Capability: Campaign performance dashboard

Capability: Campaign performance dashboard

Direct answer: Ask "show campaign performance for my active campaign" (or name a specific one — "how is Q3 Fintech Outreach doing?") and nqzai pulls that campaign's delivery data directly from the send log, computes the funnel and rates, and opens it as a report panel in the same turn. There's no separate analytics tab to navigate to and no CSV export to stitch together — the numbers come back in the chat response itself, with a chart underneath.

This is one of the read-only tools in the Campaigns group, alongside creating a campaign, listing campaigns, assigning contacts, and launching/pausing/resuming. It costs nothing beyond the turn — there's no per-lookup charge, because it's reading data nqzai already has, not calling out to buy anything.

What it actually measures

Direct answer: The dashboard is built on two distinct data sets, and it keeps them separate on purpose.

The delivery funnel — sent, opened, clicked, replied, bounced — is scoped strictly to the campaign you're looking at. Every stage comes from the record of emails this campaign sent, so a contact's activity on a different campaign can never leak into these numbers. Open rate and reply rate are both computed against the same denominator (emails actually sent by this campaign), which matters more than it sounds: a reply rate quoted without the send count behind it is close to meaningless, and the dashboard is built to never show one without the other.

The contact lifecycle — new, contacted, opened, replied, interested, unsubscribed — is a different thing entirely. It reflects where each contact currently sits in their overall status with you, which they carry between campaigns. If someone opened an email from a campaign last month and is now sitting in a new one that hasn't sent anything yet, their lifecycle status can say "opened" while this campaign's funnel correctly shows zero sent. That's not a bug in the dashboard; it's two honest views of two different things, and mixing them into one number would produce exactly the kind of contradiction ("opened: 1, sent: 0") that a merged view can't explain. The dashboard shows both, labeled separately, rather than picking one and hiding the discrepancy.

Alongside those two, the dashboard also shows:

• Audience — how many contacts are currently in the campaign.
• Cadence — how many emails have gone out today against the daily send cap, the configured send window and time zone, which days of the week it sends, and how many contacts are currently active in an enrollment.
• Campaign state — draft, active, or paused, and when it was launched.
• Goal (when one is set on the campaign) — the target type and figure you defined for it.

Zero doesn't mean "failing"

Direct answer: A campaign that hasn't launched yet, or has no contacts, or is live but hasn't sent its first batch, will show a wall of zeros — and a dashboard that just prints those zeros with no context reads as a broken or failing campaign even when nothing is actually wrong. nqzai's dashboard was rebuilt specifically to name why the numbers are zero, rather than leaving you to guess:

• No contacts assigned — every metric is zero because there's no audience yet, not because outreach underperformed. The next step is adding contacts.
• Contacts assigned but not launched — the audience is loaded and ready; the zeros just mean sending hasn't started. The next step is to launch.
• Launched but nothing sent yet — this usually means the send window, daily cap, or enrollment step is holding things up rather than the audience ignoring the mail. The next step is checking the cadence settings or the enrolled contacts.
• Actually sending, with data — here the dashboard gives you a real read: replies landing means the offer is resonating; opens without replies means the subject line is working but the body or ask needs work; neither yet just means it's early or worth revisiting the subject line.

Each of those states comes back with a plain-language note and one concrete next action, not just a number.

What the response looks like

Direct answer: The chat reply leads with a one-line verdict tied to the actual numbers — something like "opens are healthy (34%) but no replies yet" or "converting — 6% reply rate off 812 sent" — followed by the sent/open/reply/audience line and the day's send progress against the cap. The report panel that opens alongside it adds:

• A signal-overview row of cards (open rate, reply rate, audience, campaign state), sorted so whatever needs attention most surfaces first.
• A bar chart of the funnel stages. If nothing has sent yet, this renders as an explicit "no funnel activity yet" placeholder instead of an empty chart with no bars — an empty axis with nothing on it reads as a rendering bug, not as "nothing happened yet," so the dashboard says the latter outright.
• A funnel table with the raw counts per stage.
• A campaign snapshot card repeating status, audience, sent count, open rate, and reply rate together.
• The cadence line: today's sent count against the daily cap, and the send window.

Illustrative example

(Illustrative — this is a constructed example to show the shape of the output, not a real customer's data.)

Asking "how is Q3 Fintech Outreach doing?" on a campaign that's been live for two weeks might come back as:

"Q3 Fintech Outreach" is converting — 5% reply rate off 812 sent. Sent 812 · Open 31% · Reply 5% · 950 in audience Funnel: sent 812 → opened 252 → clicked 47 → replied 41 → bounced 9 Today 38/50 sent · window 9:00–17:00 America/New_York

The report panel would show open rate and reply rate as the two lead cards (both colored to reflect that opens are solid and replies are actually landing), a bar chart across the five funnel stages, and a cadence note that today's send is still under the daily cap. If the same campaign were still a draft with 950 contacts loaded but never launched, the response would instead lead with "isn't launched yet — 950 contacts ready" and skip straight to a launch suggestion, rather than showing five stages of zeros.

Getting there from chat

You don't need to know a tool name or open a separate analytics screen — natural language routes straight to it: "show campaign performance," "how's my active campaign doing," or naming a specific campaign by name all resolve to the same dashboard. Leave the campaign unspecified and it defaults to whichever one is currently active. After it responds, follow-up suggestions adapt to the state you're in: an active campaign offers to pause it or view its contacts; a paused one offers to resume; a draft with contacts loaded offers to launch; a draft with no audience offers to go find leads first. The whole loop — check performance, decide what state the campaign is in, take the next action — stays inside the same conversation.

Where it fits

Direct answer: The dashboard is deliberately read-only: it doesn't change anything about the campaign, it just tells you where it stands. That makes it safe to check as often as you want without any risk of accidentally sending, pausing, or altering anything. It sits between the tools that build a campaign (create, assign contacts, set a sequence) and the ones that act on what it shows you (launch, pause, resume, draft follow-ups) — the dashboard is the read that tells you which of those to reach for next.

Evidence, limits, and reproducible use

Direct answer: Reproducible workflow. Provide the intended audience and campaign context, inspect the proposed records or draft, then explicitly approve any action that changes a campaign or sends email. Keep suppression, consent, and sender-domain decisions outside a generated draft and review them before use.

Limit. This result does not establish identity, consent, legal basis, deliverability, or a right to contact a person. A draft or campaign configuration is a review artifact, not an instruction to send.

For the currently exposed nqzai workflow and connection limits, check the public capabilities inventory before relying on a result.

Primary references

• FTC CAN-SPAM compliance guide
• Google sender guidelines