Day 14): 15% off + free shipping + clear expiration. This is the last touch.
After the flow, if they still don't buy in 30 days, they enter a "sunset" segment and stop receiving regular campaigns — protecting list deliverability.
Step 6
Triggered on the 'Replenishment Window' segment. One email asking 'Time to reorder?' with one-click product link.
Create a flow triggered on "Replenishment Window" segment entry.
Email 1 (Day 0 — when they enter the segment, ~7 days before expected reorder): 'Time to reorder?' with the product they typically buy.
Email 2 (Day 14 — if no purchase): 10% off the same product.
Use Klaviyo dynamic content to pull the last ordered product: {{ person.most_recently_ordered_product }} or use a Product Block bound to the customer's order history.
This flow alone typically lifts repeat purchase rate 5-10 points for consumable categories.
Step 7
RFM and CLV distributions shift as the business grows. Re-pull thresholds every 90 days.
Calendar reminder every 90 days: re-pull Predicted CLV distribution and recalibrate the top-10%, top-25%, median thresholds.
Pull RFM segment counts: VIP should be ~5-10% of active subscribers. Champions ~15-25%. At-Risk High-Value ~5-15%. If your VIP count is below 2% or above 15%, the threshold needs adjusting.
Document threshold values in a shared doc — analysts and contractors need to know the boundaries to interpret reports correctly.
Common mistakes
Sending the same campaign to everyone
What goes wrong: VIPs get 10%-off campaigns they don't need (margin loss). At-Risk High-Value customers get standard messages and churn (revenue loss). Loses 15-30% of email revenue versus a properly segmented account.
How to avoid: Build the 5 RFM segments + 3 CLV tiers. Send VIPs a different cadence and offer from Bottom 50%. Stop treating the list as one audience.
Using "Total Number of Orders" instead of the metric-based "Has Placed Order" filter
What goes wrong: Profile property updates lag by hours. Segments based on profile properties exclude or include the wrong customers, especially right after a campaign send.
How to avoid: Use 'Has placed order' metric filters with timeframes, not the 'Total Number of Orders' profile property. Real-time event data is more reliable.
Building segments before Predictive Analytics has enough data
What goes wrong: Predicted CLV is null for most profiles. CLV-based segments are tiny (10-50 profiles) and unrepresentative. Decisions based on these segments are noise.
How to avoid: Wait until 180+ days of orders AND 500+ orders before building CLV segments. Until then, use only RFM (which works from any order count).
No win-back flow on the Imminent Churn segment
What goes wrong: Identified at-risk customers receive the same campaign as everyone else, churn at the normal rate. Loses the entire value of building the segment.
How to avoid: Build the 3-email win-back flow triggered on segment entry. Re-activates 8-15% of imminent-churn customers in most accounts.
Not suppressing flow members from campaigns
What goes wrong: Customer in win-back flow gets a 15% off win-back email AND the standard newsletter the same day. Both feel diluted. Recovery rate drops 30%+.
How to avoid: On every campaign send: Exclude segments → add any active flow segment. Klaviyo doesn't auto-suppress flow recipients from campaigns; you have to do it manually per campaign.
Static segment thresholds
What goes wrong: Thresholds set in Q1 don't fit Q4 distribution. VIP segment balloons to 30% of list (no longer 'VIP'). Or shrinks to <1% (no statistical signal). Either way the segment becomes meaningless.
How to avoid: Calendar reminder every 90 days to re-pull RFM and CLV distributions, adjust thresholds, document changes.
Recap
Done — what's next
How to build a Klaviyo welcome series flow that actually converts
Read the next tutorial
Hand it off
RFM + CLV segmentation is the work that separates good Klaviyo accounts from great ones. A specialist build with full segment library + win-back flow + replenishment flow is typically $700-1,200 at $14-16/hr. On a $50K-200K/mo store, this work usually compounds into 4-8x return within 6 months.
See specialist rates
180 days of orders AND 500+ orders. Both thresholds must be met. If you historical-synced 2 years of Shopify orders, you can hit both immediately. If you started Klaviyo fresh with no backfill, expect 6-12 months before Predictive populates.
A list is static — you add and remove people manually (or via signup forms). A segment is dynamic — it auto-updates daily based on rules. RFM/CLV segmentation only works with segments. Lists are for things like 'newsletter subscribers' that don't change based on behavior.
8-15 active segments for most stores. Less than 8 = under-segmented (sending the same things to different audiences). More than 15 = over-segmented (you can't reasonably target each in campaigns). 5 RFM segments + 3 CLV tiers + 2-4 use-case segments (engaged, suppressed, replenishment-window) is the sweet spot.
Lower than the general list, often higher quality. If your main list gets 3 campaigns/week, send VIPs 1-2/week with more premium content (early access, founder updates, no discount pressure). VIPs unsubscribe faster than the median if you blast them with the same offers as everyone else.
Not delete — sunset. Move lapsed (180+ days no purchase) to a separate sunset list and stop sending campaigns. This protects your sender reputation (Gmail/Outlook downrank you if engagement drops below 20%). After the win-back flow gives them one more chance, profiles that still don't engage should be marked as 'Suppressed' rather than deleted (so you keep the history).
Yes — Klaviyo CDP (the data platform tier) adds custom property syncs, external data sources, and more granular event modeling. But for stores under $5M/year in email-attributable revenue, the core Klaviyo segmentation toolkit covers 90% of the value. CDP is the next step after you've maxed out the basics.
Klaviyo
Welcome flows are the highest-revenue flow in most Klaviyo accounts — typically 30-45% of automated email revenue. A bad welcome flow leaves 60% of that on the table. This is the build that captures it.
Klaviyo
Post-purchase is where you turn a one-time buyer into a repeat customer — or don't. The flow needs different content for first-time vs returning buyers, and the timing of the review request alone shifts retention 10-15%.
Klaviyo
Open rates dropping from 28% to 16%, campaigns landing in promotions, sudden bounces? Deliverability is the silent killer of email revenue. This is the systematic diagnostic specialists run before suggesting any fix.
Klaviyo
DIY Klaviyo is a great call — until it isn't. Email should be 25-40% of e-commerce revenue. If yours is at 10-15%, the gap is your list isn't being worked. Here's the honest framework for when to hire.