The LinkedIn outreach metrics that answer “is it working” are six: leads reached, connection request acceptance rate, conversations, total reply rate, interested reply rate and the count of unlabeled replies. Each one has a fixed numerator and denominator in GTM Bud, and this article gives them exactly as the product returns them, so that the number on your slide means the same thing next quarter.
It is written for a RevOps lead who builds the outbound slide of the board deck. The CRO asked a yes or no question on Tuesday, the deck is due on Friday, and the data you have is a list of activity counts. We read each definition below from the code that computes the results, not from a help page.
Why activity counts do not answer the CRO
“We sent 1,400 messages” describes effort. The CRO asked about effect. An activity count has no denominator, so it cannot go up or down in a way that means something. A slide of activity counts invites the next question, “and is that good?”, which you cannot answer from the slide.
A funnel number has two parts and a definition. The six below were chosen with one rule: each one must change for a different reason. If two numbers always move together, one of them is decoration.
The six numbers, defined exactly
| # | Metric | Numerator | Denominator | What it describes |
|---|---|---|---|---|
| 1 | Leads reached | Leads with at least one touch on any channel | None, it is a count | Volume |
| 2 | Connection request acceptance rate | Connection requests accepted | Connection requests sent | Targeting and sender fit |
| 3 | Conversations | Leads that got a message, or that replied | None, it is a count | Real exchanges started |
| 4 | Total reply rate | Leads that replied | Conversations | First message quality |
| 5 | Interested reply rate | Leads whose strongest reply is interested | Conversations | Offer and audience match |
| 6 | Unlabeled replies | Leads whose reply has no label yet | None on the slide | Trust in rows 5 and 4 |
1. Leads reached
Leads reached is the count of leads that got at least one connection request, direct message, InMail or email from the campaign. The result also gives a split by channel: connection request, InMail and email. One lead can count on more than one channel in that split, so the channel counts can add up to more than the total. Put the total on the slide and the split in the notes.
2. Connection request acceptance rate
This rate is accepted connection requests divided by sent connection requests. Each LinkedIn conversation of a lead counts once. Requests from the last days can still be pending, so a young campaign can read lower than it will read later. The two counts come from the latest results snapshot of the campaign, and the rate carries its own as-of time. Before the first snapshot exists, the rate is not available, and the result says so.
GTM Bud returns a target with this rate: a minimum, an OK level and a great level. Treat it as a management line that GTM Bud sets for its own product. It is not a forecast for your audience.
3. Conversations
Conversations is the count of leads that got a direct message, InMail or email, or that replied. A connection request alone does not start a conversation. This is the number that surprises people: a campaign can reach many leads and have few conversations, because most of its touches so far are requests that nobody accepted yet.
4. Total reply rate
Total reply rate is leads that replied, divided by Conversations. Note the denominator. It is not leads reached and it is not messages sent. If your current slide says “reply rate” and divides by requests sent, the two numbers are not comparable, and the footnote must say which one this is.
5. Interested reply rate
Interested reply rate is leads whose strongest reply is interested, divided by Conversations. “Strongest” means a lead counts once, under the best label it earned. Replies carry one of three labels: Interested, Warm or Not interested. Warm and Not interested have their own rates with the same denominator. For a board slide, Interested is the row that matters, and Warm belongs in the notes.
6. Unlabeled replies
The results include leads whose reply is not labeled interested, warm or not interested. This is a quality check on the slide itself. If a fifth of the replies have no label, the Interested rate can be understated by an unknown amount. Report the count in small type under the table. A board member will not ask about it. Your CRO should.
What the percent values look like
Each rate comes back as a percent with one decimal place, together with the two counts and a definition text. When the denominator is zero, the percent is empty, not zero. That difference matters on a slide: “no conversations yet” is a different message than “zero percent replied”.
How to pull the six numbers in one chat
Devika is the RevOps lead at a 110-person company. She has the GTM Bud connector in ChatGPT. On Wednesday afternoon she types:
“List my campaigns. For each active or paused campaign, get the results. Build one table with leads reached, the acceptance rate with accepted and sent, conversations, total reply rate, interested reply rate and the unlabeled count. Add both as-of times. Use the definitions that come back with each number, word for word, as footnotes.”
The AI calls the campaign list, then the results tool once for each campaign. What comes back for each campaign is the set of counts and rates above. Each has its label and its definition text, and each rate has its two counts. Devika asks for the definitions word for word because a paraphrase by the AI is one more place where “Conversations” can turn into “contacts”.
Then she asks for the one thing a board reads: change.
“Here are the same six numbers from the deck three months ago. Put the two periods side by side for each campaign and mark the rows where the direction changed.”
The earlier numbers come from her own deck. GTM Bud returns totals for the whole life of a campaign, so a period comparison is a subtraction that she keeps the inputs for.
How fresh the numbers are
The numbers refresh when someone asks and can be up to about 15 minutes old. The as-of time gives the exact time. If the last refresh failed, the result says so and states that the numbers are from the as-of time. For a slide, this is simple: print the as-of time under the table. A number without a time is the reason two people quote two values for one campaign.
What the connector cannot report
A slide that names its gaps survives the meeting. These are the gaps.
- Meetings booked. Not in the campaign results. Count them in the calendar or the CRM.
- Pipeline and revenue. Not in GTM Bud. If the same chat has a CRM connector, the AI can read opportunity amounts there. That is the AI and your CRM, not a GTM Bud function.
- Numbers by rep or by sender account. No tool returns which sender account a campaign or a reply uses. For a campaign with two or more LinkedIn senders, the campaign dashboard on gtmbud.com has a By sender table, and you copy from there.
- Reply labels by channel. Reach is split by channel. Interested, Warm and Not interested are not split between connection requests and InMails.
- Numbers by message variant. The results are per campaign. They do not rank one value offer against another.
- A date range. There is no “last 30 days” input. The results are totals.
- An export. GTM Bud does not send a scheduled report, and the connector has no file download. The AI writes the table in the chat, and you paste it.
A rule for the footnote: every rate on the slide gets three things under it: the numerator, the denominator and the as-of time. If one of the three is missing, the number is an opinion.
The slide, row by row
Devika’s slide has one table and two sentences.
The table has one row per campaign and the six columns above. Under it, in small type: the definition of Conversations, the as-of time, and the unlabeled count.
The first sentence answers the CRO: which stage improved since the last deck, and which did not. The second sentence names what the slide leaves out: meetings and pipeline come from the CRM slide that follows. She does not add an outside benchmark column. For definitions that the wider market uses, and how they differ, see our guide to outbound KPIs.
Frequently asked questions about LinkedIn outreach metrics
Which LinkedIn outreach metrics belong on a board slide?
Six: leads reached, connection request acceptance rate, conversations, total reply rate, interested reply rate and the count of unlabeled replies. Together they describe volume, targeting, message quality and the reliability of the labels. Activity counts such as messages sent describe effort only.
What is the denominator of the reply rate in GTM Bud?
Conversations. That is the number of leads that got a direct message, InMail or email, or that replied. A lead that got only a connection request is not in the denominator, so the rate is not the same as replies divided by requests sent. In a campaign that sends mostly connection requests, it is higher.
Does GTM Bud report meetings booked or pipeline?
No. The campaign results end at labeled replies. Meetings and opportunity amounts live in your calendar and your CRM. If your AI has a CRM connector in the same chat, it can read those numbers there and put them next to the GTM Bud numbers.
Why do two people quote different numbers for the same campaign?
Usually because they asked at different times. The lead counts carry an as-of time, and the acceptance rate carries its own. Write both next to the numbers on the slide, and a difference between two readers has a visible cause.
Should the slide show a benchmark next to each number?
Show your own earlier periods first. An outside benchmark mixes audiences, offers and account ages that are not yours. GTM Bud returns a target with the acceptance rate, and that target is a line to manage against, not a forecast for your campaign.
Put six defined numbers on the slide
LinkedIn outreach metrics answer the CRO only when each one has a numerator, a denominator and a time. Take the six from one chat, copy the definitions as footnotes, and say in one sentence what the slide does not cover. Setup takes a few minutes with the steps on GTM Bud for ChatGPT. To turn the same numbers into a weekly habit, read the 20 minute outbound review for RevOps, and for the controls around the numbers see AI LinkedIn outreach for RevOps on GTM Bud.