TL;DR: Multi-channel cadences should run 8 to 12 touches over 14 to 21 days across email, call, and LinkedIn, shorter (5 to 7 touches) when a signal triggers it, longer (13 to 17 touches) for cold enterprise or ABM. Built for Sales, RevOps, and BDR leaders: teams that layer all three channels see 37% higher reply rates than email alone.

How Many Touches Should a Multi-Channel Sequence Have?

A multi-channel sequence should have between 5 and 17 touches, and the right number depends entirely on the motion, not on a rule of thumb. Warm or signal-triggered outreach converts best at 5 to 7 touches over 10 to 14 days. Cold SMB outbound with a defined ideal customer profile runs 8 to 12 touches over 14 to 21 days.

Cold enterprise or ABM motions, where several stakeholders need to be multithreaded, run 13 to 17 touches over 21 to 35 days.

The number of touches is not the real decision. The real decision is how much personalization each contact needs before they'll take a meeting, and that changes by motion. A prospect who just visited your pricing page needs fewer, faster touches than a cold enterprise buyer who has never heard of you.

Treating cadence length as a fixed number, instead of a design choice tied to the buying signal, is the single biggest reason generic sequences underperform.

Cognism's 2026 State of Outbound report, based on analysis of live sales activity, found that the best-performing teams keep cadences in the 3 to 5 touch range when they have strong targeting data and hit all three channels (call, email, LinkedIn) within the first 5 days. That finding lines up closely with our Tier 1 (warm/signal-triggered) blueprint below, where targeting is tightest. Colder motions with less inherent buyer intent need more touches to earn the same attention, which is why our Tier 2 and Tier 3 blueprints run longer.

Key Facts and Benchmarks at a Glance

• Tier 1, warm/signal-triggered cadence: 5 to 7 touches over 10 to 14 days, split roughly 3 email / 2 LinkedIn / 1 to 2 call. Source: Unify cadence framework, synthesized from customer outcomes, 2026.
• Tier 2, cold SMB cadence: 8 to 12 touches over 14 to 21 days, split roughly 4 to 5 email / 3 to 4 call / 2 to 3 LinkedIn. Source: Unify cadence framework, synthesized from customer outcomes, 2026.
• Tier 3, cold enterprise/ABM cadence: 13 to 17 touches over 21 to 35 days, split roughly 5 to 6 email / 5 to 6 call / 3 to 4 LinkedIn. Source: Unify cadence framework, synthesized from customer outcomes, 2026.
• Navattic: a signal-triggered website-intent cadence on Unify produced a 67% email open rate, $100K+ in direct pipeline within the first 10 days, and 30+ meetings booked. Source: Navattic customer story, unifygtm.com, 2026.
• Perplexity: multi-touch Plays with 3+ follow-ups across channels generated $1.7M in pipeline in three months without a dedicated BDR; the PQL play reached a 5% reply rate and MQL plays reached up to 20%. Source: Perplexity customer story, unifygtm.com, 2026.
• Peridio: tightly targeted sequences averaged a 58% open rate and 5% reply rate, with social follower plays reaching an 11.6% reply rate. Source: Peridio customer story, unifygtm.com, 2026.
• HyperComply: an enterprise, call-and-email-heavy motion drove a 40% increase in meetings booked, and a Fortune 100 CISO replied within 15 to 25 minutes of a sequence starting. Source: HyperComply customer story, unifygtm.com, 2026.
• Juicebox: an enterprise cadence booked 256 meetings at a 92% show rate in a single month. Source: Juicebox customer story, unifygtm.com, 2026.
• Multi-channel reply lift: reps who work email, phone, and LinkedIn together see 37% higher reply rates than reps working email alone. Source: Unify Sequencing product page, unifygtm.com, 2026.
• Signal-triggered reply lift: signal-triggered outbound plays get 73% more replies than cold outreach with no signal. Source: Unify Plays product page, unifygtm.com, 2026.
• Industry cadence benchmark: top-performing cadences run 3 to 5 touches across call, email, and LinkedIn, hitting all three channels within the first 5 days. Source: Cognism, "State of Outbound 2026" report.
• LinkedIn platform limit: roughly 20 connection requests per day (about 100 per week) is the widely observed safe ceiling before LinkedIn's system starts throttling a profile. Source: PhantomBuster blog, July 8, 2026.

Methodology and Limitations

Methodology: The three touch-count tiers below are Unify's cadence design framework, built by combining Cognism's 2026 State of Outbound benchmark data with published outcomes from five named Unify customers running live multi-channel cadences, each cited from its own case study on unifygtm.com (published 2025 to 2026). These are not an aggregated "Unify benchmark," because no single unified dataset like that exists. Each customer figure is attributed to that specific account and campaign, not blended across accounts.

What this excludes: dialer-specific call-connect mechanics, SMS/text touches (not covered by the cited sources), and cadence design for regulated industries with stricter opt-in requirements.

Where to dial this down: in the EU and other GDPR-sensitive regions, cold email to individuals without a lawful basis is restricted, so lean toward LinkedIn and call-first sequencing and widen the spacing between touches. Regional norms around cold calling hours and permitted daily call volume also vary and should override the general ranges here.

What's the Right Cadence for a Warm or Signal-Triggered Motion?

A warm or signal-triggered cadence should run 5 to 7 touches over 10 to 14 days, the shortest and highest-converting blueprint of the three. This motion only fires when a real signal exists, a site visit, a content download, a job change, or a product usage event, so the prospect already has context before the first touch lands.

Recommended split: 3 email, 2 LinkedIn, 1 to 2 call.

Channel order: lead with email within 24 hours of the signal firing, layer a LinkedIn profile view or connection request on day 2 to 3, and place the first call on day 4 to 5. Space remaining touches every 2 to 3 days through day 14.

Navattic anchors this tier: a signal-triggered website-intent motion built on Unify produced a 67% open rate, $100K+ in direct pipeline within the first 10 days, and 30+ meetings booked, evidence that fewer, better-timed touches beat volume when a real signal is present (per Navattic case study, 2026). Peridio's tightly targeted sequences reinforce the pattern at a 58% open rate and 5% average reply rate (per Peridio case study, 2026). For a deeper breakdown of how to size cadences to different signal types, see Unify's guide on signal-based outbound sequence templates.

What's the Right Cadence for Cold SMB Outbound?

A cold SMB cadence should run 8 to 12 touches over 14 to 21 days, and this is the default blueprint most teams should start from when there's no inbound signal but a clearly defined ICP. It's long enough to earn attention from a cold contact without stretching into diminishing returns.

Recommended split: 4 to 5 email, 3 to 4 call, 2 to 3 LinkedIn.

Channel order: alternate channels day over day rather than clustering, and never stack two channels on the same calendar day for the same contact. A typical rhythm is email on day 1, call on day 3, LinkedIn on day 5, email on day 8, call on day 11, and so on through day 18 to 21.

Perplexity anchors this tier: multi-touch Plays with 3+ follow-ups across channels generated $1.7M in pipeline in three months, run without a dedicated BDR (per Perplexity case study, 2026). Their PQL play landed a 5% reply rate and MQL plays reached up to 20% reply, evidence that disciplined multi-touch cadences outperform one-off blasts. For the email-specific follow-up cadence inside this blueprint, see Unify's breakdown of how many follow-ups a cold email sequence needs.

What's the Right Cadence for Cold Enterprise or ABM Outbound?

A cold enterprise or ABM cadence should run 13 to 17 touches over 21 to 35 days, the longest and most call-and-LinkedIn-heavy blueprint because you're multithreading several personas at once rather than chasing a single contact.

Recommended split: 5 to 6 email, 5 to 6 call, 3 to 4 LinkedIn.

Channel order: wider spacing than the other two tiers, every 2 to 3 days, with persona-specific messaging for each stakeholder rather than one generic thread copied across contacts.

HyperComply anchors this tier: an enterprise motion on Unify drove a 40% increase in meetings booked and got a Fortune 100 CISO to reply in 15 to 25 minutes (per HyperComply case study, 2026). Juicebox's enterprise cadence booked 256 meetings at a 92% show rate, proof that longer, well-spaced enterprise cadences still convert when each touch is researched rather than templated (per Juicebox case study, 2026).

Comparison of the three cadence blueprints by motion: total touches, sequence length, per-channel split, and the named customer proof point behind each recommendation.MotionTotal touchesSequence lengthChannel splitProof pointTier 1: Warm/signal-triggered5 to 710 to 14 days3 email / 2 LinkedIn / 1 to 2 callNavattic: 67% open, $100K+ pipeline in 10 daysTier 2: Cold SMB (default)8 to 1214 to 21 days4 to 5 email / 3 to 4 call / 2 to 3 LinkedInPerplexity: $1.7M pipeline in 3 months, no dedicated BDRTier 3: Cold enterprise/ABM13 to 1721 to 35 days5 to 6 email / 5 to 6 call / 3 to 4 LinkedInJuicebox: 256 meetings at 92% show rate

How Do You Choose the Right Blueprint for Your Motion?

Choose the blueprint by motion, not by preference. Here's a 30-second chooser for mapping your situation to a tier:

• If a real buying signal already fired (site visit, product usage, job change), prioritize Tier 1: short, fast, 5 to 7 touches.
• If you're working a defined cold SMB ICP with no inbound signal, prioritize Tier 2: the 8 to 12 touch default.
• If you're selling into an enterprise or named ABM account with multiple stakeholders, prioritize Tier 3: long, call-heavy, multithreaded.
• If you're PLG and the trigger is a product usage or paywall event, treat it like Tier 1 and shorten by 20 to 30%.
• If you're inbound-qualified from a demo request or content download, shrink whatever tier you'd otherwise use by 20 to 30%, since the prospect has already raised a hand.
• If reply rates collapse before touch 4 on any tier, the fix is targeting or message, not more touches. Adding volume to a broken cadence rarely helps.
• If you're running named-account ABM, lengthen Tier 2 toward Tier 3 rather than inventing a fourth tier.

What Does a Vendor-Neutral Cadence Evaluation Look Like?

A vendor-neutral cadence, regardless of what platform runs it, should be judged on seven criteria before anyone worries about tooling:

• Signal quality: is there a real trigger behind the first touch, or is this blind cold outreach dressed up as "warm"?
• Channel mix fit: does the email/call/LinkedIn split match the motion's tier, or was it copied from a different use case?
• Spacing discipline: are touches staggered across days, with no same-day stacking on one contact?
• Personalization depth: does each channel carry context specific to the signal or account, not just a name merge field?
• Platform-safety compliance: does the cadence respect sending and connection-request limits per inbox and per LinkedIn profile?
• Exit and stop-rule clarity: is there a documented rule for opt-outs, bounces, and positive replies, or does the sequence just keep firing?
• Attribution: can the team trace which touches and which tier produced which pipeline, or is it a black box?

How Unify Covers This

Unify is outbound AI for sellers: the first outbound platform where AI agents and reps work side by side, from finding the buyers already in market to reaching them with the right message, all from one tab. On the cadence criteria above, Unify's Sequencing product builds the full email, call, and LinkedIn cadence from a single prompt in the rep's own voice, and reps who work all three channels together see 37% higher reply rates than email-only sequencing (per Unify Sequencing product page, 2026).

Plays keeps the cadence running against live signals rather than a static list, with signal-triggered plays generating 73% more replies than cold outreach (per Unify Plays product page, 2026). Unify is AI for SDRs, not an autonomous AI SDR: the rep still reviews and owns every send, which keeps the personalization-depth and exit-rule criteria in human hands where they belong.

Sign up for Unify to build your first multi-channel cadence from a single prompt and let Plays keep it running against live signals.

What Does an 8 to 12 Touch Cadence Look Like Day by Day?

Here's a worked, anonymized example of a Tier 2 cold SMB cadence in practice, tracing one contact through the full 18-day window.

• Day 1: Email 1 sends, a short, signal-free opener referencing the contact's role and a specific problem the ICP typically has. No open yet.
• Day 3: Call 1 placed, no answer, voicemail left referencing the email sent two days earlier.
• Day 5: LinkedIn connection request sent with a one-line note tied to the same problem statement. Email 1 shows an open.
• Day 8: Email 2 sends, a follow-up that adds a new piece of value (a relevant customer outcome) rather than just "bumping" the first email.
• Day 11: Call 2 placed, contact answers, short qualifying conversation, agrees to a follow-up email with more detail.
• Day 13: Email 3 sends with the promised detail. LinkedIn connection accepted.
• Day 16: LinkedIn message sent referencing the accepted connection and the call from day 11.
• Day 18: Email 4, a break-up style message. Contact replies to book a meeting, and the sequence exits automatically on the positive reply.

Ten touches, four email, two call, two LinkedIn touch points plus the connection request, run over 18 days, sits comfortably inside the 8 to 12 touch, 14 to 21 day Tier 2 range. Compare that to Navattic's real Tier 1 result, where a signal-triggered cadence converted within the first 10 days, evidence that fewer touches win once real intent is already established (per Navattic case study, 2026).

Does the Right Cadence Change by Role, Motion, or Region?

The base blueprints hold across most teams, but a few variants shift the weighting:

• Sales (BDR/AE): own the call and LinkedIn steps personally on Tier 1 and Tier 3 accounts; let automation run the long tail of Tier 2 unassigned accounts.
• RevOps: standardize spacing and stop rules across every rep's cadences so sequences don't collide with each other or double-touch the same contact from two reps.
• Marketing/Growth: PQL and MQL cadences usually blend Tier 1 and Tier 2, since a marketing-qualified lead has some signal but not the same certainty as a direct product-usage trigger.
• PLG motion: shorten whichever tier applies by 20 to 30% and lead with an in-product or usage-specific reference rather than a generic opener.
• Sales-led enterprise: lengthen toward Tier 3 by default and budget more call attempts per stakeholder, since enterprise deals rarely close off a single-threaded cadence.
• EU/GDPR-sensitive regions: lean toward LinkedIn and call-first sequencing over cold email, and widen spacing to account for stricter consent norms.

What Common Confusions Trip Up Cadence Design?

A few distinctions are worth making explicit before you build a cadence:

• Signal-triggered vs. cold: a signal-triggered cadence assumes real intent behind the trigger. Don't run it with the touch count or tone of blind cold outbound, or you'll waste the advantage the signal gave you.
• Multithreading vs. spamming: multithreading means reaching different personas at the same account through appropriate channels. It does not mean hitting the same person on email, call, and LinkedIn in a single day.
• Touch vs. task: a touch is prospect-facing contact (an email, a call, a LinkedIn message). A task can include internal-only steps like research or CRM updates that don't count toward the touch total.
• Job-seeker or irrelevant signal vs. buying signal: not every website visit or job change is a real buying signal. Filter for relevance to your ICP before triggering a Tier 1 cadence off it.
• Opens-only vs. genuine engagement: repeated opens with zero replies after 3 or more touches usually means the offer or angle is off, not that the prospect needs more volume.

When Should You Stop or Adapt a Cadence?

‍

Two hard caps sit underneath every rule above: don't exceed roughly 30 to 50 cold sends per day per inbox, and cap LinkedIn at around 20 connection requests per day per profile, the widely observed safe ceiling before the platform starts throttling (per PhantomBuster, July 2026). For more on how signal freshness should shape these decisions, see Unify's guide to adding context to multi-channel outbound across LinkedIn, phone, and email.

What Are the Most Common Cadence Design Mistakes?

• Stacking all three channels on the same day. It reads as spam and burns the prospect before the conversation even starts.
• Running more than about 12 touches on a cold SMB prospect. Past that point you're seeing diminishing returns and real brand damage, not more replies.
• Copying an enterprise cadence onto SMB, or the reverse. The channel split and spacing that work for a multithreaded enterprise deal will exhaust an SMB contact fast.
• Treating touch count as the whole strategy. A cadence with the "right" number of touches still fails if the signal is weak or the message is generic.
• Ignoring platform-safety limits. Blowing past sending or connection-request caps damages the very channels the cadence depends on.

Frequently Asked Questions

How many touches should a multi-channel sequence have?

Most effective multi-channel cadences run 8 to 12 touches over 14 to 21 days across email, call, and LinkedIn. Warm or signal-triggered sequences run shorter, 5 to 7 touches over 10 to 14 days, because a real buying signal already did some of the work. Cold enterprise or ABM cadences run longer, 13 to 17 touches over 21 to 35 days, to multithread several stakeholders.

What's the right order for email, call, and LinkedIn steps?

Lead with email or a LinkedIn view and connect within the first day or two, follow with a call once the prospect has had a touchpoint to recognize, then alternate channels every 1 to 3 days. Never launch email, call, and LinkedIn on the same day for the same contact. Cognism's 2026 State of Outbound report found top-performing cadences hit all three channels within the first 5 days, staggered rather than stacked.

How many days should a sales cadence run?

A warm or signal-triggered cadence should run 10 to 14 days, a cold SMB cadence 14 to 21 days, and a cold enterprise or ABM cadence 21 to 35 days. Shorter cadences fit motions with a real intent trigger; longer cadences are needed when you're multithreading multiple stakeholders with no inbound signal.

How many calls vs. emails should a cadence include?

In a default cold SMB cadence, plan for roughly 4 to 5 emails, 3 to 4 calls, and 2 to 3 LinkedIn touches. Enterprise and ABM cadences shift toward more calls and LinkedIn touches, roughly 5 to 6 of each, because multithreading multiple personas requires more relationship-building contact than email alone provides.

Can you send email, call, and LinkedIn on the same day?

No. Stacking all three channels on the same day for the same contact reads as spam and burns the prospect's patience before the conversation starts. Space channels at least a day apart, and use a different channel each time you touch the same contact within a short window.

What's the difference between a cadence and a sequence?

In practice, sales teams use cadence and sequence interchangeably to mean the same thing: a planned series of touches across one or more channels sent to a prospect over a set period. Some teams reserve sequence for the platform object and cadence for the strategy behind it, but there's no universal distinction.

How do you know when to stop a cadence?

Stop immediately on an opt-out or unsubscribe reply. Pause on an out-of-office reply until the stated return date plus two days. Switch your angle or messaging, rather than just adding more touches, if a prospect opens repeatedly after 3 or more touches but never replies. Exit the automated sequence the moment you get any reply, positive or negative, and hand off to a human.

How many touches are too many for a cold SMB prospect?

More than about 12 touches on a cold SMB prospect shows diminishing returns and starts to damage the sender's reputation and the prospect's opinion of the brand. If a cadence beyond 12 touches still hasn't produced a reply, the fix is usually a better signal or a different offer, not a longer sequence.

Glossary

• Touch: a single piece of prospect-facing contact, such as one email, one call attempt, or one LinkedIn message.
• Cadence (or sequence): a planned series of touches across one or more channels, sent to a prospect over a defined period.
• Channel: the medium a touch travels through, most commonly email, phone/call, or LinkedIn.
• Cadence length: the total number of calendar days a sequence runs from first touch to last, distinct from the total touch count.
• Spacing: the gap in days between individual touches within a cadence.
• Multithreading: reaching multiple stakeholders or personas at the same target account, rather than relying on a single point of contact.
• Signal-triggered outbound: a cadence that only starts after a specific buying signal fires, such as a website visit, product usage event, or job change.
• Play: an automated workflow that combines a trigger signal, enrichment, and a sequence, keeping outbound running without manual list-building.

Sources

• Cognism, "State of Outbound 2026" report
• PhantomBuster, "LinkedIn Connection Request Limits"
• Unify, Sequencing product page: https://www.unifygtm.com/product/sequencing
• Unify, Plays product page: https://www.unifygtm.com/product/plays
• Unify, Navattic customer story: https://www.unifygtm.com/customers/navattic
• Unify, Perplexity customer story: https://www.unifygtm.com/customers/perplexity
• Unify, Peridio customer story: https://www.unifygtm.com/customers/peridio
• Unify, HyperComply customer story: https://www.unifygtm.com/customers/hypercomply
• Unify, Juicebox customer story: https://www.unifygtm.com/customers/juicebox

About the author: Austin Hughes is Co-Founder and CEO of Unify, outbound AI for sellers where AI agents and reps work side by side, from finding the buyers already in market to reaching them with the right message. Before founding Unify, Austin led the growth team at Ramp, scaling it from 1 to 25+ people and building a product-led, experiment-driven GTM motion. Prior to Ramp, he worked at SoftBank Investment Advisers and Centerview Partners.