Analytics
What every metric on the Analytics tab actually measures, the difference between the unique-leads count and the per-step send count, and how to read a campaign's health at a glance.
The Analytics tab is the main place you'll go to see how a campaign is performing. This page explains what each number actually measures, the subtle distinction between unique-leads metrics and per-email metrics (they're both useful — for different things), and the troubleshooting moves when something looks off.
It's the default view when you open a campaign.
What's on the page
Top to bottom:
- Four KPI cards — Leads Contacted, Replies, Positive Replies, Booked Calls. Each has an info tooltip (
?icon) explaining what it counts. - Campaign Performance Trend chart — three area lines plotted over time, with a date-range selector.
- Per-step breakdown — one table per step in the sequence, showing how each variant at that step has performed.
The four KPI cards
Each card has an info tooltip on the label — hover the ? icon if you want the in-app definition alongside the number.
Leads Contacted
Unique leads who have received at least one email in this campaign's sequence. Shows as a fraction: 3,387 / 3,949 leads uploaded, with a progress bar so you can see how far the campaign has reached into the list.
This number is capped at your total leads uploaded — it can never exceed it. It counts people, not emails, so a follow-up sent to someone who already got Step 1 doesn't push the number up.
When the campaign has contacted every lead in the list (the fraction reads 3,949 / 3,949), the campaign keeps running while it sends follow-ups. Once every follow-up has gone out too, the campaign moves to Completed.
Replies
The count of replies received, plus a percentage showing reply rate against leads contacted.
The card has an Include auto replies toggle. It's on by default:
| Toggle state | What's counted |
|---|---|
| On (default) | All replies — including out-of-office and other auto-responders. Useful as a deliverability signal: an OoO at least proves the email reached an inbox. |
| Off | Manual replies only — no auto-responders. Use this when you want to see genuine human engagement. |
Numbers update as soon as you flip the toggle.
Positive Replies
Replies the AI has classified as showing genuine interest — "send me more info", "let's book a call", "interested, what are next steps", etc.
Below the count is the positive reply rate as a percentage of leads contacted. The card displays the campaign's target (the Healthy PRR threshold from the Options tab — defaults vary by method; podcast campaigns default to 1%).
A progress bar fills toward the target. Above target, the campaign-management agent considers the campaign healthy. Below target, it may flag the campaign as underperforming.
A "positive" reply is not just any reply — out-of-office, "not interested", and unsubscribes don't count. See Lead lifecycle for the full classification rules.
Booked Calls
The count of meetings actually booked from this campaign, plus the booking conversion rate (booked calls as a percentage of positive replies).
The card displays the target (default 25%, configurable on the Options tab). A progress bar fills toward the target.
If positive replies are healthy but booked calls lag, the bottleneck isn't your outreach — it's the handoff between reply and booking (calendar link, response speed, follow-up tone after the positive reply lands).
A note on "Leads Contacted" vs "Contacted" on the trend chart
These look like the same metric but they're not — and the distinction matters once you start reading the trend chart.
| Where | What it counts |
|---|---|
| Leads Contacted KPI card | Unique leads who've received at least one email. Capped at total leads uploaded. Counts people. |
| Contacted line on the trend chart | Emails sent per day — every send to every recipient, including follow-ups. Counts emails. Can spike to a few hundred a day even on a campaign with a small lead pool. |
The KPI tells you reach. The chart line tells you daily send volume. Both are useful — just for different questions.
The Performance Trend chart
Three area lines plotted over time:
| Line | What it shows |
|---|---|
| Contacted (blue) | Emails sent each day (touchpoints, including follow-ups) |
| Replied (orange) | Replies received each day |
| Positive (green) | Positive replies each day |
The time-range selector in the top-right narrows the window. Options: Last 7 days, Last 14 days, Last 30 days, Last 90 days, Last year, All time.
What to read in the chart:
- A steady upward trend on Contacted means the campaign is sending on schedule. A flat stretch where you expected sends usually means the campaign paused, hit a sending-window cap, or ran out of leads to contact at that step.
- A Replied line that tracks Contacted is the healthy pattern — replies should rise roughly in proportion with sends.
- A Replied line that flattens while Contacted keeps climbing is reply fatigue: later steps in the sequence aren't getting replies. Revisit the follow-up copy or shorten the sequence.
The per-step breakdown
Below the trend chart, every step in the campaign's sequence has its own table titled Step 1, Step 2, etc.
Each step's table shows one row per variant plus a Total row at the top aggregating across the variants. The columns:
| Column | What it shows |
|---|---|
| Variant | Variant label (A, B, C, …) with an enable/disable toggle to turn the variant on or off without leaving this page |
| Sent | How many emails of this specific variant went out |
| Replied | Reply count + reply rate as a percentage of Sent |
| Positive Responses | Positive-reply count + positive-reply rate as a percentage of Sent |
| Booked Calls | Booked-call count + booked-call rate as a percentage of positive responses |
Important: every metric on this table is per-email, not per-lead. Sent on the Step 1 table is "Step 1 emails sent", not "people who got Step 1". For most campaigns those are the same number, but on edge cases (paused-then-resumed campaigns, leads removed mid-sequence) they can drift apart.
How to use it:
- Find your winner. If one variant of Step 1 has a reply rate twice another's, the platform should be sending more through that variant. Disable the laggards via the toggle and add fresh variants modelled on the winner.
- Don't judge variants below the volume threshold. A variant with 100 sends still only has the minimum amount of data. The campaign-management agent only acts once a variant clears the Minimum Threshold of sends — see thresholds.
- Spot a dead step. If every variant of Step 3 has poor engagement, the timing or the message is the problem — not the individual variants. Often Step 3 needs to be shorter, sent sooner after Step 2, or removed entirely.
- Switch off underperforming variants with the per-variant toggle so future sends only go through the better ones. The historical data stays on the table either way — disabling stops the variant being used going forward.
You'll know your campaign is healthy when…
- Replied line tracks Contacted line on the trend chart — sends and replies both trending up.
- Positive Replies card progress bar is at or above its target (campaign-specific Healthy PRR).
- Booked Calls progress bar is climbing as positive replies accumulate.
- At least one variant per step has a clear lead in reply rate after a week or so of sending.
If most of those aren't happening, see the troubleshooting below.
If something's not right
My reply rate is zero after sending 1,000+ emails
This is almost always a deliverability issue, not the campaign content. Emails are going out but landing in spam.
Check in this order:
- Bounce rate on the Leads tab. Over 5% means the list is the problem; the campaign auto-pauses at 5% via bounce protection.
- Inbox health in Email Accounts. If health scores are below 80%, sender reputation is the bottleneck — see warmup.
- Domain authentication. If you're using inboxes purchased through Acquisity (pre-warmed or DFY), SPF, DKIM, and DMARC are already configured and this is rarely the cause. If Email Accounts shows an inbox Error, see domain provisioning.
- Send a test to yourself. If your own test lands in spam, every prospect gets the same. Run through the deliverability playbook.
The trend chart shows no activity for 3+ days
Check in order:
- Campaign status at the top of the detail page — Active, Paused, or Completed?
- Sending schedule on the Options tab — are the days/hours set so the campaign can actually send today?
- Inbox health in Email Accounts — are the inboxes attached to this campaign Active with green flames?
- Attention Needed banners at the top of the campaign detail — if there's one, it tells you the specific fix.
Step 2 (or 3) shows much lower reply rates than Step 1
Two common causes:
- Reply fatigue. A follow-up sent too soon after the opener can feel pushy, while one sent too late lands when the prospect has lost context. Try widening or narrowing the gap and watch the next batch.
- Repeating Step 1's argument. If the follow-up just rephrases the original pitch, recipients tune out. The strongest follow-ups add a new angle, a question, or a softer ask.
One variant has way more sends than the others
That's the platform working as designed. Once a variant clears the Healthy PRR threshold, the agent may disable weaker variants if AI campaign management allows variant optimization. Or you may have manually disabled the others via the per-step table toggles. Either way, the variant with most sends is your current champion.
The Replies count looks wrong
Toggle Include auto replies off and on once. The number recalculates each time. If you're surprised by the count with the toggle on, you're seeing OoO and bounce auto-responders included — which is the default. Off-only counts manual replies.
Related
- AI Agent tab — what the AI thinks of these numbers and what it can act on automatically
- Sequences tab — where you edit the variants whose performance shows on this page
- Variants and A/B testing — the full theory behind running variants
- Thresholds — the workspace-level numbers that drive Healthy PRR and target booked-call rate
- Deliverability — when zero replies means something's broken
- Lead lifecycle — what makes a reply count as "positive"