60-Min Sales Training: Multichannel Prospecting Cadences

Two ways to run the hour: framework-first or script-first

Every sales manager who sits down to build this session lands on the same fork. You can teach the *architecture* of a multichannel cadence — why day 2 is a phone call and not another email, what a trigger tier is, how channel mix distributes — and let reps derive their own copy. Or you can hand out finished scripts, drill them until they sound natural, and let the architecture emerge from repetition. Both work. They fail differently, and that difference is what decides which one belongs in your hour.

Framework-first front-loads the whiteboard. You draw the 14 days, you argue for the sequencing, you explain why the second voice on a different channel outperforms the third email in the same thread. Reps leave understanding the *why*. The failure mode is beautiful theory and zero output: everyone nods, nobody sends. Managers who run framework-first sessions and skip the build assignment routinely find, two weeks later, that the sequence was never created in Outreach or Salesloft at all. The cadence lived on a whiteboard, got photographed, and died in a camera roll.

Script-first inverts it. You open with a printed one-pager, reps highlight the version they'll use, and by minute 20 someone is reading a voicemail out loud badly and then less badly. Reps leave with muscle memory. The failure mode is brittleness: the moment a prospect goes off-script — "I already have an SDR tool," "bad timing, circle back in Q4" — a rep who memorized words without the underlying logic freezes or, worse, keeps reading. Script-first teams also drift toward copy-paste across channels, because nobody explained that email carries the proof and LinkedIn carries the relationship.

There's a third option worth naming because it's what most experienced managers actually do: framework-then-script, with the framework compressed to the minimum viable explanation. Fifteen minutes of architecture is enough to make the scripts make sense, and it leaves 15 for the scripts themselves and 15 for role-play. That's the split this page recommends — 5 setup, 15 framework, 15 scripts, 15 role-play, 5 pitfalls, 5 action items — but it's a recommendation, not a law, and the tradeoff is real. A team of tenured AEs who already understand cadence theory can cut framework to 5 minutes and spend 25 on role-play. A brand-new SDR pod that has never run a sequence needs the full 15, possibly 20, and should sacrifice pitfalls to get it.

The adjacent question — one that comes up in nearly every team's second or third session — is whether to train the cadence at all versus just shipping a manager-built sequence and telling reps to run it. Shipping the sequence is faster. It also produces reps who send touches they don't believe in, and the tell is voicemail quality: a rep running someone else's cadence leaves a 40-second voicemail that recites the email. A rep who was in the room when the play was designed leaves 20 seconds that reference the email without repeating it. That gap shows up in reply rate on the day-5 email, which is the single most sensitive downstream metric in the whole sequence.

Choosing your hour: a decision path

The choice isn't really "framework or scripts." It's a function of three inputs: rep tenure, whether a sequence already exists in your engagement platform, and how much of your pipeline sits in enterprise versus mid-market. Enterprise cadences run longer and lean harder on the phone; mid-market cadences compress into two to four weeks and lean on email plus LinkedIn. If you're training one hour and your pipeline is mixed, teach the mid-market version — it fits the majority of most B2B pipelines and enterprise reps can stretch it, whereas the reverse (teaching a 50-day enterprise cadence to a mid-market pod) produces reps who under-touch and give up on day 9.

Two branches in that diagram deserve defending. First, the live-build branch: if no sequence exists in your platform, spending 20 of your 60 minutes watching reps actually construct the steps in Outreach, Salesloft, Apollo, or HubSpot Sequences is worth more than role-play. Configuration is where cadences die. Reps set the day-3 LinkedIn step as an automated task with no due-date enforcement, it silently skips, and the thread rule breaks — the day-5 email references a voicemail the prospect got but a LinkedIn touch they never saw. Watching the build catches that.

Second, the manager-approval gate. It sounds bureaucratic for a 10-step sequence. It exists because the most common post-training regression is a rep who rebuilds the cadence "their way" — usually by deleting the phone steps, because dials are uncomfortable and connect rates are low. If you don't inspect the sequence before it activates, you'll discover in week three that half the team is running a two-channel email-and-LinkedIn play and calling it multichannel. Approval takes four minutes per rep and prevents the single most expensive drift.

One more decision that lives just outside the hour: who gets cadenced at all. A trigger taxonomy — new exec hire, funding, acquisition, public earnings miss as the top tier; job posts, tech-stack signals, podcast appearances as the second; ICP-fit-only as the third — determines whether the cadence is worth running. Full multichannel effort on a cold ICP-only list burns the list and the rep. The standard rule is that the bottom tier earns email and LinkedIn only, and unlocks phone and video after any engagement signal. Teach this in setup, not as a separate session; it takes 90 seconds and it prevents the most common list-burn pattern.

The numbers each option has to clear

Whichever structure you choose, the arithmetic underneath the cadence is the same, and reps should leave the room knowing it. Cadence design is a math problem dressed as a communication problem.

Touch count. Industry SDR benchmarking has trended upward for a decade — the working number for enterprise has moved from roughly a dozen attempts to the high teens or low twenties across a six-to-eight-week window, while mid-market and SMB compress to a lower double-digit count over two to four weeks. The direction of travel matters more than any single vendor's figure: attempts per contact go up, per-touch response goes down, and single-channel sequences absorb that decline worst. Treat published benchmarks as directional. Your own sequence stats after 200 contacts are more reliable than anyone's report.

Reply rate. Cold outbound email reply rates in most B2B categories sit in the low-to-mid single digits and have been declining year over year as inbox volume rises. That decline accelerated with the spread of AI-drafted outbound — buyers now pattern-match generic personalization instantly. A well-run multichannel cadence should beat your single-channel baseline meaningfully, but be careful how you set the target. "+30% relative lift" on a 5% baseline means 6.5%, not 35%. Reps hear the percentage and anchor wrong. Write the absolute target on the board: if your email-only baseline is 5%, the multichannel target is 6.5–7%, and anything above 8% is excellent.

Connect rate. Cold dial connect rates land in the low-to-mid single digits for most B2B segments. This is the number that causes reps to abandon the phone, and it's the number you have to reframe in the session. At a 5% connect rate, 95% of dials produce a voicemail — which means the voicemail *is* the deliverable, not a consolation prize. Score voicemails in one-on-ones. A rep who leaves a 20-second voicemail that plants one specific question sees a materially higher reply rate on the follow-up email that references it, because the email is no longer cold; it's the second half of a conversation the prospect already half-heard.

Video. Video steps are expensive in rep minutes — recording, re-recording, hosting, embedding — and cheap in prospect attention if the first five seconds land. The economics only work if you cap effort: a 45-second take, one recording, one CTA, no editing. Teams that let reps spend 15 minutes per video produce beautiful videos at a volume too low to matter. Cap it at three minutes of rep time per video including send, and put video only on tier-A and tier-B accounts where the account value justifies it.

Channel mix. A workable distribution across the 14 days is roughly a third phone, a third email, a quarter LinkedIn, and the remainder video. What matters isn't the exact split — it's that no single channel exceeds about 40% of touches. The moment email crosses 50%, the cadence is an email sequence with decorations, and its performance regresses to the email baseline.

Time cost, honestly stated. Ten touches across 14 days runs roughly 25–40 minutes of rep time per contact including research, if the scripts are pre-written and the sequence is templated. At 25 new contacts per rep per week, that's 10–17 hours — more than half a workweek. Managers who load reps with 60 contacts a week and a 10-touch cadence are prescribing an impossible schedule, and what actually happens is reps silently skip the phone and video steps. Size the list to the cadence, not the other way around.

Deliverability. One under-discussed number: sending domain reputation. A rep running a 10-touch cadence on 25 contacts a week is sending roughly 75 emails weekly from their personal domain — fine. The same rep at 200 contacts a week is at 600, which without a warmed secondary domain and SPF/DKIM/DMARC configured properly will start landing in spam. Multichannel cadences are partly a deliverability hedge: when email degrades, phone and LinkedIn carry the thread. That's a real strategic reason for the mix, not just a response-rate argument, and it's worth 60 seconds in the framework teach.

What each channel is actually for

The single biggest execution failure in multichannel Prospecting is treating the channels as four delivery mechanisms for one message. Reps write a good email, then paste it into LinkedIn, then read it as a voicemail. Buyers notice immediately, and the effect is worse than sending nothing on the extra channels — it signals automation and gets the whole thread muted.

Assign each channel a distinct job and enforce it in the scripts:

Email carries the proof. It's the only channel where a prospect can read at their own pace, forward to a colleague, and check a link. Put the specific claim here — the named outcome, the comparable company, the metric. Keep it under 75 words for the first touch. Longer emails aren't punished by algorithms; they're punished by attention.

Phone carries the ask. A voicemail that tries to prove something is a bad voicemail. Under 22 seconds, name the prior touch, plant one question, state a callback number and a specific next attempt time, hang up. The goal isn't a callback — callbacks from cold voicemail are rare. The goal is to make the next email land as a follow-up rather than a first contact.

LinkedIn carries the relationship. A connection request with no pitch, sent after the prospect has already seen an email and heard a voicemail, converts far better than a cold request. The message that follows an accepted connection should be value-only: a relevant post, a question, an observation about something they wrote. Pitching in the connection note is the fastest way to get ignored on the channel where you have the fewest touches available.

Video carries the human. The whole argument for video is that it proves a person, not a system, is behind the sequence. That means the handwritten name on paper in the first eight seconds isn't a gimmick — it's the proof of manual effort. It means no script-reading, because visible reading destroys the effect. One CTA, 45 seconds, done.

The thread rule ties all four together: every touch explicitly references the prior one. "I left you a voicemail Tuesday — figured I'd try a different lane." A prospect who reads the email, hears the voicemail, and sees the LinkedIn note should recognize one human running one play, not four systems firing at them. This is the highest-leverage single sentence in the entire Training, and it's the one reps forget first. Put it on the whiteboard and leave it up.

Adjacent channels worth a mention if your team has them: SMS works in some segments and is legally fraught in others — get it cleared before you teach it. Direct mail and gifting platforms function as a fifth channel on high-value enterprise accounts, usually inserted around day 8 in place of video. Community and Slack-group presence is a slow channel that doesn't fit a 14-day cadence but strongly improves the acceptance rate on the LinkedIn steps. Mention them, don't teach them; one hour won't hold five channels.

Running the hour, minute by minute, and the week after

The sequencing of the session itself matters as much as the sequencing of the cadence. Put the agenda on the board before reps sit down — Setup, Framework, Scripts, Role-Play, Pitfalls, Action Items — so nobody spends the hour wondering when it ends.

Setup, five minutes. Open with one question around the room: what was the last cold touch that actually got you a meeting, and what channel was it? Write the answers on the whiteboard. You will reference them during the framework teach, and it converts the session from a lecture into an argument reps have already joined. Then state the mindset shift in one line: you are not sending emails, you are running a 14-day play.

Framework, fifteen minutes. Draw the 14 days. Walk the touch order, explain the tier system, explain the thread rule, explain the channel-mix cap. Take the objection you will definitely get — "nobody answers the phone" — head-on with the voicemail-is-the-product reframe rather than deferring it to the pitfalls section.

Scripts, fifteen minutes. Printed one-pager, every script on it. Reps highlight the variant they'll use this week. The non-negotiable rule: read every script out loud before it ships. If it doesn't sound like a human texting a colleague, rewrite it. Reading aloud catches the "I just wanted to circle back" reflex that no amount of silent editing will.

Role-play, fifteen minutes. Pairs, three scenarios, four minutes each plus a minute of observer feedback, rotating roles. The three scenarios that cover the most failure surface: a live phone pickup where the buyer objects that they already have a tool; a live video recording graded on the first five seconds, proof density, and single-CTA discipline; and a breakup-email reply that says "bad timing, circle back in Q4," where the rep must lock a specific date and send one piece of value rather than saying "great, I'll reach out then." Walk the room with a stopwatch. If a rep runs past the script's time budget, stop them. Speed is a skill and it only develops under a clock.

Pitfalls, five minutes. Five, stated as fix pairs. Same message on different channels — fix: one job per channel. Skipping the phone — fix: score voicemails in one-on-ones. Generic "circling back" on the day-5 email — fix: reference a specific detail from the voicemail. A breakup email that begs — fix: short, dignified, one-word reopen path. Full cadence on cold ICP-only contacts — fix: bottom tier gets email and LinkedIn until they engage.

Action items, five minutes. Four assignments, all due Friday: build the sequence in the platform using the verbatim scripts and get manager approval before activation; load 25 top-tier or mid-tier triggered contacts, no cold-only; record one video and send it to the manager for grading before any prospect sees it; log every live phone pickup in a shared channel with one sentence on what happened. Name the single accountability metric out loud — sequence reply rate at end of week two — and write the absolute target number, not the relative lift.

The daily drill in that diagram is where the training actually sticks. Seven minutes before standup: read yesterday's voicemails aloud, score any videos sent, pair-drill three live-pickup openers, post the sequence reply rate. A single 60-minute session with no daily reinforcement decays to near-zero within three weeks — reps revert to whatever they were doing before, because that's what's comfortable. Seven minutes a day for two weeks is roughly 70 minutes of reinforcement, more than the original session, and it costs nothing on the calendar because it fits inside the standup slot you already own.

When week-two numbers come in below target, resist the urge to rewrite the scripts. The diagnostic order is: step completion first, then reply rate. Pull the sequence report and look for steps with near-zero completions. It is almost always the phone steps or the video step — the uncomfortable ones. A cadence where 40% of the touches never fire isn't underperforming on messaging; it isn't running. Fix completion before you touch copy, and check it in the Friday one-on-one where every rep brings sequence stats: reply rate, meetings booked, and one specific script line that's working. That last item is how good copy propagates across the team without a second training session.

Related questions

How often should we re-run this training?

Once as a full 60-minute install, then quarterly as a 20-minute refresh built around whatever the sequence data shows is breaking. Between refreshes, the daily seven-minute drill carries the load. Re-running the full hour more than quarterly signals a reinforcement problem, not a knowledge problem.

Does this work for inbound follow-up, not just cold outbound?

Yes, compressed. Inbound leads deserve a 5–7 day version with the same channel logic and much tighter spacing — first touch within minutes, phone on day one rather than day two. The thread rule and one-job-per-channel principle transfer directly; the 14-day patience does not.

What if reps refuse to make the phone calls?

Treat it as a skill gap, not a compliance problem. Reps avoid dials because they have no reliable opener and no voicemail they're proud of. Drill the 12-second live-pickup opener and the 20-second voicemail daily until both are automatic. Completion rates on phone steps usually recover within two weeks.

Can AI tools write the sequence copy for us?

They can draft it, and they'll produce competent, forgettable copy that reads exactly like everyone else's. Use them for research and first drafts, then have the rep rewrite the opener and the specific proof point in their own words. The read-it-aloud test catches AI-flat copy reliably.

How do we adapt this for a two-person sales team?

Run the same hour, cut role-play to one scenario, and drop the manager-approval gate to a peer review. Size the list to capacity — two reps cannot service 25 new contacts each per week on top of active deals. Ten to fifteen is realistic, and the cadence quality matters more than volume at that scale.

FAQ

How long until we can measure whether the cadence is working?

Two weeks gives you directional signal on step completion and early reply rate. Meaningful reply-rate comparison needs roughly 150–200 contacts through the full sequence, which for a small team is closer to four to six weeks. Judge completion at two weeks and reply rate at six; judging reply rate early on 40 contacts produces noise you'll overreact to.

Which sales engagement platform do we need?

Any platform that supports multi-step sequences with mixed manual and automated tasks will run this — Outreach, Salesloft, Apollo, HubSpot Sequences, and several others all qualify. The requirement is that manual tasks (dials, LinkedIn steps, video sends) surface as enforced to-dos with due dates rather than optional suggestions, and that per-step completion is reportable. Without step-level completion reporting you cannot diagnose a failing cadence.

Should every prospect get all four channels?

No. Reserve the full multichannel play for accounts with a real trigger event or strong tier fit. Cold ICP-only contacts get email and LinkedIn until they show an engagement signal, at which point phone and video unlock. Running the full effort on an untriggered list burns both the list and the rep's willingness to run the play at all.

What's the right list size per rep per week?

Twenty-five new contacts per rep per week is a defensible starting point for a dedicated SDR running a 10-touch cadence, and it consumes 10–17 hours. Full-cycle AEs carrying active deals should start at 10–15. If step-completion data shows phone and video steps being skipped, the list is too large — cut it before you cut the cadence.

How do we keep the scripts from going stale?

Rotate the proof point and the opener monthly; keep the structure fixed. The Friday one-on-one where each rep names one script line that's working is the harvesting mechanism — good lines propagate laterally and the shared one-pager gets updated with them. Rewriting the whole sequence quarterly destroys the comparison baseline you need to know whether anything improved.

Is a 14-day cadence too short for enterprise deals?

Usually, yes. Enterprise buying committees and longer evaluation cycles justify a six-to-eight-week cadence with more touches spread wider, often across multiple contacts at the same account. The channel logic and thread rule are identical; only the spacing and touch count change. Teach the compressed version first — reps who can run 14 days well can stretch it, but reps taught a 50-day cadence first tend to under-touch early and lose the thread.

Sources

• https://blog.hubspot.com/sales/sales-prospecting
• https://www.salesloft.com/resources
• https://www.outreach.io/resources
• https://blog.close.com/sales-cadence/
• https://www.gong.io/resources/
• https://blog.hubspot.com/sales/cold-calling-tips
• https://www.linkedin.com/business/sales/blog
• https://blog.hubspot.com/sales/sales-email-templates

Related on PULSE

• [Prospecting on LinkedIn: Interactive Template for a 20-Minute Sales Session](/knowledge/st0795)
• [Top 10 prospecting role-play scenarios for sales teams](/knowledge/st0544)
• [Top 10 prospecting training drills for B2B sales reps](/knowledge/st0543)
• [60-Min Sales Training: Video Prospecting (Loom, Vidyard)](/knowledge/st0439)
• [Top 10 Ready-to-Use Sessions for Prospecting Email Writing](/knowledge/st0681)
• [The Social Sell: Template for a Team Workshop on LinkedIn Prospecting](/knowledge/st0672)