7 min read

How to Build RFM Segments in Klaviyo

How to Build RFM Segments in Klaviyo

RFM segmentation divides customers into groups based on Recency (days since last purchase), Frequency (total orders), and Monetary (total spend), enabling targeted retention and reactivation campaigns.

Why RFM Matters for DTC Operators

RFM is the operational backbone of email retention. Unlike demographic or behavioral tags, RFM captures purchase intent and engagement in a single, updatable metric. A customer who bought 60 days ago and spent $500 across 5 orders behaves differently than one who bought 6 months ago and spent $100 once. Klaviyo's native RFM tools let operators build this segmentation without custom SQL or external tools.

The payoff is measurable. High-frequency, recent, high-value customers (top-right of the RFM matrix) typically generate 80% of repeat revenue. Low-recency, low-frequency customers are churn risks. Segmenting by RFM lets operators allocate email budget to the segments most likely to convert, rather than broadcasting the same message to everyone.

Setting Up RFM Scoring in Klaviyo

Klaviyo calculates RFM automatically once you connect purchase data. The platform assigns each customer a score from 1 to 5 on each dimension (R, F, M), where 5 is best. Recency is measured in days since the most recent purchase; Frequency is the count of orders; Monetary is the sum of order value (before refunds, by default).

Before building segments, audit your data. Check that Klaviyo is correctly attributing orders to customer profiles. If your backend uses email as the primary identifier and Klaviyo uses a different key, RFM scores will be fragmented across ghost profiles. Run a cohort report on a known high-value customer to verify their R, F, M scores match reality.

Configure Klaviyo's RFM calculation settings under Analytics > RFM. Decide whether to include refunded orders (most operators exclude them). Set the lookback window - 365 days is standard, but if your product is seasonal or you launched recently, adjust accordingly. Once saved, Klaviyo updates scores nightly.

  • Recency: 1 = >180 days, 5 = <30 days (adjust thresholds if needed)
  • Frequency: 1 = 1 order, 5 = 5+ orders (scale to your median repeat rate)
  • Monetary: 1 = lowest quartile, 5 = highest quartile (Klaviyo auto-calculates)

Building Segment Logic: The RFM Matrix

The RFM matrix is a 5x5x5 grid, but operators don't need to segment all 125 cells. Focus on 6 - 8 high-impact segments that drive different behaviors. Start with these core groups: Champions (R5, F5, M5), Loyal Customers (R4-5, F4-5, any M), At-Risk (R1-2, F2-4, any M), and Lost (R1, F any, M any).

In Klaviyo, build each segment as a static list or dynamic segment. Static lists snapshot RFM at creation time; dynamic segments update nightly as scores change. For retention campaigns, use dynamic segments so a customer who just purchased automatically moves out of 'At-Risk' and into 'Recent Buyer'. For one-time reactivation blasts, static lists prevent duplicate sends.

Layer additional filters on top of RFM. Exclude unsubscribed users, filter by product category, or add a minimum email engagement threshold (e.g., opened an email in the last 90 days). This prevents sending reactivation emails to hard bounces or users who never engaged with your list.

  • Champions: R >= 4, F >= 4, M >= 4 - VIP retention, early access, exclusive offers
  • Loyal: R >= 3, F >= 3, M any - upsell and cross-sell campaigns
  • At-Risk: R <= 2, F >= 2, M any - win-back, incentive-driven
  • Lost: R <= 1, F any, M any - final reactivation or suppress

Campaign Mapping and Cadence

Once segments are built, map campaigns to each. Champions should receive your highest-touch, highest-value campaigns: early access to new products, loyalty rewards, and exclusive content. Send weekly or bi-weekly. Loyal customers get standard promotional campaigns and educational content, 1 - 2x per week. At-Risk customers get a dedicated win-back series: a soft re-engagement email, followed by a discount or incentive, then a final 'we miss you' message over 2 - 3 weeks. Space these out to avoid fatigue.

Lost customers are the hardest call. Some operators suppress them entirely to protect sender reputation. Others run a single, high-incentive reactivation campaign (e.g., 30% off) and then suppress. If you do target Lost, use a separate send list and monitor bounce rates closely. A 5% bounce rate on a Lost segment can tank your domain reputation.

Build automation workflows that move customers between segments. When a customer in At-Risk makes a purchase, they should automatically exit the win-back series and enter a post-purchase nurture flow. Use Klaviyo's conditional logic to trigger these transitions based on RFM score changes.

  • Champions: 1 - 2x weekly, high-value offers, VIP content
  • Loyal: 1 - 2x weekly, standard promotions, educational
  • At-Risk: 2 - 3 emails over 3 weeks, discount-driven
  • Lost: 1 final campaign or suppress entirely

Scoring Rules and Thresholds

Klaviyo's default RFM thresholds work for most DTC brands, but operators should validate them against their own data. If your median customer purchases every 45 days, Klaviyo's default Recency threshold (30 days for a score of 5) may be too aggressive. Run a cohort analysis: segment customers by purchase frequency and calculate the median days between orders. Use that to set realistic Recency tiers.

Frequency thresholds depend on your repeat rate. If 30% of customers make a second purchase, a Frequency score of 3 should represent that 30th percentile, not a fixed number like 3 orders. Klaviyo allows custom thresholds - use them. The goal is to ensure that R, F, and M scores are evenly distributed across your customer base, not bunched at the low end.

Monetary thresholds are auto-calculated by Klaviyo as quartiles, which is usually correct. However, if you have a small number of whales who skew the distribution, consider capping Monetary at a reasonable threshold (e.g., top 10% of spend = M5) to avoid over-indexing on a handful of customers.

  • Validate Recency thresholds against your median purchase interval
  • Set Frequency thresholds to match your repeat purchase distribution
  • Use Monetary quartiles unless whales distort the data - then cap at top decile

Monitoring and Optimization

Track segment performance weekly. In Klaviyo, pull a report on open rate, click rate, and conversion rate by RFM segment. Champions should have the highest open and conversion rates; Lost should have the lowest. If At-Risk has a higher open rate than Loyal, your messaging may be off - the win-back offer is too strong or the Loyal campaign is stale.

Monitor segment churn. Calculate the percentage of customers who move from Champions to Loyal to At-Risk each month. A healthy churn rate is 5 - 10% month-over-month. If 30% of Champions drop to Loyal in a single month, either your product quality is declining or your retention campaigns are ineffective. Dig into the data.

A/B test messaging by segment. Send two versions of a win-back email to At-Risk: one with a 20% discount, one with free shipping. Measure which drives higher conversion. RFM segments are large enough to support statistical significance, so use them as testing cohorts. Over time, you'll learn which messaging resonates with each segment.

  • Weekly: compare open, click, and conversion rates across segments
  • Monthly: track segment migration and churn rates
  • Quarterly: A/B test messaging and offers by segment

Common Pitfalls and How to Avoid Them

The biggest mistake is over-segmentation. Operators build 20 RFM segments and then struggle to create unique campaigns for each. Start with 6 - 8 core segments. Once you have data on performance, expand. Complexity without insight wastes time.

Second pitfall: ignoring data quality. If your Klaviyo list has duplicate profiles or ghost customers with no purchase history, RFM scores are meaningless. Before launching campaigns, deduplicate and clean your list. Merge profiles by email address and phone number.

Third: setting and forgetting. RFM thresholds should be reviewed quarterly. As your business grows, purchase patterns change. A Frequency score of 3 might have meant 3 orders in year one, but 2 orders in year two if your repeat rate declined. Update thresholds to keep segments balanced.

Finally, operators often suppress Lost customers too aggressively. A single reactivation campaign can be profitable even with a 2% conversion rate. Test a final win-back offer before suppressing entirely. The lifetime value of a reactivated customer often exceeds the cost of the email.

FAQ

How often does Klaviyo update RFM scores?

Klaviyo updates RFM scores nightly. If a customer makes a purchase today, their Recency and Frequency scores will update tomorrow. This means dynamic segments refresh overnight, so a customer who just purchased will automatically exit an 'At-Risk' segment and move to 'Recent Buyer' the next day.

Should I use static or dynamic RFM segments?

Use dynamic segments for ongoing retention campaigns (Champions, Loyal, At-Risk) so customers flow in and out automatically. Use static segments for one-time reactivation blasts or limited-time offers to prevent duplicate sends. For example, a 'Lost Customer Reactivation' campaign should be a static list snapshot to ensure each customer receives the offer only once.

What if my RFM scores are heavily skewed toward low values?

This usually means your repeat rate is low or your customer base is young. Check your Frequency distribution - if 80% of customers have F1 or F2, your product may have a retention problem unrelated to email. In the short term, adjust Frequency thresholds downward so F3 represents your median repeat customer. Long-term, focus on product and post-purchase experience to increase repeat rate.

Can I use RFM alongside other segmentation methods?

Yes. Combine RFM with behavioral segments (e.g., browsed category X), demographic data (e.g., location), or engagement metrics (e.g., email opens in last 30 days). For example, segment as 'Champions who browsed winter coats' for a hyper-targeted campaign. Just avoid over-segmenting - each additional filter shrinks your audience and reduces statistical power.

FAQ

How often does Klaviyo update RFM scores?

Klaviyo updates RFM scores nightly. If a customer makes a purchase today, their Recency and Frequency scores will update tomorrow. This means dynamic segments refresh overnight, so a customer who just purchased will automatically exit an 'At-Risk' segment and move to 'Recent Buyer' the next day.

Should I use static or dynamic RFM segments?

Use dynamic segments for ongoing retention campaigns (Champions, Loyal, At-Risk) so customers flow in and out automatically. Use static segments for one-time reactivation blasts or limited-time offers to prevent duplicate sends. For example, a 'Lost Customer Reactivation' campaign should be a static list snapshot to ensure each customer receives the offer only once.

What if my RFM scores are heavily skewed toward low values?

This usually means your repeat rate is low or your customer base is young. Check your Frequency distribution - if 80% of customers have F1 or F2, your product may have a retention problem unrelated to email. In the short term, adjust Frequency thresholds downward so F3 represents your median repeat customer. Long-term, focus on product and post-purchase experience to increase repeat rate.

Can I use RFM alongside other segmentation methods?

Yes. Combine RFM with behavioral segments (e.g., browsed category X), demographic data (e.g., location), or engagement metrics (e.g., email opens in last 30 days). For example, segment as 'Champions who browsed winter coats' for a hyper-targeted campaign. Just avoid over-segmenting - each additional filter shrinks your audience and reduces statistical power.

andrebrov