A sales cadence is the planned rhythm a sales team follows to reach a prospect: how many touches, on which channels (email, phone, LinkedIn), how far apart, over how many weeks, and what ends it. An email sequence covers one channel. The cadence is the whole schedule, calls and social touches included.

Without one, follow-up runs on mood. A rep sends a good first email, maybe a second, hears nothing and moves on to fresh names, reading silence as a no when it usually means the prospect had a busy week.

Teams that do have a cadence often have exactly one, built for cold outbound, and push demo requests, no-shows and stalled deals through it too. Those run on different clocks, so much of this entry is about running more than one.

TL;DR

• Five choices set a cadence: how many touches, over how many weeks, on which channels, with what gaps, and what stops it. Day numbers follow your sending windows, so in Outreach a step due on a Sunday waits for Monday and the later steps shift with it.

• Reps give up early because silence feels like an answer. Fixing the stopping point before the first send takes that call away from a bad afternoon.

• Most teams need a small set of cadences, not one: cold, signal-triggered, inbound, content leads, no-shows, stalled deals and recycling, each with its own entry rule and pace.

• Inbound follow-up cadences are measured in minutes and days, not weeks. HubSpot lets a contact sit in only one sequence at a time, so a cold prospect who fills in a demo form has to leave the cold cadence before the inbound one can start.

• The cadence sets the daily workload. Fifteen new prospects a day in a cadence with five emails and two calls means about 75 emails and 30 calls a day once it's full, and at 10 to 20 cold emails per inbox (follow-ups included) that takes four to eight inboxes.

• Park non-responders for 60 to 90 days, bring them back only with a new reason to write, and make an opt-out on one channel stop every channel.

• Nebor designs cadences around a team's channels and rep capacity, builds them in the sequencer and CRM the client already runs, and writes down the entry rules so the next change doesn't need us.

What sets a sales cadence's rhythm

Whatever tool runs it, every cadence comes down to five settings.

The count is the number of touches, and the length is how many weeks they're spread across. Together they give you density, which is what the prospect actually feels. Nine touches over three weeks is a steady presence. Nine touches in one week is a siege.

The channel mix decides which touches carry the argument and which carry familiarity. Email holds the detail, a connection request puts a face to the name, and a call is the one touch where the prospect can push back in real time. What each channel does on its own is covered under multi-channel outreach. For cadence design, the point is that the mix decides the order: a call lands better once the person has seen your name twice.

The gaps give the cadence its shape. Nearly every one that works is front-loaded, with touches close together in the first week while your first message is still fresh, wider gaps after that, and a longer pause before the last one. The opposite shape, a touch every four days for six weeks, feels like a machine emptying a queue.

The exits are the fifth setting and the one most often left on the default. A reply on any channel, a booked meeting, a bounce or an opt-out should each end the automated part for that person.

One practical wrinkle: "day 3" is rarely a calendar date. Sending tools only run steps inside the time windows you allow. Outreach's sequence schedule documentation shows the effect with a step that lands on a Sunday: it waits for the first time block on Monday, and the next step is then counted from Monday. A cadence started on a Thursday can lose its tight opening cluster to the weekend, which is why we start new cohorts early in the week.

Why reps stop after two touches

Ask a rep why they dropped a prospect after two emails and the answer is nearly always "they weren't interested". Usually nobody said so. The prospect was travelling, meant to reply later, or never saw the email among ninety others on a Monday morning.

Silence feels like a verdict, and it gets heavier with every touch. By the third unanswered message most people feel they're bothering someone, and that feeling, not any evidence, is what ends the follow-up. When we go back through client cadences and check which touch produced each reply, the first email is rarely where most of them came from, and a fair number arrive on a different channel from the one that opened.

A cadence fixes this by moving the decision. The stopping point is agreed before the first send, by someone looking at the whole list rather than at one silent prospect. The rep's job becomes running whatever step is due, which is far easier to keep doing on a slow day than deciding again, prospect by prospect, whether to carry on.

Structure doesn't make a weak touch worth sending, though. A third email that says "just bumping this" teaches the prospect that your name means nothing new. Each step needs its own reason to exist, whether that's a fresh angle, a customer story from their industry or a question nobody asked yet. Writing those is the craft of a good follow-up sequence, and doing it for hundreds of prospects without sounding templated is what personalization at scale is for.

One team, several cadences

A cadence built for strangers is the wrong tool for someone who asked for a demo an hour ago, and wrong again for a buyer who went silent after a proposal. Each starts at a different temperature, so each deserves its own rhythm.

This is the set we'd usually sketch for a B2B team selling mid-market deals. The numbers are working ranges, not benchmarks.

Two things make a set like this hold together. Each cadence needs an entry rule that lives in the CRM rather than in a rep's head, so the same person isn't picked up by two cadences at once. And the cadences need a priority order. If someone in the cold cadence downloads a guide, the warmer cadence wins and the cold one stops.

The outbound playbook is where those entry rules and priorities belong in writing. Starting cadences from real buying signals and trigger events, instead of from whoever was in last week's list upload, is also how a team moves from batch outbound to outreach that's timed to the buyer.

Inbound follow-up cadences run on a different clock

An inbound lead turns the cold logic around. You don't have to earn attention, because the person just gave it to you. You have to respond before that attention goes somewhere else, which makes the first hour worth more than the following two weeks. That first response is the subject of speed to lead. The inbound cadence is everything around and after it.

In practice it's short and dense. A call and a personal email within minutes, a second call the same day if the first went to voicemail, an email the next morning that answers whatever the form asked and offers a booking link, then a couple of lighter touches over the following days. If there's still nothing after a week, the lead goes to nurture, not into a cold cadence, because a cold opener would read oddly to someone who already knows you.

The no-show cadence is the inbound variant teams forget. Someone booked a meeting and didn't turn up. More often than not that's a calendar collision, not a change of heart. An email within the hour offering two new times, a call that afternoon and one more nudge a couple of days later is often enough to rebook, and none of it happens reliably if it depends on the rep remembering.

The tooling has a catch. In HubSpot, each contact can only be enrolled in one sequence at a time, and enrolling contacts through a workflow (the way an inbound cadence starts without a rep pressing a button) requires a Sales Hub or Service Hub Enterprise subscription. So when a prospect halfway through a cold cadence fills in a demo form, something has to take them out of the cold sequence first, or the inbound one never starts. We build that handover as an explicit rule and test it with a real form fill.

The inbound meetings workflow shows how routing, the first touch and this follow-up cadence fit together for demo requests.

The capacity math behind a sales cadence

Every cadence comes with a workload attached, and it's easy to design one no rep could actually run. The arithmetic fits on the back of an envelope.

Take a cold cadence shaped like the one above: five emails, two calls and three LinkedIn actions (a profile visit, a connection request and a message) over about two and a half weeks. Say each rep starts 15 new prospects every working day. Within three weeks around 200 people are in flight at once, and from then on every day carries the same load: about 75 emails, 30 calls and 45 LinkedIn actions, a little less as replies take people out early.

Hold that against what each channel can carry. In our experience a cold inbox stays healthy at 10 to 20 emails a day, follow-ups included, so 75 emails a day needs four to eight warmed inboxes. Email throttling and inbox rotation cover how that sending gets paced and spread. Thirty calls fills a solid block of a rep's day.

LinkedIn is usually where the plan breaks first. The steps run from a real person's profile, and LinkedIn's help page names a burst of invitations, or a pile of ignored ones, among the reasons it restricts an account. Forty-five actions a day from one profile is more than we'd ask of it, which is why the LinkedIn outreach steps tend to set the ceiling.

So work backwards. Start from the channel with the least room, divide by how often the cadence uses it, and you have the number of new prospects a rep can start each day. If a rep can sensibly manage 20 LinkedIn actions a day and the cadence uses three per prospect, that's six or seven new prospects a day, not fifteen, and the email and call volumes follow from there. Skip this and the overdue tasks pile up by week two, which is how a cadence slides back into improvised follow-up.

Automation changes the numbers but not the logic. Emails can go out from a sending tool, which is part of how you automate sales prospecting, while calls and replies stay with people. The question becomes how many inboxes and rep hours the cadence needs, a budget conversation better had before launch.

After the final touch

Every cadence ends, and each way out needs its own rule.

A reply takes the prospect out of automation and puts them with a person, with the conversation logged on the record so nobody else picks them up by accident. A soft no ("not this quarter") deserves a date rather than a deletion. Set a task or a property for the month they mentioned and come back then with something new.

Silence after the last touch means park, not delete. We'd leave a non-responder alone for 60 to 90 days and bring them back only with a fresh reason to write, such as a new role, a funding round or a run of pricing-page visits. The same cadence with the same emails a second time turns an indifferent prospect into an irritated one.

An opt-out has to travel across channels. Someone who replies "please stop" to an email shouldn't get a call the next morning because the phone step lived in another tool. Calling rules differ by country. In the UK, the ICO's telephone marketing guidance tells business-to-business callers to screen numbers against both the TPS and the corporate CTPS register, and to keep their own list of people who've asked not to be called. A cadence with phone steps should check those lists before the call task is created, not after the call.

Finished cadences are also the best input for the next version: which step drew the replies, which channel opened the conversations, which segment never answered.

In short

A sales cadence is a decision made in advance about how often to show up, where, for how long and when to stop. Made in advance, it survives a hectic week and a discouraged rep. Made on the fly, it tends to end after the second touch.

What most teams miss is that one schedule can't serve every situation. A stranger, a demo request, a no-show and a stalled deal each need their own rhythm, a rule for which one wins, and enough inboxes and rep hours to run. With that settled, the emails and calls themselves finally have room to be good, and the cadences become the clock the rest of your GTM system keeps time by.

If you'd like a second pair of eyes on yours, we map the cadences a team actually needs, build them in the sequencer and CRM it already runs, and leave the entry rules and capacity numbers written down for whoever adjusts them next.

Frequently asked questions