Sent → Delivered → Opened → Clicked. Drop-offs expose deliverability vs content weakness.
At a glance
The shape of the funnel tells you what’s wrong. Sharp drop Sent→Delivered = list-hygiene or sender-reputation problem. Sharp drop Delivered→Opened = subject-line or content-relevance problem (now muddied by MPP, see below). Sharp drop Opened→Clicked = creative or CTA problem. The diagnostic cost-of-investigation is shaped by which step the funnel narrows hardest at.
| What it counts | A 4-step waterfall: recipients (sent) → deliveries → unique_opens → unique_clicks. Each step is an absolute count; the rate from one step to the next is the diagnostic. |
| Definition of “sent” | Klaviyo’s recipients count, the number of delivery attempts the platform made. Includes bounces; excludes contacts that were on the suppression list at send time. |
| Open rate basis | Klaviyo’s unique_opens / deliveries. Counts each recipient at most once per send regardless of how many times they opened. iOS Mail Privacy Protection (MPP) inflates this stage by ~10-15 percentage points for brands with significant Apple-Mail recipient share. |
| Bounce handling | Bounces are the gap between Sent and Delivered. The funnel surfaces bounce volume implicitly through the Sent→Delivered drop; the Bounce Rate card breaks bounces down by hard/soft/reason. |
| Attribution model | n/a, this is a behavioural funnel, not a revenue-attribution card. Pair with Revenue per Recipient and Conversion Rate for the revenue-side view. |
| Currency | n/a. |
| Sent→Delivered healthy band | >=98 percent. Below 96 percent is amber (list-hygiene drift). Below 95 percent is red (active deliverability issue). |
| Delivered→Opened healthy band | >=20 percent post-MPP for ecommerce verticals (15-25 percent inflation built in). Pre-MPP benchmarks of 12-18 percent no longer apply meaningfully; brands need to re-baseline against their own post-MPP history. |
| Opened→Clicked healthy band | >=8 percent (click-to-open ratio) for engaged ecommerce sends. Below 5 percent suggests the email opens but the CTA does not pull; below 3 percent suggests the recipient regrets opening. |
| Sent→Clicked composite | The end-to-end click rate, typically 1-3 percent for ecommerce campaigns and 5-15 percent for triggered flows. The blended figure is what most brands quote externally; the per-step view is what reveals where to optimise. |
| Time window | 30D blended. Per-campaign breakdown via Campaign Status Breakdown. |
| Alert trigger | - (no single trigger; sub-step alerts fire via dedicated cards). |
| 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 UK fashion brand on Shopify with 64,000 active Klaviyo subscribers running 4 campaigns per week plus 5 always-on flows. Campaign-only snapshot for the 30-day window ending Tuesday 14 May 26.| Funnel step | Count | Step rate | Compared to brand baseline |
|---|---|---|---|
| Sent (recipients) | 768,000 | , | normal |
| Delivered | 729,600 | 95.0% | down 3 points from 98 percent baseline |
| Unique opens | 153,200 | 21.0% (open rate) | normal post-MPP |
| Unique clicks | 14,592 | 9.5% (click-to-open) | normal |
| End-to-end click rate | 1.9% | down from 2.4 percent baseline |
- The break is at Sent→Delivered. A 95 percent delivery rate means 5 percent of sends bounced or were rejected by receiving ISPs. That’s the entire drop in end-to-end click rate; opens and clicks per delivered email are normal. The funnel is not telling us we have a content problem; it’s telling us we have a deliverability problem.
- Why this matters operationally. The brand might have noticed “click rate is down 20 percent” and started workshopping new subject lines, new creative, new CTAs. That work would not move the needle; the issue is upstream. The funnel view prevents misdirected effort.
- What’s actually causing it. 5 percent bounce rate this far above the 2 percent healthy baseline points to either (a) accumulated hard-bounced contacts not yet suppressed, (b) a recent list-acquisition push that brought low-quality addresses, or (c) sender-reputation degradation at receiving ISPs that’s increasing rejections-on-arrival. The Bounce Rate card breaks this down by reason; the Alert Sender Reputation card catches the third pattern.
- What 21 percent open rate at Step 3 means. Post-MPP, this is in the green band for fashion brands. The opens are inflated by MPP-triggered preloads (estimate ~15 percentage points of the 21 percent are MPP rather than human); the human-confirmed engaged share is still meaningful, around 6 percent of delivered. Don’t try to “improve” opens; the metric is too noisy post-MPP to optimise against.
- What 9.5 percent click-to-open means. This is the metric that genuinely matters post-MPP. Click-to-open ratio (clicks divided by opens) is robust against MPP because the same MPP-triggered open inflation appears in both numerator-eligible cohort and denominator. 9.5 percent click-to-open on fashion campaigns is healthy; brands with strong creative push 12-15 percent.
- End-to-end click rate at 1.9 percent is the executive-summary number. The brand should treat this as the headline funnel-health metric and use the per-step view to diagnose causes. The recovery target after fixing deliverability: 2.5 percent end-to-end click rate within 30 days.
- Identify which step has the largest unhealthy drop. Compare each step rate to the brand’s own 90-day baseline rather than to industry averages; brand-specific patterns are more reliable diagnostics.
- Match the narrow step to the cause class. Sent→Delivered = deliverability or list-hygiene. Delivered→Opened = subject-line, send-time, or sender-name (with MPP caveat above). Opened→Clicked = email content, CTA, offer.
- Cross-reference with the Email Health KPIs composite. A narrow Sent→Delivered step paired with amber-bounce-rate and amber-delivery-rate confirms list-hygiene. Same narrow step paired with red-spam-complaint-rate confirms ISP-reputation degradation.
- Plan the fix at the upstream layer. Don’t optimise content for a deliverability problem; don’t optimise deliverability for a content problem. The funnel is the diagnostic; the fix happens elsewhere.
Sibling cards merchants should reference together
- Email Health KPIs is the rolled-up composite that summarises whether any step in this funnel needs attention. Glance at the composite first, drill into the funnel when amber.
- Bounce Rate and Delivery Rate decompose the Sent→Delivered drop into hard bounces, soft bounces, and ISP rejections.
- Open Rate isolates the Delivered→Opened step, useful for trend analysis even with the MPP caveat applied.
- Click Rate and Click-to-Open Rate decompose the Opened→Clicked step. Click-to-open is the post-MPP-resilient metric.
- Conversion Rate extends the funnel one further step: Clicked → Converted (placed order). The full Sent→Converted view is the commercial-impact version of this engagement-only funnel.
- Revenue per Recipient is the headline commercial metric this funnel ultimately drives. Use the funnel to diagnose; use revenue per recipient to measure impact.
- Campaign Status Breakdown for the campaign-by-campaign view of the same funnel, surfacing whether issues are systematic or campaign-specific.
- Flow Step Drop-off is the flow-side counterpart, showing where in multi-step automated sequences subscribers fall out.
Reconciling against the vendor’s own dashboard
Where to look in Klaviyo’s own dashboard:- Analytics → Email Performance Dashboard shows the same step counts (Sent, Delivered, Opens, Clicks) for the merchant’s account-wide view across the configured time window.
- Per-campaign view: Analytics → Campaigns → individual campaign → Performance. Klaviyo presents the same funnel at campaign granularity, useful for isolating campaign-specific drop-off patterns.
- Per-flow view: Flows → individual flow → Analytics tab. Funnel-equivalent presented per flow message rather than per campaign.
- Aggregation period boundaries. Klaviyo’s dashboard rolls up to calendar-day boundaries in the merchant’s account time zone; Vortex IQ uses the merchant’s profile time zone for boundary alignment. A campaign sent at 11pm UK time on Friday appears in Friday for Klaviyo (UK timezone account) but Saturday for Vortex IQ (US-East profile), accounting for small per-day step-rate differences.
- Send-vs-delivery sync timing. Klaviyo updates delivered counts in near-real-time from receiving-ISP confirmations; Vortex IQ syncs every 30 minutes. During an active send, the two views can show slightly different counts that converge once both have ingested the same delivery confirmations.
- Open-tracking pixel reliability. Both systems rely on Klaviyo’s open-tracking pixel; the underlying counts are identical. Differences appear only in how the rate is rendered (e.g. open rate based on deliveries vs based on recipients), and Vortex IQ uses Klaviyo’s recommended denominator (deliveries) consistently.
- Funnel-step inclusion rules. Klaviyo’s UI sometimes hides funnel steps where the count is zero or near-zero; Vortex IQ surfaces all four steps regardless to keep the diagnostic view consistent.
- Klaviyo
klv_engagement_funnelvs Mailchimp, Brevo, SendGrid equivalents: definitional twins on different platforms. Brands operating multiple ESPs for distinct audiences should compare funnel shapes; identical merchant audiences across two platforms should produce nearly-identical funnel rates, and meaningful drift signals platform-side issues at one or the other. - Klaviyo funnel last-step (clicks) vs Shopify or BigCommerce session data: Vortex IQ joins Klaviyo’s click count with the merchant’s ecommerce-platform session data on the same UTM-tagged Klaviyo links. The platform-side data should show roughly the same click count (within attribution-window differences); large gaps indicate UTM-tagging issues or pixel-tracking gaps on the merchant’s site.
- Klaviyo funnel composite metrics vs SendGrid transactional funnel: NOT definitional twins. Marketing email funnels and transactional email funnels behave differently. Useful for triangulation but not for direct comparison.