If you are working out how to test LinkedIn outreach, test the prospect list before you touch the copy. The list drives roughly 70% of outreach results. The words drive the other 30%.
Here is the situation we hear most often. Three weeks of sending, a few polite replies, no booked calls, and a document holding four rewritten versions of your opening message.
This post is for coaches running their own LinkedIn outreach, and for coaches checking whether an agency's testing is sound. Nothing below needs a tool you do not already have.
We cover why copy feels like the obvious first change, how to run two lists against one message, how long to leave a test, when copy testing finally earns its place, and what no amount of testing will shorten.
Copy is tempting because you control it
Copy is tempting because it is the one part of a campaign you can change in ten minutes. Building a list takes hours. Rewriting a first message takes one coffee. So it becomes the first change most coaches make, and it is the change that teaches them the least.
The problem is what a copy test measures. On a list of people who do not have the problem you solve, better words produce a clearer no.
The objection we hear is reasonable. Coaches say their message works fine in warm conversations, so the copy cannot be the issue. That is usually true, and it points straight at the audience.
A rewritten opener also hides the damage. Reply rate ticks up by half a point, you feel progress, and six weeks disappear before anyone asks who is actually on the list.
How to test LinkedIn outreach in the right order
Test in this order: the offer, your acceptance rate, the list, then the copy. Each step depends on the one before it. Skip a step and the next test gives you a number you cannot trust.
-
Check the offer. Outbound finds demand that exists. It does not create demand for something the market does not want.
-
Fix acceptance rate. Your profile has one job in outreach, getting the connection request accepted.
-
Test the list. Two audiences, one message, the same time window.
-
Test the copy, once one list is clearly ahead.
Low acceptance is a root problem, not a vanity metric. Fewer accepts means fewer people who can receive your offer, which means fewer conversations and fewer sales. Your profile needs credentials and social proof, not a pitch.
Across 1000+ campaigns across all industries, Coaching Accelerator tests the list first. The copy already follows a proven, tested framework, so the audience is the live variable.
How to run two lists against one message
Build two lists of similar size, send both the identical message sequence, and start them in the same week. One message, two audiences. Whichever list produces more positive replies tells you where your offer actually lands.
Make the two audiences genuinely different. Business coaches in London against business coaches in Manchester is not an A/B test of a prospect list. Change something that affects whether the person has your problem: industry, seniority, company size, or funding stage.
Track four numbers per list: accepts, replies, positive replies, and meetings booked. Positive replies matter more than total replies. A list that returns 20 replies and no interest is worse than one producing 8 replies and 3 calls.
Open the first 20 profiles on each list by hand before you send anything. Among coaches who built their own Sales Navigator lists, we keep finding students and career changers sitting behind job titles the filter marked as senior.
How long to run each list before judging it
Give each list at least three weeks and at least 100 accepted connections before you pick a winner. LinkedIn limits most accounts to roughly 100 connection requests a week, so two lists of 300 people take about six weeks to work through on a single account.
The sample floor exists because small numbers lie. At 30 accepted connections, one extra reply swings your reply rate by more than three points. At 150, the same reply moves it by less than one.
Replies also arrive late. A coach who reads your message on a Tuesday may answer the following Monday, after a full client day and a weekend. Calling a test after five days measures inbox habits, nothing else.
Do not touch the message mid test. Editing the opener in week two ends the test and restarts the clock, which is how most outreach testing quietly produces no answer at all.
Copy testing becomes useful once a list wins
Test copy after one audience is clearly ahead on positive replies. Now the question is useful: how do the right people respond to different words? Before that point you are measuring how the wrong people respond, and that answer changes nothing.
Change one element at a time, and start with the ask rather than the first line. Moving from a direct pitch to a short question usually shifts reply quality more than any rewritten opener.
Keep everything else identical, including send timing and follow up spacing. Two variables changed at once gives you a result with no cause.
If replies are strong and calls still do not convert, the copy is not your problem. The profile and the message generate the lead. Your website and your call structure do the closing.
What testing does not fix or shorten
Testing finds the best audience for an offer. It does not create demand for the offer itself. If two properly run list tests produce no segment with real interest, the positioning needs work before you add volume.
Market fit comes before volume, always. Sending more messages to more segments cannot rescue an offer the market does not want.
User interviews and go to market work only apply to early stage coaches still shaping what they sell and who they sell it to. If you have paying clients and a clear niche, skip that step.
Testing also does not shorten the sales cycle. Month one produces conversations and booked meetings, not signed clients. Coaching decisions take weeks, and a first call rarely ends in a yes.
Judge the cost over a year, not a month. One closed client typically covers a full year of outreach spend, and that is the comparison that matters when you are five weeks in and waiting.
Your next step if the numbers are poor
Pick two audiences, build both lists, and run your existing message against both for three weeks. You will learn more from that than from a month of rewriting first lines.
If you would rather not spend six weeks learning this by hand, it is the work Coaching Accelerator does for clients. Lists, profile optimization, outreach, and meetings booked into your calendar, with campaigns reviewed and adjusted every cycle instead of set once and left alone.
We have run this process for 300+ coaching businesses, and 95% of clients continue after their first month trial. If you want a second opinion on your current list, book a call and we will tell you what we would test first and why.