Back to blog

Email CRO

E-commerce Email Segmentation: Beyond Basic Demographics

March 18, 2026 · 10 min read

Most e-commerce brands segment their email lists by demographics — age, location, gender — and call it a day. That worked well enough five years ago. Today, it is table stakes at best and actively wasteful at worst. When every subscriber receives the same promotional blast regardless of where they are in their customer journey, open rates plateau, unsubscribes climb, and revenue per email stagnates. I have seen this pattern across dozens of Shopify and subscription brands, and the fix is always the same: move beyond demographics toward behavioral and predictive segmentation.

This is not a theoretical exercise. I will walk through the segmentation models I actually implement for clients — including the Klaviyo-specific mechanics — and show you the performance differences between blasts and properly segmented campaigns.

---

The Segmentation Hierarchy

Think of segmentation as a pyramid. Each layer adds precision — and revenue. Most brands live on the bottom layer and never climb higher.

Higher layers compound the layers below. You do not discard demographics — you build on top of them.

Demographic segmentation tells you who someone is. Behavioral segmentation tells you what they have done. Predictive segmentation tells you what they are likely to do next. The revenue impact compounds at each level — not because demographics stop mattering, but because behavior is a far more reliable predictor of future purchases than any demographic attribute.

A 35-year-old woman in London and a 35-year-old woman in Manchester might have completely different purchase behaviors, product preferences, and lifetime values. Treating them identically because they share an age and gender is where most segmentation strategies break down.

---

Behavioral Segmentation: Where the Leverage Lives

Behavioral segmentation uses observable actions — what someone browses, how often they purchase, which emails they open, whether they abandon carts — to group customers into meaningful segments. These signals are available in every Klaviyo account. The problem is that most brands never build segments around them.

The four behavioral dimensions I start with on every new client engagement are:

---

RFM Analysis: The Framework That Unifies Behavioral Segmentation

RFM stands for Recency (how recently a customer purchased), Frequency (how often they purchase), and Monetary value (how much they spend). It is a decades-old framework from direct-mail marketing that translates perfectly to e-commerce email. The idea is simple: score each customer on a 1-5 scale across all three dimensions, then use the composite score to create actionable segments.

Rather than trying to manage 125 possible RFM combinations (5 x 5 x 5), I consolidate them into the segments that actually drive different campaign strategies.

Consolidated RFM matrix. Each cell maps to a distinct email strategy — from loyalty rewards (Champions) to reactivation offers (Lost VIPs).

The power of RFM is that it turns three separate metrics into a single, actionable view of your customer base. A "Champion" customer (high recency, high frequency, high monetary value) should receive loyalty perks and early access. A "Can't Lose" customer (low recency but historically high frequency and spend) is your most urgent win-back priority — they were valuable and they have gone quiet.

In Klaviyo, you build these segments using custom properties synced from your e-commerce platform or calculated via a data warehouse. The key is updating RFM scores regularly — I run a weekly recalculation for most clients. Stale RFM data defeats the purpose of the framework entirely.

---

Subscription-Specific Segments

Subscription brands have an additional layer of segmentation complexity that one-time purchase brands do not. The subscriber lifecycle creates its own segments, each requiring fundamentally different email strategies. Treating active subscribers the same as at-risk subscribers is the email equivalent of ignoring a fire alarm.

These are the subscription lifecycle segments I build for every client running a subscription program.

Each stage maps to specific email flows and campaign strategies. The transition points between stages are where intervention matters most.

The transitions between these stages are where the money is made or lost. A new subscriber who skips their second order has a dramatically higher churn probability than one who receives both of their first two deliveries. Detecting that skip event and triggering an intervention email — not a generic promotional blast — is worth more than any demographic segment you can build.

For high-LTV subscribers, the strategy shifts entirely. These are your most valuable customers, and the last thing you want to do is annoy them with irrelevant promotions. I typically build a "VIP" Klaviyo segment defined by total revenue in the top 20th percentile and active subscription tenure of six months or more. This segment receives early access to new products, personalized content based on their purchase history, and absolutely no discount-led campaigns — they are already converted and loyal.

At-risk subscribers require a different approach again. In Klaviyo, I define "at-risk" as subscribers who have skipped two or more consecutive orders or whose email engagement (opens plus clicks) has dropped below 10% over the last 30 days. The intervention for this segment is not a discount code — it is a satisfaction check. A simple "Is everything okay with your subscription?" email with a one-click survey link outperforms discount-led retention emails by a significant margin in my experience.

---

Blasts vs Segmented Campaigns: The Performance Gap

The best argument for segmentation is the data. These are representative numbers from campaigns I have managed across subscription DTC brands. The pattern is consistent: segmented campaigns outperform blasts on every metric that matters.

Representative ranges from subscription DTC brands. Segmented campaigns consistently deliver 2-3x the revenue per recipient with a fraction of the unsubscribes.

The revenue-per-recipient gap is the metric that should get your attention. A segmented campaign targeting "Champions" with early access to a new product launch will typically generate $0.40-0.60 per recipient. The same product launch sent as a full-list blast — including disengaged subscribers, recent churns, and people who have never purchased in that category — drops to $0.08-0.15. You are not just leaving money on the table; you are actively damaging your sender reputation by emailing people who will not engage.

The unsubscribe rate difference matters more than most people realise. A 0.4% unsubscribe rate on a 100,000-subscriber list means 400 people leave with every blast. Over a year of weekly sends, that is 20,800 lost subscribers — many of whom might have converted with the right message at the right time. Segmentation does not just improve campaign performance. It preserves your list.

---

Klaviyo Implementation: Practical Steps

If you are on Klaviyo — and most Shopify subscription brands are — here is how I implement these segments in practice. Klaviyo's segment builder is powerful but requires thoughtful setup. The out-of-the-box segments Klaviyo suggests are a starting point, not a destination.

Step 1: Ensure your event tracking is complete. Before building a single segment, verify that Klaviyo is receiving the right events. At minimum, you need: Placed Order (with line items and revenue), Viewed Product, Added to Cart, Started Checkout, and — for subscription brands — Subscription Created, Subscription Canceled, and Order Skipped. If your subscription platform does not send these events natively, use Klaviyo's API to push them from your backend. Every segment I describe below depends on having clean, consistent event data.

Step 2: Build your engagement tiers. Create three segments based on email engagement over the last 90 days. "Engaged" profiles have opened or clicked at least one email in the last 30 days. "Semi-engaged" have opened or clicked in the last 31-90 days but not the last 30. "Unengaged" have no opens or clicks in the last 90 days. Every campaign you send should target Engaged and Semi-engaged at most. Sending to Unengaged profiles is a deliverability liability.

Step 3: Layer in purchase behavior. Using Placed Order events, create segments for: zero purchases (prospects), one purchase (one-time buyers), two-plus purchases (repeat buyers), and five-plus purchases (loyal customers). Cross-reference these with your engagement tiers. A "Repeat Buyer + Engaged" segment is your highest-value campaign audience. A "One-Time Buyer + Semi-engaged" segment is your highest-potential growth audience.

Step 4: Add subscription lifecycle properties. Sync subscription status (active, paused, canceled), subscription tenure (months since first subscription order), total subscription orders, and last skip date as custom properties on each Klaviyo profile. These properties power the lifecycle segments I described earlier. In Klaviyo, use profile properties rather than events for status-based segmentation — it is faster and more reliable.

Step 5: Build your RFM scores. Klaviyo does not natively calculate RFM scores, so you have two options. Option one: use Klaviyo's "Predictive Analytics" features (available on higher-tier plans) which provide a predicted customer lifetime value and churn risk score. These are not RFM per se, but they serve a similar purpose. Option two: calculate RFM scores externally — in a data warehouse, a Google Sheet with exported data, or a custom script — and sync the scores back to Klaviyo as custom properties. I prefer option two because it gives you full control over the scoring methodology and update frequency.

---

Segment-Driven Campaigns That Outperformed Blasts

Theory is useful but results matter more. Here are four campaigns where segmentation made a measurable difference, drawn from real client work.

---

Mistakes I See Repeatedly

Over-segmenting too early. Segmentation is powerful, but creating 40 micro-segments when you have a 5,000-person list is counterproductive. Each segment needs enough volume to produce statistically meaningful results. I start with four to six core segments and add granularity as the list grows. If a segment has fewer than 500 profiles, it is too small to run reliable campaigns against.

Ignoring engagement tiers in campaigns. This is the single most damaging mistake. Sending campaigns to your entire list — including profiles who have not opened an email in six months — destroys your sender reputation. Gmail and Outlook watch engagement signals closely. If a large percentage of your recipients never open your emails, future emails land in spam for everyone, including your engaged subscribers. Always exclude unengaged profiles from campaigns.

Static segments with no refresh cadence. A customer who was "at-risk" three months ago may have reactivated since then. If your segments are not updating dynamically, you are sending the wrong messages to the wrong people. In Klaviyo, use segment conditions based on rolling time windows (e.g., "has placed order zero times in the last 90 days") rather than fixed dates. And for custom properties like RFM scores, establish a weekly or biweekly sync schedule.

Treating all churned subscribers identically. A subscriber who canceled after one month because the product did not suit them is fundamentally different from one who canceled after 14 months because they wanted a break. The first needs product education or a different product recommendation. The second needs a "we miss you" message and perhaps a reminder of what has changed since they left. Segment your churned subscribers by tenure, cancelation reason, and LTV at the point of cancelation.

---

Where to Start

If you are currently sending campaigns to your full list with minimal segmentation, do not try to implement everything in this article at once. Start with these three changes that deliver the highest impact for the least effort:

First, create engagement tiers and stop emailing unengaged profiles in campaigns. This alone will improve your open rates, click rates, and deliverability within two to four weeks. It is the single highest-leverage change you can make.

Second, segment your subscribers by lifecycle stage. At minimum, separate new subscribers (first 60 days), active subscribers, and at-risk subscribers. Build one flow for each. The at-risk intervention flow will pay for itself within the first month.

Third, split one of your existing flows by product category. Pick your highest-revenue flow — typically cart abandonment or browse abandonment — and create category-specific variants. Even two or three variants will meaningfully outperform a single generic flow.

Once these three foundations are in place, you can layer in RFM analysis, predictive segmentation, and the more sophisticated campaigns I have described. But without engagement tiers, lifecycle stages, and basic behavioral splits, the advanced strategies have nothing to build on.

If you want help implementing these segmentation strategies for your e-commerce or subscription brand, get in touch. I will audit your current Klaviyo setup and show you exactly where the segmentation gaps are costing you revenue.