If a LinkedIn campaign gets fewer replies this week, changing the first message may be the wrong response. The audience may have changed, recent invitations may still be waiting for acceptance, or replies may belong to messages sent last week.
Read the counts, their timing and the message sequence together. In OutreachGenie Pro, open a campaign's Analytics tab and find campaign history below the existing metrics. It shows twelve Monday-to-Sunday weeks in UTC, invitation acceptance and reply trends, and a table for successive messages. The general dashboard trends remain available on Free.
Check what each denominator contains
Invitation acceptance rate and message reply rate measure different stages. Someone cannot accept an invitation that you never sent, and an accepted connection does not mean they received a campaign message.
In the campaign history, invitation acceptance is grouped by the week the invitation was sent. A later acceptance can therefore change an earlier week's result. Recent invitation groups have had less time to receive a response than older groups.
The weekly message view uses completed sends recorded in that week and observed replies recorded in that week. A reply this Monday can belong to a message sent last Friday. These weekly totals describe recorded activity in each period; they are not a fixed group of messages followed for an equal number of days.
That distinction matters with a quiet week. If you sent few new messages but received several older replies, dividing this week's replies by this week's sends can produce an unusually high rate. It is not evidence that the small batch of new messages performed that well. Read the underlying counts and inspect the relevant conversations before comparing weeks.
Understand how message numbers are assigned
OutreachGenie's campaign history numbers each prospect's completed messages in send order within the campaign. The first completed message is Message 1, the second is Message 2, and so on. Message 5 and later groups the remaining sends.
The table identifies send order. It does not identify which visual flow node sent a message. A branch, an edited flow or a prospect who reached a different path can make those concepts diverge. Use the campaign-editing guide when you need to understand changes to an active sequence.
A recorded reply is credited to the last message sent to that prospect before the reply. If a person received two messages before answering, the table credits the second one. The first message may still have helped. The table is a consistent attribution rule, not a causal test of the last message.
Only completed sends in the recorded history contribute to the send counts. A queued action or prepared message is not a completed send. Check the underlying campaign activity if a number differs from what you expected to happen.
Compare counts before comparing percentages
Consider this fictional sequence:
| Message | Completed sends | Credited replies | Recorded reply rate |
|---|---|---|---|
| Message 1 | 80 | 8 | 10% |
| Message 2 | 40 | 6 | 15% |
| Message 3 | 10 | 2 | 20% |
Message 3 has the highest rate, but it reached only ten people who had already passed through earlier stages. It would be a mistake to move its text to the start of the flow solely because 20% exceeds 10%.
The groups differ in size, prior exposure and selection. A later message might reach only people who remained eligible and had not replied. It also has less or more observation time depending on the campaign's schedule.
Keep a review note with the audience, send period, counts, follow-up delay and the change you are considering. Compare similar groups after the same response window. For a small campaign, reading the conversations may teach you more than a percentage movement based on one extra reply.
Match the next change to the weak stage
Few invitations accepted: inspect targeting, profile relevance and the invitation context before rewriting a follow-up that many people never reached. Check an older invitation group so pending responses do not look like failures.
Connections accepted, few messages completed: inspect the campaign's schedule and recorded activity. A delivery or eligibility issue needs a different fix from weak writing.
Messages completed, few replies: review whether the message gives this audience a concrete reason to answer. Look at several actual conversations and distinguish positive interest, polite refusals and requests to stop. A reply count alone cannot tell you whether the campaign produced useful conversations.
Later messages receive replies: inspect the full sequence. A reminder may help, but the table's last-message rule cannot prove that another follow-up will improve results. Preserve stop conditions and avoid extending the sequence just to increase the total number of responses.
For a broader measurement plan, use the LinkedIn reply-rate guide. Keep qualified conversations and business outcomes separate from an undifferentiated reply total.
Handle missing history before interpreting a zero
Free workspaces see the Pro preview. That is an access boundary, not evidence that their campaigns have no activity. On Pro, an empty twelve-week window means the history has no relevant recorded sends for that view; it does not describe all past LinkedIn conversations.
Check that you opened the intended campaign and workspace. If loading fails, use the retry control before drawing a conclusion. History depends on the events and reply observations available to OutreachGenie; it is not a complete export of your LinkedIn inbox.
Questions that come up in practice
Common questions
Where do I find campaign history in OutreachGenie?
Open the campaign's Analytics tab. Pro workspaces can see twelve Monday-to-Sunday weeks in UTC, acceptance and reply trends, and a per-message table. Free workspaces retain general dashboard trends and see a Pro preview for campaign history.
Does Message 2 mean the second node in my flow?
Not necessarily. Message numbers come from each prospect's completed sends in that campaign. They can differ from visual flow positions, particularly after branches or edits. Message 5 and all later completed sends share a grouped row.
Can I use the highest follow-up rate as proof of the best message?
No. Later messages reach different, often smaller groups, and the last-message attribution rule does not establish causality. Compare similar audiences with equal observation time and inspect the actual conversations.
Research used for this guide
Open the workspace and try it with a small, reviewed list.
Try the workflow