How many times should we try to reach a new lead before giving up?
More than once, and across more than one channel. A single unanswered call is not a dead lead; it is an unanswered call. A workable default is several attempts spread over the first few days, mixing call, text and email, with the first two attempts inside the first hour. What matters more than the exact count is that the cadence is defined in the system rather than left to whoever has time.
The default is one attempt, and nobody chose it
Ask most service businesses their follow-up policy and you will hear a number. Pull the activity log and you will usually find that a large share of unreached leads received exactly one call attempt and nothing else.
That is not a decision, it is what happens when follow-up depends on free time. The fix is structural: the attempt schedule belongs to the system, which creates the task, assigns it, and escalates when it is skipped.
Design the cadence around how people actually behave
A few patterns hold up well enough to build on.
- Front-load hard. The first attempt should be immediate. The second should be within the hour, from a different channel — usually a text, because an unknown number that just called is more likely to be answered as a message.
- Vary the time of day. Three calls at 10am on three consecutive days mostly proves the person is not available at 10am.
- Leave something actionable. A voicemail with a specific callback window and a name outperforms a generic one, and a follow-up text referencing the voicemail outperforms either alone.
- Stop cleanly. Define the last attempt and say so in it. Then move the record to a nurture track rather than leaving it open forever.
Open forever is worse than closed
Leads that stay in an active stage indefinitely quietly destroy your pipeline reporting. Everything looks like it might still close, so nothing gets a real forecast, and the team learns to ignore the list.
A defined exit — unreached after N attempts, closed with a reason code — does two things. It makes stage counts mean something, and it creates a clean population you can market to later. That population is one of the most useful segments a customer data layer can produce, because you know exactly what happened to it.
Track attempts per lead, not just contact rate
Contact rate alone cannot tell you whether the process is working. If contact rate is low and average attempts per lead is one point two, you do not have a lead quality problem, you have a follow-through problem.
The report worth building is contact rate by attempt number: what share of eventually-reached leads were reached on the first try, the second, the third. When the fourth attempt still converts meaningfully, stopping at two is leaving booked work on the table. That analysis needs call outcomes joined to lead records, which is exactly what call analysis plus CRM data makes possible.
Topics: follow-up cadence · lead follow-up · no answer · process
Have a version of this question about your own business?
The useful answer usually depends on which systems you run and how they're connected. That's a conversation, not a blog post.