Not on the night. On the night everyone means it. The message dies in the gap between meaning it and having anything specific left to say.
Most follow-ups are not late because of laziness. They are late because the specifics have gone. On the night you could have written something worth reading; by Thursday you have had nine more conversations and all that survives is a name and a job title, which produces a message indistinguishable from a mailshot. The fix is to capture the specifics at the moment, not to try harder later.
By Cardr · Published 4 August 2026 · 5 min read
It is not that people are rude, and it is not that you are disorganised. It is that the useful parts go first.
“Great to meet you at the event, let me know if I can ever help” contains no evidence that the conversation happened. It could have been sent to everyone in the room, and the recipient assumes it was, because it usually was.
A message that references one specific thing from the conversation does something different. It proves you were listening, which is the actual scarce quality, and it gives the person something to reply to.
The trouble is the order things leave in. What you keep longest is what is least useful.
The only row that survives the month is the one you could have got off a badge. Everything that would have made the message worth reading has gone by Thursday, which is roughly when most people sit down to write it.
You do not need a better memory. You need to write one sentence down while you are still standing there.The specific thing does not have to be profound. The size of the problem they described, the city they are moving to, the thing they said they were dreading. All of it is evidence.
It works with a notebook. Nothing here needs software.
Not at the end of the night, when the conversations have merged. While you are still standing there, or within a minute of leaving. The problem they described is the whole message you will eventually send.
"After the funding closes", "in the new year", "once they hire someone". Timing tells you when to write, which is usually worth more than what to write.
Three or four sentences, short enough to read on a phone in a lift. The specific detail is the whole proof that you were listening, and it is what makes the message answerable.
A message with no ask is often the one that gets a reply, because it does not read as the opening of a sales sequence. If you do ask, ask for something that costs them a sentence.
This is the step almost nobody takes and where most of the value is. In January, the person who said "come back after Christmas" will have forgotten they said it to anyone. Being the one who came back is often the entire advantage.
Almost nobody sends it, and it is where most of the value is.
Send the first message within 48 hours, while they can still place you. After about a week you are reintroducing yourself, which is a much harder message to write and a much easier one to ignore.
Keep it short enough to read on a phone in a lift. Reference the specific thing. Make the ask small or make no ask at all; a message with no ask is often the one that gets answered, because it does not feel like the opening of a sales sequence.
Then, if the timing detail said so, come back at the actual moment. The person who said “come back to me after Christmas” meant it, and in January they will have forgotten they said it to anyone. Being the one who came back is often the entire competitive advantage, and it costs one line in a calendar.
Being straight about this, since we sell one.
No tool makes you interesting, and no tool can invent a specific detail you did not capture. If you take nothing else from this page, take step one; it works with a notebook and a pen.
What a tool does is remove the friction from the two steps that fail most: getting their details at all, and having the context still attached when the moment comes round. Cardr does the first by making the exchange two-way, so a conversation ends with a contact rather than a hope, and the second by holding your note against the name and drafting the message from it later.
It does not make the follow-up good. It makes the good version cheap enough that you actually send it, which turns out to be the binding constraint for almost everybody.
No. And the places paper still wins are specific enough to name.
GuideNFC vs QR for business cardsThey fail in different places, and the distance between you decides which one matters.
GuideDigital business cards on AndroidThe most common objection, answered straight rather than sold around.
By professionWho it is forSixteen trades, each with the moment that decides whether a card works.
Short enough to read on a phone without scrolling. Three or four sentences is plenty. Length reads as effort to the writer and as work to the reader.
Within 48 hours, while they can still place you. After about a week you are reintroducing yourself, which is a harder message to write and an easier one to ignore.
Then the conversation was probably not one worth following up, and sending a generic message will not change that. Spend the time on the three conversations that were.
Once, later, with a genuine reason. Not a bump, not a "just floating this to the top of your inbox". If the timing detail you captured has arrived, that is a real reason and a legitimate second message.
It drafts one from the note you captured, on the Pro plan, fifty a month. You edit and send it. The hard part of a follow-up is starting it from an empty box, and that is the part it removes.
Nothing for the other person to install, and no card details to start.