Documentation / Execution

Sequences

What the lead experiences over time — steps, waits, channel switches, and the limits that protect your sending account.

A Sequence is a lead’s contact plan over time: which message, on which channel, after which wait. It is the third of the three core nouns:

Table = per row · Sequence = per lead over time · Workflow = per event.

What a step carries

Every step has a type, a text and a wait:

TypeChannelWhat happens
connection_requestLinkedInConnection request, optionally with a note
messageLinkedIn, WhatsAppDirect message — requires an existing connection
emailEmailSent through the configured sending account

The wait counts from the moment the previous step was sent, not from enrollment. A plan of 0 / 3 / 5 days therefore spans eight days. A random spread of up to three minutes is added on top, so sending doesn’t look like clockwork.

Texts carry placeholders such as {{firstName}} or {{city}}, filled per lead at send time.

Character limits LinkedIn enforces hard

  • Connection note: 200 characters. Applies to free and Sales Navigator accounts. LinkedIn rejects longer notes at send time — the platform checks this at enrollment so it doesn’t surface live.
  • Direct message: 8,000 characters technically. In practice, short messages perform considerably better.

Connecting without a note: leave the note empty. This is required when the account has no Sales Navigator — a note needs premium there, or the send fails.

Status decides whether anything goes out

draftapprovedactivecompleted. Only active sends.

The most common cause of “nothing is happening” lives right here: a sequence still sitting in draft, or an active one with zero steps. Both look like a running campaign from the outside.

Limits protect your account, not the system

The real ceiling is the weekly connection-request limit — 80 per account by default. On top of that: a daily limit, distribution across hours, a warm-up curve for new accounts, and a cap on open, unanswered requests.

The practical consequence: volume comes from more sending accounts — which scales linearly and stays safe — not from raising limits. A dormant second account halves your throughput without producing an error.

Branching and switching channels

Optional per sequence: a graph instead of a fixed list. Steps then hang off conditions and a lead can change channel — for example from LinkedIn to WhatsApp when the connection was accepted but no reply came. Linear stays the default; the graph is the exception for cases that genuinely need it.

Enrollment, replies, outcome

Enrollment happens from a table: the final enroll column hands the finished row to a sequence. That is the single transition from bulk work into the contact plan.

When a lead replies, their sequence pauses — the system does not keep talking while a human answers. The reply lands in the Feed, and a positive reply becomes a Deal.

Metrics

Per sequence: sent, opened, replied, positively replied. Weighted variants allow A/B tests within one sequence, with metrics kept separate per variant.

Every send leaves behind a Run.