The outbound reporting dashboard: metrics worth tracking weekly

Most outbound reports are built to look busy rather than to be read. A dashboard that earns its weekly slot is short enough to scan in two minutes and ordered so that one bad number tells you which part of the machine to open.

What is an outbound reporting dashboard?

An outbound reporting dashboard is a single view of the metrics that track a cold outreach campaign from the moment an email leaves a mailbox to the moment a deal closes. It is not an analytics suite and it is not a monthly slide deck. It is a short chain of numbers, each one the denominator of the next, reviewed on a fixed cadence so that changes get noticed while they are still cheap to fix.

The structure matters more than the tooling. A spreadsheet with nine rows and a column per week beats a beautiful platform view that nobody reads on a Monday morning.

The nine metrics worth tracking weekly

Every number below sits downstream of the one before it. Keep that order on the page, because the order is what turns reporting into diagnosis.

• Sends. How many emails actually went out, per campaign and per mailbox. This is a capacity check, not a performance metric. A sudden drop usually means a paused sequence, an exhausted list or a mailbox that stopped sending.
• Delivery rate. Delivered divided by sent. Anything short of the high nineties means mail is being rejected or silently dropped, and every number below it becomes unreadable until you fix it.
• Bounce rate. Hard bounces divided by sent. Under 2% is where you want to sit. Above 5% you are actively damaging sender reputation, and the cause is almost always an unverified list rather than bad luck.
• Reply rate. Replies divided by delivered, not sent. Across B2B cold email, reply rates typically land between 1% and 5%, with the spread driven by how narrow the market is, how well the list is matched and how specific the offer sounds.
• Positive reply rate. The share of replies that show interest, plus the share of delivered emails that produce one. This is the number to protect. A campaign with plenty of replies and few positive ones has a targeting problem wearing the costume of success.
• Meetings booked. Positive replies that converted into a slot in a calendar. The gap between the two is almost entirely about reply speed and how the handover is written.
• Show rate. Meetings that actually happened. Most teams see somewhere between 60% and 85%, pushed upward by short lead times and reminders and downward by bookings made three weeks out.
• Opportunities created. Meetings that produced a real next step with a scope and a rough value. This is the first number a finance team recognises.
• Closed deals and pipeline value. The lagging number. It confirms whether the whole chain was pointed at buyers rather than at people who were merely polite.

How do you read the dashboard as a diagnosis?

Start at the top and stop at the first metric outside its usual range. Because each number depends on the one above it, a problem high in the chain corrupts everything below and will send you chasing symptoms if you start at the bottom. A campaign with no meetings and a 91% delivery rate does not have a booking problem. It has a mail problem that has not been named yet.

This is also why a dashboard beats a single headline metric. Reply rate on its own tells you something is wrong. Reply rate sitting under an intact delivery rate and beside a healthy positive share tells you where to go next.

Deliverability problem, targeting problem or copy problem?

Three failures produce similar-looking bad weeks and need completely different fixes. The dashboard separates them if you know the signature of each.

• Deliverability. Delivery rate falls, bounces climb, and reply rate collapses evenly across every segment, language and mailbox at the same time. Deliverability breaks suddenly and across the board. Confirm it with seed-inbox placement tests and authentication checks rather than guessing, which is the whole point of ongoing deliverability monitoring.
• Targeting. Delivery is clean and total replies look normal, but the positive share is thin and the negative replies say a version of "we do not do that". Wrong people, right machine. Fix the list and the filters before you touch a single sentence of copy.
• Copy. Delivery is clean, the list is verified and well matched, segments perform alike, and the total reply rate is simply low. Nobody is objecting because nobody is engaged. This is the only one of the three where rewriting the email is the correct first move.
• Offer. Replies, meetings and show rate all hold up, and then nothing becomes an opportunity. Outbound is doing its job and the conversation is failing. No sequence edit repairs this.

Two of these four have nothing to do with writing, which is worth remembering the next time a quiet week turns into a request for new subject lines.

A worked week

Numbers become useful when you chain them. Take a week of 2,000 sends to a verified list. At a 1.5% bounce rate, roughly 1,970 emails deliver. A 3% reply rate gives about 59 replies. If a third of those are positive, that is around 20 interested conversations. Convert half into meetings and you book 10; at a 75% show rate, 7 or 8 happen; if a third produce a real next step, the week added two or three opportunities to pipeline.

Run the same arithmetic in reverse and the dashboard becomes a planning tool. Two opportunities a week from 2,000 sends means a revenue target maps to a required list size, and the required list size tells you whether your market is even big enough for the volume you were planning. It also shows where effort pays: lifting the positive share from a third to a half moves the outcome as much as adding another 1,000 sends, without the deliverability cost.

What does not belong on the weekly dashboard

Every metric on the page costs attention, so the exclusions matter as much as the inclusions.

• Open rate. Privacy proxies pre-fetch tracking pixels and inflate the number, clients block them inconsistently, and the pixel itself can hurt deliverability. It is neither accurate nor free.
• Emails sent per rep. A workload metric dressed as a performance metric. It rewards volume, which is the one behaviour outbound does not need encouraging.
• LinkedIn impressions and connection counts. Useful for a content review, meaningless for pipeline.
• Total leads in the CRM. A count of rows, not of interest.
• Any metric nobody would act on. If a number moving 20% in either direction would change nothing you do, it is decoration.

How often should each number be read?

Weekly for the top of the funnel, monthly for the bottom. Sends, delivery, bounce and reply rate accumulate enough volume in seven days to mean something, and those are exactly the metrics where a week of damage compounds into a month of recovery. Meetings, show rate, opportunities and closed deals move too slowly to read weekly; look at them on a rolling four to eight week window so that one quiet week does not trigger a rebuild of a campaign that was working.

Sample size is the constant caution. Twelve replies is not a trend, and a positive share calculated on six replies will swing wildly week to week for reasons that have nothing to do with your campaign. Watch direction over several weeks and compare against your own history rather than a published average, which is the argument made at more length in our piece on outbound benchmarks.

Segment before you conclude

A blended campaign number hides the thing you need to see. Split the same nine metrics by ICP segment, by country and language, by sequence step and by sending mailbox, and the diagnosis usually falls out on its own. One mailbox with a collapsed delivery rate drags the campaign average down while three healthy ones look fine. One country segment replying at half the rate of the others is a language or positioning signal, not a copy problem. One sequence step producing most of the replies tells you where to spend the next edit.

Segmentation is also how you avoid the most common reporting mistake, which is switching denominators mid-comparison. Pick delivered as the base for reply rate, write it on the dashboard, and never quietly change it because a different base makes a week look better.

Who is the dashboard for?

Three audiences read the same page differently. The person running the campaign reads the top four rows and acts within a day. The founder or sales lead reads meetings, show rate and opportunities, and cares about whether the calendar is filling with the right companies. Whoever signs the invoice reads pipeline value against cost and wants a straight answer on payback. One page can serve all three if it is ordered by funnel rather than by whose department owns which number, and if it connects cleanly to the wider set of B2B sales KPIs the business already tracks.

If your agency will not show you these nine numbers on a fixed cadence, that is information too. We report the full chain to clients every week as part of the flat monthly retainer, including the weeks where the honest answer is that a segment did not work and we are changing it.

Build the dashboard before the campaign

The worst time to define your metrics is after a bad month, when every definition becomes an argument. Agree the nine rows, the denominators and the review cadence during setup, while nobody has a result to defend. Then the first weak week produces a diagnosis instead of a debate, and the campaign gets fixed while the fix is still small.