TL;DR: An outbound follow-up cadence is a pre-planned sequence of touches - calls, texts, and emails, spaced over days - that you run against every new lead until they convert or clearly opt out. Most deals are lost not to a bad pitch but to a cadence that quits after two touches or forgets which leads it already worked. Build a cadence with a fast first touch (minutes, not days), 8 to 12 total touches across at least two channels over two to three weeks, and a single system tracking state so nobody falls through the cracks.
What is an outbound follow-up cadence?
An outbound follow-up cadence is the fixed rhythm of contact attempts a rep or system makes after a lead enters the pipeline: touch one at minute zero, touch two a few hours later, and so on across voice, SMS, and email until the lead responds or exits.
The cadence is not the script. The script is what you say. The cadence is when you say it, on which channel, and how many times before you stop. Get the cadence wrong and even a perfect script never gets heard.
Here is the uncomfortable rule of thumb the whole discipline rests on: most leads that eventually convert do so after multiple follow-ups, yet most outbound efforts stop after one or two attempts. The gap between those two facts is where revenue leaks. A cadence exists to close that gap on purpose instead of leaving it to whoever remembers to call back.
Why do leads get dropped in the first place?
Leads rarely die because they said no. They die from silence, and silence has predictable causes.
- The cadence quits too early. A rep leaves one voicemail, sends one email, hears nothing, and moves on. The lead was never a no - it was a not-yet.
- No memory of state. Nobody knows lead #418 got a text on Tuesday and is due for a call Thursday, so the touch never happens or happens twice.
- Single-channel tunnel vision. You call three times, get no answer, and give up - never trying the text the lead would have replied to in seconds.
- Slow first touch. The first attempt lands hours or days late, after the lead's intent has already cooled or a competitor has answered.
Every one of these is an operations failure, not a talent failure. A good cadence, tracked in one place, removes all four. That is why teams increasingly run the whole motion through one system where every touch is logged automatically and the self-driving pipeline advances itself, rather than relying on humans to remember who is owed what.
The anatomy of a cadence that doesn't drop leads
A durable cadence has five design decisions. Make them once, then let the system execute.
1. Speed to first touch
The first attempt should happen in minutes, not the next business day. Lead response time is widely cited as one of the strongest predictors of contact and conversion - the odds of reaching a lead fall off sharply as minutes turn into hours. Your cadence's touch one is the single highest-leverage moment in the whole sequence.
If a lead comes in at 9 p.m. or during your team's busiest hour, a human queue guarantees a slow touch one. An AI voice agent or an instant text closes that gap because it doesn't sleep, break, or triage.
2. Channel spread
Single-channel cadences leave replies on the table. Some people answer the phone. Some will only text. Some read email at 6 a.m. and never pick up a call. A cadence that touches at least two channels catches more of your audience than one that hammers a single wire.
3. Touch count and spacing
Aim for 8 to 12 touches over two to three weeks. Front-load the early days when intent is hot, then taper. A workable default:
- Day 0: Call within minutes; if no answer, send an immediate text and a short email.
- Day 0-1: Second call a few hours later, different time of day.
- Day 2: Text with a specific question that invites a reply.
- Day 4: Call plus a value email (case-relevant, not "just checking in").
- Day 7: Text.
- Day 10: Call.
- Day 14: "Break-up" email or text that gives an easy yes/no.
4. Exit and opt-out rules
A cadence needs clean stopping conditions: a booked meeting, an explicit no, or an opt-out. Persistence is not harassment only if you honor exits instantly. That means DNC scrubbing, per-state opt-out handling, and TCPA timing rules baked in - not bolted on. If you want the full rulebook, see the outreach compliance guide.
5. State tracking
Every touch, on every channel, has to be logged against the lead the instant it happens. Without shared state, a cadence is a wish. With it, the cadence runs itself and no lead sits in limbo.
When to call vs. text vs. email inside the cadence
Each channel does a different job. Sequence them to their strengths instead of blasting all three at once.
| Channel | Best for | Speed | Use in cadence |
|---|---|---|---|
| Voice | Qualifying, building rapport, closing | Real-time | Touch one and every decision moment |
| SMS | Quick replies, confirmations, nudges | Seconds | Between calls, reminders, re-engagement |
| Context, proof, documents, long-form | Hours | Value drops and the break-up touch |
The pattern that works: lead with a call for the human moment, use text to keep the thread warm between calls, and use email to carry the context a text can't. For a deeper split on this, the call vs. text vs. email playbook breaks it down by scenario.
How to run the whole cadence from one system
Here is where cadences fall apart in practice. A dialer runs the calls. A separate SMS tool runs the texts. An email platform runs the sequences. A CRM is supposed to tie it together, but a rep has to log everything by hand. The cadence lives in four tools that don't talk, so touches get missed, double-sent, or mistimed.
The fix is to run voice, SMS, and email against one audience, on one schedule, with one record of truth. Tools like DialEcho run the entire cadence from a single system: the AI voice agent places the calls, two-way texting handles the replies in seconds, email sequences carry the context, and every touch logs itself to the pipeline with no data entry. Because it all draws from one token wallet, a call minute and a text and an email batch spend from the same balance - so channel choice is a strategy decision, not a budgeting headache.
Running the motion in one place is also what makes speed-to-lead real. When touch one can fire the moment a lead lands - after hours, during a rush, on the hundredth lead of the day - your cadence starts on time every time. See how the end-to-end motion fits together for the full picture, or the AI calling agent if the voice side is your bottleneck.
Where a human still beats the automation
Be honest about the line. Automation wins on persistence, speed, and never forgetting a touch - the exact things humans are bad at. It should own touch one, the reminders, the re-engagement, and the qualifying.
A closer wins on the moment a genuinely interested buyer is ready to talk numbers, handle a real objection, or feel understood. The smartest cadences let the system work the long, patient sequence and hot-transfer the buyer to a human at the exact moment it matters. That is the point of automating the cadence: not to remove the human, but to make sure the human only spends time on live, qualified conversations instead of dialing voicemails.
A quick cadence checklist
Copy this and pressure-test your current sequence against it:
- Does touch one fire in minutes, 24/7?
- Are at least two channels in the sequence?
- Are there 8 to 12 touches over two to three weeks?
- Is spacing front-loaded early, tapered later?
- Is every touch logged automatically against the lead?
- Are opt-outs, DNC, and timing rules enforced on every channel?
- Is there a clean break-up touch that ends the sequence?
- Does a ready buyer reach a human fast?
Miss any of these and you have a leak. Close all eight and your cadence stops dropping leads - it works them until they close or clearly say no.