Segments

Segments are query-based cohorts computed from the shopper profiles currently stored in Zubby. Membership is evaluated when the query runs, rather than maintained as a separate member list. Source synchronization and scheduled profile updates can delay changes; the count is not an instantaneous view of activity in your store. Segments are the audience layer for broadcasts and the fastest way to inspect a cohort before acting on it.

Where it lives

Dashboard → Segments. The header shows your total shopper count. Below it: a custom-segment builder, your saved custom segments, and a grid of the built-in catalog and supported saved custom segments. Each catalog card shows the segment size, its share of all shoppers, a 30-day change indicator, a one-line commercial rationale, a View shoppers link, and a suggested playbook to activate it.

The built-in catalog

Six segments ship out of the box, each in a category that drives its badge colour:

Segment	Definition	Category
VIP · $500+ lifetime value	Lifetime spend ≥ $500.	vip
New customers (30d)	Exactly 1 order, placed in the last 30 days.	new
At-risk repeat buyers	2+ orders, last order 60–180 days ago.	at_risk
Dormant shoppers (180d)	No activity in the last 180 days.	dormant
High-intent prospects	3+ chat conversations, still 0 orders.	engaged
Marketing opt-ins with email	Consented to marketing and has a verified email.	marketing

The marketing opt-ins segment is the baseline audience for every broadcast and lifecycle send - keep it healthy, because hygiene matters more than raw size.

The 30-day change is an approximation

The “+/− vs. 30d” figure on each card is a fast single-query estimate of trend direction, not a true historical snapshot. Read it as “growing” or “shrinking,” not as an exact delta.

Custom segments

Build your own from the Build a custom segment card:

Give the segment a name that states its intent (“Lapsed 60d+, $200+ spend”) - you’ll be reading it in a list months from now.
Pick a field, an operator, and a value. The builder type-checks the pair before saving, so an unrunnable filter can never be stored.
Save. The segment appears below with a live member count - if the count looks wrong, check the units (spend is in cents, recency in days) before doubting the data.

Custom segments evaluate live, exactly like the built-ins. Note that broadcasts can target supported saved custom segments from your store as well as built-in audiences. Journey enrollment uses configured behavioral triggers rather than custom-segment membership.

Fields you can filter on

Field	Type	Note
total orders	number	Lifetime order count.
total spend cents	number	Lifetime spend - in cents (50000 = $500).
average order value cents	number	AOV in cents.
days since last order	number	Computed from last order date.
days since last visit	number	Computed from last seen date.
accepts marketing	boolean	Value is `true` or `false`.
lifecycle stage	string	Matches the shopper’s current stage.

Operators

Numeric fields accept the full set: =, ≠, >, ≥, <, ≤. String fields accept = and ≠ only. Boolean fields accept = only. The builder type-checks the field/value pair before saving, so an unrunnable filter can never be stored.

Money and time are not in obvious units

Spend and AOV are cents - filter on total spend cents ≥ 50000 for $500, not 500. Recency fields are days - days since last order > 60 means lapsed two months.

What segments power

Broadcasts - the segment dropdown in the broadcast composer is this exact built-in catalog and supported saved custom segments.
Shopper drill-in - View shoppers on each card opens the Shoppers list filtered to that live cohort.
Activation - each catalog card links to a suggested playbook (welcome series, win-back, browse abandonment, post-purchase thank-you) so a segment leads straight to a next action.

Recurring lifecycle sends belong in journeys, which enroll shoppers by behavioural trigger (cart abandoned, winback, post purchase) rather than by segment membership.

Gotchas

Membership is evaluated against currently stored shopper datawhen queried. Source sync and scheduled updates, including VIP recalculation, can delay changes. Preview counts can differ from the send-time audience as records, consent and delivery eligibility change.
View shoppers deep-links into Shoppers filtered by the segment, so you can inspect exactly who matches.
Custom segments are scoped to your store and editable only by owners and admins. Every create and delete is written to the audit log.
A custom filter spec supports up to 10 conditions combined with AND or OR; the quick builder on the page creates single-condition segments.

Shoppers, lifecycle & VIP - the profiles and lifecycle stages segments query.
Broadcasts - send a one-off campaign to a segment.
Journeys & sequences - enroll shoppers from supported behavior triggers.
Cart & browse rescue - recovery uses behavioural triggers alongside segments.

Choose inspection or activation

Use a saved custom segment to inspect shoppers who match a condition, then select a supported saved segment in the broadcast composer. Broadcasts accept built-in and supported store-owned saved audiences; channel consent and plan eligibility still apply. Journeys enroll from their configured event triggers. The quick builder creates one condition; do not assume it offers a multi-condition visual editor. Test membership with one shopper who should match and one who should not, checking cents versus currency units and days versus dates.

Contribution value and VIP are different filters

The advanced contribution_ltv_cents field is a Pro contribution-profit condition, not revenue CLV. It is available only when the shopper’s cost coverage reaches 90%; missing coverage does not mean zero value. The quick builder does not expose every advanced field. Built-in VIP spend criteria and configured VIP tiers are separate definitions, so verify the relevant card before treating them as interchangeable.