What to send after the first reply to a cold message

A first reply runs sixty nine characters

Across those 139 conversations, the median first reply ran 69 characters. The mean is 118, pulled up by a handful of long ones, and 28 of the 139 came in under 25 characters.

Sixty nine characters is "Our offices." It is "Yeah". It is an email address with nothing around it. A reply that short carries almost no information about intent, so the intent gets supplied by whoever reads it, and the reading drives the next message. Sorting the replies by what the person was actually doing puts a floor under that guess. Reading all 139 by hand, they break down like this:

• 57 gave a partial answer, a correction, or a handoff to someone else
• 46 declined outright
• 18 were a bare line: a greeting, a thanks, or an emoji reaction
• 15 asked a question back
• 3 asked for a later date

One share moved a lot since our July and August window, where the same classification over 69 conversations found 10 outright declines. Declines are now 33% of first replies against 15% then. Volume per profile went up over those weeks and more markets came online, and the honest reading is that a larger share of people now answer a cold message to close it rather than to open something.

The family of the first reply carries the thread

Split continuation by what the person sent first, and the families separate harder than anything in our own copy:

• bare line: 10 of 18 continued, 56%
• asked a question back: 6 of 15, 40%
• partial answer, correction, or handoff: 18 of 57, 32%
• declined outright: 3 of 46, 6.5%

Counting only the threads where we did answer, the same order holds: 10 of 12 bare lines kept going, 6 of 12 questions, 18 of 44 partial answers, and 3 of 16 declines. A person who writes "hi, thanks for connecting" has given you almost nothing and left the door open, and that is the most promising thing a first reply can be. Four polite sentences explaining why the timing is wrong usually finish the conversation, and 3 of 16 of those reopened when we answered anyway. Those three are why the first "no" gets one honest push rather than a goodbye.

The steepest remaining drop sits right after the reply

The depth of those 139 conversations, counted by how many messages the other person sent:

• 139 sent a first reply
• 39 sent a second, 28%
• 19 sent a third, 14%
• 16 sent a fourth, 12%

Between the first reply and the second, 72% of conversations end. That is a larger proportional loss than anything downstream, and it lands on people who have already accepted a connection request and already typed something to a stranger.

The first controllable variable is whether anyone answers. We answered the first reply in 87 of the 139 conversations. In the 52 we left alone, two people wrote again unprompted, and 30 of the 52 were outright declines where silence is a reasonable answer. Three of them were questions aimed straight at us: two were counter pitches we chose to leave, and one is a drafted answer that has been sitting open past its clock for over a month. Where we did answer, 36 of 87 conversations continued, 41%. The machinery that keeps every reply answered is covered in how we run reply handling as a system.

What our own answer changes

In the July and August window, answers containing a question continued 55.6% of the time against 42.9% for answers without one, and we published it as a direction rather than a finding. On a sample twice the size the two groups sit a point and a half apart: 25 of 61 answers with a question continued, 41.0%, against 11 of 26 without, 42.3%. Across a whole thread the count runs the other way, 57.6% of our messages in conversations that continued carried a question against 71.4% in conversations that stopped, so a question mark earns its place by what it asks rather than by being there.

The shape we still write is the same three moves every time, because it decides what the person reads even when the aggregate stays flat. Answer the literal thing they said, in one line. Concede anything they corrected, without defending it. Then ask one question about their situation that they can answer from memory.

Worked examples from conversations that continued, with the identifying parts removed:

• They wrote "Our offices." The answer described the specific case we get called on, one hire starting in a city where the company has no room yet, and asked whether that had come up.
• They wrote that our checkable fact about their employer was out of date. The answer was "Fair hit, that is clearly out of date, my bad", then one question about where they had landed.
• They handed over a colleague's email. The answer confirmed the exact address back to them and asked one qualifying question before using it.

Speed is worth naming because it moved while continuation stayed put. Our median time from their reply to our answer is now 5.0 hours, the 75th percentile 24.1 hours and the 90th 57.5. Twenty nine of the 87 answers went out inside an hour, 74 inside 48 hours, 13 later. The reason is the clock we put on ourselves: a reply is drafted immediately, the owner of the account can send it, edit it or hold it, and at the end of 48 hours it goes on its own. In the earlier window our median was 14 minutes, and continuation followed latency in neither window. So answer quickly for the ordinary reason, that people forget what they wrote to a stranger, and keep a deadline so nothing sits forever. The operator layer around that clock is laid out in what a managed outbound engine does.

Count the characters each side wrote

Add up the characters each side wrote across a whole conversation and the two groups separate hard.

• conversations that stopped at one reply: 569 characters from us, 129 from them, a ratio of 4.4 to 1
• conversations that continued: 886 from us, 534 from them, 1.7 to 1

At the single turn right after the first reply the two groups sit close together: our answer averaged 212 characters against their 114 in conversations that went quiet, and 207 against 96 in the ones that continued. So the imbalance is cumulative. It builds one message at a time, and by the time it shows up in the totals the conversation is already over. The worst thread in this window ran 2,503 characters from us against 18 from them across six of our messages, five of which carried a question.

The rule that follows is easy to hold. Match theirs. A person who writes one line gets one line. A person who writes three paragraphs has invited three paragraphs. Your own explanation of what you do is worth roughly one sentence until they ask for the second one.

The second message, now with a number

Our answer was followed by silence in 56 of the 87 conversations we answered. In 15 of those 56 we sent a second message, and 5 of the 15 brought a reply. In the other 41 our answer sat there and the conversation ended with our name at the bottom of it.

One reply for every three second messages is the first real evidence we have on this turn. In July and August the same count was five second messages and zero replies, too thin to claim anything. Fifteen is still thin, and it points somewhere we can act: the cheapest pipeline in a quiet thread is the message we already decided against sending.

What that message carries is the part worth getting right. The person read the ask once already, so a repeat of it adds pressure and the same information. The second message earns its place by adding something absent from the first: a number from a comparable account, a specific artifact you can send today, a short list of companies you would go after for them. Our own copy rules hold this as a hard requirement, because a follow up that restates the previous message is the machine tell that ends threads.

One related rule, learned the expensive way. A booking link goes into a thread once. Ask for the conversation in words as often as the situation warrants, put the link down a single time, and repeat it only after the person themselves returns to the subject of timing. They saw it the first time.

Deferrals need a date said out loud

Three of the 139 first replies asked for a later date, down from six of 69, so the practice here rests on the logic more than on the count. A polite acknowledgement is a soft close that both sides forget inside a week. The version that holds is one line longer: name the date out loud, put it in a calendar where it will fire without you, and say what you will bring when you come back. In our logs the answers that named a specific month and a specific thing to return with are the ones still moving.

What to run on Monday

Five checks, all on data you already have.

1. Count how many of your replied conversations produced a second reply from the other person. Ours is 39 of 139. A share under 30% points the copy work at the message after the reply.
2. Sort last month's first replies into the five families above. The share that declined outright is the number to watch, because it moves with how broadly you are targeting, and it moved from 15% to 33% for us in two months.
3. List every first reply and mark whether anyone answered it at all. Ours was 52 unanswered out of 139, and three of those were direct questions, which is the kind of miss that only a query finds.
4. Add up the characters each side has written in your live threads. Anything past three to one is a signal to cut before you send the next one.
5. Pull every conversation that has been quiet for more than four days since your answer and send a second message that adds a fact. Ours returned one reply for every three sent.

All five run on the same habit: treating a reply as the start of a conversation with its own numbers, and measuring those numbers on the same schedule you measure sends. If you would rather have this operated for you, with the reply handling, the follow up rules and the counters already built, that is our work. The full cycle is described on how the system works, and the levels are laid out on our levels and pricing page.

Questions buyers ask