How many people are leaving your list, plotted over time. Spikes almost always sit one or two days after a specific send, which is exactly what makes the cause easy to find.
At a glance
A time-series of the unsubscribe COUNT across the selected period, drawn as a line so you can read the rhythm of churn and spot the days it jumps. This is the count, not the rate and not the alert: the Unsubscribe Rate card normalises these unsubscribes against how many emails went out, and the Unsubscribe Spike card fires when a single point breaks out of the normal band. The value of the raw count line is timing: because unsubscribes register immediately when a recipient clicks, a spike on the line almost always lines up with a send a day or two earlier, which makes the offending campaign or flow easy to identify.
| What it counts | The number of profiles that unsubscribed in each interval of the selected period, plotted over time. A voluntary opt-out, not a bounce or spam-driven suppression. |
| API endpoint + statistics field | Derived from consent changes via GET /api/profiles and list membership via GET /api/lists, bucketed per interval. Optionally cross-read against the unsubscribe metric event through GET /api/metrics. |
| Distinct from rate and alert | Count, not percentage. The Unsubscribe Rate divides these by sends; the Unsubscribe Spike is the breakout alert on top of this series. |
| Lists vs segments | Measures profiles leaving lists (static opt-in groups). Segment exits are automatic rule changes, not unsubscribes, and are not counted. |
| Consent status | Counts the transition to unsubscribed. Suppression from hard bounces or spam complaints is a separate signal on the suppressed-profiles card. |
| Chart type | Line (time-series). |
| Time window | Selected period, bucketed by day or week. |
| Alert trigger | The line itself surfaces movement; the breakout alert is handled by the dedicated Unsubscribe Spike card. |
| Roles | owner, marketing |
Calculation
Calculated automatically from your Klaviyo data. See the At a glance summary above for what the metric tracks and the worked example below for a typical reading.Worked example
A pet-supplies brand on Shopify sending three campaigns a week. The 30-day window covers 14 Mar 26 to 12 Apr 26. All figures are illustrative.| Day (in window) | Unsubscribes | Send that day |
|---|---|---|
| Typical campaign day | 40 to 70 | One segmented campaign |
| Non-send day | 5 to 15 | Flows only |
| 26 Mar | 310 | Full-list “clearance” blast |
| 27 Mar | 95 | Tail of the blast |
| 11 Apr | 58 | Normal segmented campaign |
- The 26 Mar point is a clear spike, about 5.6x a normal campaign day. The line makes it unmistakable. The cause is the full-list clearance blast sent that day: a broad, untargeted send to the entire list, including disengaged profiles who do not normally receive mail.
- The spike has a one-day tail. The 27 Mar reading of 95 is recipients who opened the blast late and unsubscribed the next day. A spike followed by an elevated next day is the classic single-send signature.
- Baseline churn is healthy and steady. 40 to 70 on campaign days and single digits on flow-only days is a normal rhythm. That steadiness is what makes the spike legible; without a stable baseline you cannot tell a spike from noise.
- The count line tells you when; the rate card tells you whether it was disproportionate. 310 unsubscribes off a full-list blast to, say, 60,000 recipients is a different story than 310 off a 5,000-recipient segment. Read the Unsubscribe Rate alongside to judge severity per send.
- The lesson is targeting, not silence. The fix for a clearance-blast spike is not to stop emailing, it is to stop blasting the full list. Segment to engaged profiles, suppress long-dormant ones, and the same promotion will churn a fraction of the audience while protecting deliverability.
Sibling cards merchants should reference together
Unsubscribe Trend is the count-over-time view of churn. Pair it with these:| Card | Why pair it with Unsubscribe Trend |
|---|---|
| Unsubscribe Rate | The normalised percentage. A spike in count may be fine if sends rose; the rate separates volume from genuine churn. |
| Unsubscribe Spike | The alert layer on this series. The trend shows the shape; the spike card fires when a point breaks the band. |
| Added vs Unsubscribed | Sets churn against acquisition so you can see net list movement. |
| Subscriber Growth Rate | The net outcome. Sustained high unsubscribes drag the growth rate down. |
| List Health Summary | The condition roll-up that churn feeds into. |
Reconciling against Klaviyo
Where to look in Klaviyo:- Analytics → Metrics → Unsubscribed (or Unsubscribed from List) for the event-level time-series of opt-outs, the closest direct comparison to this line.
- Audiences → Lists & Segments → [the list] → Growth for the removal side of the per-list growth chart.
- Campaigns → [a specific campaign] → Performance to attribute a spike to a single send, where Klaviyo reports the unsubscribe count for that campaign directly.
| Reason | Direction of divergence |
|---|---|
| Time-zone. Klaviyo buckets in account time zone; Vortex IQ buckets in UTC. A point near midnight can move a day, which matters most when attributing a spike to a send. | Per-day count shifts; window total close. |
| Suppression vs unsubscribe. Hard bounces and spam complaints suppress a profile but are not voluntary unsubscribes. This line counts opt-outs only. | Excludes suppression, so lower than a generic “removals” figure. |
| List vs account scope. A profile can be on several lists; an account-level unsubscribe differs from leaving a single list. Klaviyo’s per-list and global views can disagree, and so will this card depending on scope. | Variable. |
| Metric availability. If the merchant does not track the unsubscribe metric event, the count is derived from consent changes, which can bucket slightly differently from the event stream. | Marginal. |
| Page caps. Profile and membership reads paginate at 50 per page; large windows are assembled across pages with minor mid-pull drift possible. | Marginal. |