Loading
Follow up on a cadence you set, across channels, and stop permanently the moment they reply or ask it to. Silence is not consent to keep going forever.
Published December 2, 2025
Most leads do not say no. They stop replying, which feels like a no but is not one, and the honest reason is usually that something more urgent happened. The question is what a reasonable follow-up looks like when you cannot tell the difference between busy and uninterested.
Most human teams stop after two attempts, not because two is the right number but because the third attempt is nobody's favorite task and there is always something else to do. The lead who did not answer on Tuesday is still a lead on Thursday, and often still a lead the following week.
An agent does not find the third attempt awkward, which means the decision about how many times to follow up becomes a policy question rather than a motivation question. That is a genuine improvement, provided somebody actually sets the policy.
The same thing that makes an agent good at persistence makes it dangerous without limits. A system that will happily send a fifteenth message is a system that will annoy your buyers and damage your name, and it will do it at scale and without noticing.
So the stop rules matter more than the cadence. Ours are deliberately blunt.
A follow-up that arrives on the channel the buyer chose reads as service. The same message on a channel they did not choose reads as intrusion. Somebody who filled in a web form and gave an email address has not invited a phone call, and treating those as equivalent is the fastest way to make follow-up feel like pursuit.
The practical rule is to follow up first on the channel the inquiry arrived on, and only widen if they have explicitly given you the other one. That single constraint removes most of the complaints that automated follow-up generates.
The third message in a sequence should not read like the first with the word 'still' added. It should say something new, or acknowledge that it is a follow-up, or offer an easy way out. The cheapest improvement available to most sequences is simply making the last message an explicit off-ramp: here is how to tell me to stop, no hard feelings.
Counterintuitively, giving people an easy exit makes the sequence perform better, because it removes the feeling of being cornered. A buyer who was genuinely busy is more likely to re-engage with a business that did not make them feel pursued.
Every attempt, on what channel, at what time, and what came back. Not for the reporting, though that is useful, but because the audit trail is what lets you answer honestly if somebody ever asks how many times you contacted them. If you cannot answer that question from the record, the cadence is running without supervision.
On the site
Keep reading