Percent of members who started a multi-step automation and reached the final email. Sharp drops = broken trigger or template-render error.
At a glance
The share of members who entered a multi-step automation (Customer Journey or Classic Automation) and reached its final email. Computed across active multi-step automations as completers divided by entrants. A multi-step journey only earns its full revenue if members move through all of its steps, so a falling completion rate means the back half of every journey, often the highest-converting offers, is being seen by fewer and fewer people. A gentle decline is usually natural attrition (people unsubscribe, lose interest, or convert and exit early). A sharp drop is almost always mechanical: a broken step trigger, a template-render error mid-journey, a wait-step misconfiguration, or a segment-exit rule that is ejecting members too aggressively. The card is read as a structural-health gauge for the automation programme, not a campaign performance metric.
| What it counts | Across active multi-step automations: (members who reached the final step ÷ members who entered the first step) × 100, blended into a single programme-level rate. Single-email automations are excluded, they have no “series” to complete. |
| Why it matters | The most valuable content in a journey is rarely the first email. Welcome offers, escalating discounts, and the strongest post-purchase cross-sells sit later in the sequence. A low completion rate means that high-value tail is reaching a fraction of entrants. Lifting completion lifts revenue from steps you have already built. |
| What drives natural attrition | Members unsubscribe, become inactive, convert and exit via a goal rule, or are removed by a segment-exit condition. Some attrition is healthy: a member who buys after email two should not need emails three through five. |
| What drives a mechanical drop | (1) A broken trigger between steps after a Customer Journey version migration. (2) A template-render error mid-journey, so every member queued for that step fails and stalls. (3) A wait-step set far too long, so members appear “incomplete” simply because they are still in the queue. (4) An over-aggressive exit rule ejecting members who should continue. |
| Currency | n/a, this is a percentage. The revenue impact surfaces in top-automations-by-revenue and automation-vs-campaign-revenue-mix. |
| Time window | 30D vsP (30-day rolling vs prior 30-day period). The vs-prior comparison is what surfaces a sharp drop. |
| Alert trigger | < 80% (a broken flow is likely below this threshold for most multi-step journeys), or a sharp drop versus the prior period. |
| Sentiment key | mc_automation_completion_rate |
| Roles | owner, marketing |
Calculation
Calculated automatically from your Mailchimp 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 skincare brand on Shopify running Mailchimp Standard with three multi-step Customer Journeys. Snapshot for the 30-day window ending Thursday 11 Jun 26, with the prior 30-day period for comparison.| Journey | Steps | Entrants | Completers | Completion rate | Prior period | Change |
|---|---|---|---|---|---|---|
| Welcome Series | 4 | 6,420 | 5,010 | 78.0% | 79.5% | -1.5 pts |
| Post-Purchase Nurture | 3 | 4,880 | 4,150 | 85.0% | 84.2% | +0.8 pts |
| Win-Back | 5 | 2,310 | 990 | 42.9% | 71.0% | -28.1 pts |
| Blended programme | 13,610 | 10,150 | 74.6% | 78.9% | -4.3 pts |
- The blended rate of 74.6 percent is below the 80 percent threshold and down 4.3 points vs prior. The headline says “investigate”, but the blended figure hides where the problem actually is. Decomposing by journey is the first move.
- Win-Back is the entire story. It collapsed from 71.0 percent to 42.9 percent, a 28-point drop. That is not natural attrition, attrition moves a point or two, not nearly thirty. This is a mechanical failure. Investigate: open the Win-Back journey and check (a) whether a step trigger disconnected after a UI migration, (b) whether step three or four throws a template-render error (send a test through each step), and (c) whether a wait step was lengthened, leaving members stuck mid-queue and counting as incomplete.
- Welcome and Post-Purchase are healthy. Welcome at 78 percent is slightly under the generic threshold but stable vs prior and entirely normal for a four-step welcome series, some new subscribers convert early and exit, which is the desired outcome. Post-Purchase actually improved. Neither needs action.
- The blended figure was misleading on its own. Two healthy journeys plus one broken journey averaged to a “soft amber” 74.6 percent, which understates how broken Win-Back is and overstates any problem with the rest. Always decompose before acting.
-
The revenue at risk is in the tail. If Win-Back’s final two steps carry its strongest re-activation offer, fewer than half of entrants are now seeing it. Cross-reference
top-automations-by-revenue, Win-Back’s contribution will have fallen over the same window.
- Decompose the blended rate by journey. The composite always hides a load-bearing weak point; find the journey dragging it.
- For the dragging journey, identify the step where members stall. A completion cliff between two specific steps points straight at the broken or mis-rendered step.
- Send a test through each step. A template-render error surfaces immediately in a test send.
- Check wait-step durations. A lengthened wait makes members look incomplete when they are simply still queued; this is a false alarm, not a break.
- Check exit and goal rules. An over-aggressive exit condition ejects members who should continue; a goal rule firing early is desirable (they converted) and not a problem.
| Time horizon | Action |
|---|---|
| First hour | Decompose the blended rate; identify which journey dropped and between which steps members stall. |
| First 4 hours | Send tests through the suspect steps; confirm whether the cause is a render error, a broken trigger, or a benign wait-step artefact. |
| First day | Fix and republish; for a wait-step artefact, confirm the rate normalises once queued members clear. |
| First week | Measure the 7-day-rolling completion rate for early recovery signal; review whether the journey’s exit rules are ejecting members prematurely. |
Sibling cards merchants should reference together
| Card | Why merchants reach for it |
|---|---|
automation-status | The status inventory of every automation. A completion drop and a status problem usually share a root cause. |
automation-stopped-firing-24h | The acute version of a mechanical failure: a journey that has stopped entirely. A sharp completion drop and this alert often fire together. |
welcome-series-completion-rate | The welcome-specific completion rate. The welcome journey is the most-watched single series; this card isolates it. |
win-back-automation-recovery-rate | Win-back completion and win-back recovery move together; a completion collapse drags recovery with it. |
top-automations-by-revenue | Where the revenue cost of a completion drop lands. Triage the broken journey that earns the most. |
automation-vs-campaign-revenue-mix | If automation revenue share is slipping, falling completion is a common cause. |
customer-journeys | The Customer Journey inventory; most multi-step automations are Journeys and stall step-by-step. |
engagement-funnel | The Sent to Converted funnel. Completion is journey-internal; the engagement funnel is the per-send view of the same drop-off logic. |
Reconciling against Mailchimp
Where to look in Mailchimp’s own dashboard:- Mailchimp → Automations → Customer Journeys, open a journey, and use the journey map view to see how many members are at each step. The drop-off between steps is the visual equivalent of the completion-rate decomposition.
- Mailchimp → Reports → Automations for per-automation and per-step send and engagement counts.
- The journey map’s per-step member counts are the raw material: entrants at step one versus members reaching the final step.
| Reason | Direction | What to do |
|---|---|---|
| Blended vs per-journey. Vortex IQ reports a programme-level blend; Mailchimp shows one journey at a time. | Not directly comparable | Decompose the Vortex IQ figure by journey, then compare each journey to its Mailchimp map. |
| In-flight members. Members still inside a wait step have entered but not completed; they depress completion until they exit. | Vortex IQ may read lower during long-wait journeys | A journey with a multi-day wait step will show lower completion simply because members are mid-queue, not dropped. Account for wait duration. |
| Goal-rule early exits. A member who converts and exits via a goal rule counts as not completing the email series, by design. | Vortex IQ may read lower | Early conversion is a good outcome, not a failure. A journey with a strong early-conversion goal will show lower completion and that is healthy. |
| Window alignment. Vortex IQ uses a 30-day rolling cohort; Mailchimp’s map shows lifetime-to-date member positions. | Either direction | Compare like cohorts; the rolling window will differ from the lifetime map. |
| Refresh lag. Completion recalculates each sync; the Mailchimp map updates as members move. | Vortex IQ moves slowly | Wait for the next sync; check last_synced_at. |