How to find which prospects reply to your outreach

Q: Should you measure reply rate or meeting rate?

Use both. Reply rate counts every reply, including 'not interested', so it shows who engages. Meeting rate shows who buys, but meetings are rarer, so you need larger groups before the numbers mean anything. Start with reply rate and confirm with meeting rate once you have enough meetings.

To find which prospects reply, sort the people you've already contacted into groups, such as seniority, industry, or company size, then compare the reply and meeting rate for each group. Only trust a difference once each group has at least 30 people. Then build your next list from the groups that reply most.

Most teams try to fix low reply rates by rewriting the message. But the list is often the bigger lever. If directors at 50-person companies reply three times as often as VPs at enterprises, no rewrite of the VP message will close that gap. Your results already show which people answer. You just have to sort them.

What should you compare?

Pick one question at a time. Good first questions are about who the person is, because that's what you control when you build the next list.

Question	Example tags
Do directors reply more than VPs?	Director, VP or above, Everyone else
Which roles reply most?	Sales, Marketing, Engineering, Operations
Do smaller companies reply more?	Under 50 employees, 50–500 employees, Over 500 employees
Which industries book meetings?	Software, Financial services, Healthcare, Other
Does the source matter?	Engaged with a post, Event attendee, Cold list

Keep each question to three or four tags. More than that and every group gets too small to read.

How do you tag prospects consistently?

Write tags so every prospect fits exactly one, and always include a catch-all like "Everyone else" or "Other." Without it, the people who don't fit get forced into the nearest tag and skew its rate.

Tag from what's on the profile: title, headline, company, and company size. Decide edge cases once and write them down. Is a "Head of Sales" a director or a VP? Whichever you pick, apply it to everyone.

For a few dozen prospects, a spreadsheet column works. For hundreds or thousands, use an AI classifier that reads each profile and picks one of your tags. Decision models like TypeSafe's Jev are built for this: they pick from a fixed list instead of writing text, so they can't invent a tag, and tagging a thousand prospects costs about a cent.

How many prospects do you need before trusting a reply rate?

Use these thresholds:

Fewer than 10 people: too thin. One extra reply moves the rate by 10 points or more.
10 to 29 people: an early signal. Worth watching, not worth rebuilding your list around.
30 or more: solid. Big enough to act on.

Look at the size of the gap too. 4 of 40 against 5 of 40 is noise. 2 of 40 against 10 of 40 is a real difference.

Should you measure reply rate or meeting rate?

Use both, in order. Reply rate counts every reply, including "not interested," so it tells you who engages. Meeting rate tells you who buys, but meetings are rarer, so you need bigger groups before it means anything.

Start with reply rate to spot which groups engage. Once you have enough meetings, check that the same groups also book them. If a group replies a lot but never books, the replies are mostly polite no's, and it isn't the group to chase.

On LinkedIn, also check the connection acceptance rate per group. A group that rarely accepts never sees your message, which is a list problem, not a copy problem.

What do you do with the results?

Shift your next list toward the winners. If directors reply at twice the rate of VPs, prospect more directors.
Stop prospecting the losers. A group with a solid sample and a near-zero rate is costing you sends.
Then test the message inside the best group. Once you know who replies, A/B test your messages on that group, so the test isn't muddied by who you sent to.
Re-check every few weeks. New prospects keep coming in, and a group that looked strong at 30 people can fade at 100.

How to automate finding which prospects reply with an AI agent

Sliq can do the tagging and the math for you. In Insights, ask a question like "Do directors reply more than VPs?" and list the tags. Sliq tags everyone you've contacted using TypeSafe's Jev, then shows the acceptance, reply, interested, and meeting rate for each tag, marked thin, early, or solid by sample size. New prospects added to your campaigns are tagged automatically, so the numbers stay current.

Sliq runs LinkedIn and email outbound from Claude, ChatGPT, or any other AI agent. Sliq finds prospects, researches them, and drafts every message, then adds each one to an approval queue for you to review. So once you know which groups reply, the agent can build the next list from them.

Delegate this to a Sliq agent ->

Best AI lead scoring tools for outbound to score prospects before you reach out
How to A/B test LinkedIn messages once you know which group to test on
Jev use cases in sales and go-to-market for other jobs a decision model can do
Find customers in your LinkedIn network to build the next list from warm connections
Browse all GTM plays

Frequently asked questions

How do you find which prospects reply to your outreach?

Sort the people you've already contacted into groups, such as seniority, industry, or company size, then compare the reply and meeting rate for each group. Only trust a difference once each group has at least 30 people. Then build your next list from the groups that reply most, and stop prospecting the ones that don't.

How many prospects do you need before trusting a reply rate?

Treat a group with fewer than 10 people as too thin to read, 10 to 29 as an early signal, and 30 or more as solid enough to act on. With 10 people, one extra reply moves the rate by 10 points, so small groups swing on luck.

Should you measure reply rate or meeting rate?

Use both. Reply rate counts every reply, including "not interested," so it shows who engages. Meeting rate shows who buys, but meetings are rarer, so you need larger groups before the numbers mean anything. Start with reply rate and confirm with meeting rate once you have enough meetings.

How do you tag prospects into groups without doing it by hand?

Use an AI classifier that reads each prospect's title, headline, and company and picks one of the tags you defined. Decision models like TypeSafe's Jev are fast and cheap enough to tag thousands of prospects. Sliq does this for you in Insights and shows the reply and meeting rate for each tag.