Pre-conference outreach

Pre-conference outreach

Template placeholders

Replace every {{...}} before enabling. See the setup checklist reference for the full setup list.

• {{EVENT_NAME}} - The conference as people say it in a DM (e.g. "GTM2026", "SaaStr")
• {{EVENT_CITY}} - Host city, used to pick "hitting the conf?" vs "Making the trip?"
• {{EVENT_DATES}} - Date range, used to pick "this week" vs "next week" at send time
• {{EXEC_SIDE_EVENT}} - Your executive side-door offer, as it reads in a sentence (e.g. "an exec dinner on Tuesday night")
• {{PRACTITIONER_SIDE_EVENT}} - Your hands-on side-door offer (e.g. "a skills hackathon during the conference", "a build session")
• {{EXEC_SENDER}} - Teammate with the authority lens: talks to whoever owns the problem and its budget
• {{SALES_LEADER_SENDER}} - Teammate with the seller-management lens: talks to people who manage sellers
• {{PRACTITIONER_SENDER}} - Teammate with the practitioner lens: talks to hands-on operators of the GTM stack
• {{EXEC_SENDER_MIN_ORG_SIZE}} - Employee count above which {{EXEC_SENDER}} may send (default: 100)
• {{CRM}} - Your CRM (e.g. HubSpot, Attio, Salesforce)
• {{SIGNAL_LIBRARY}} - Optional: your list of verified third-party angles and how to verify each (funding, hiring, product launches, stack changes)
• {{ATTENDEE_TAG}} - The tag or memory line you use to record "this person attended {{EVENT_NAME}}"
• {{M1_CHAR_CAP}} - Hard character cap on the first email body (default: 250)
• {{VOICE_GUIDE}} - Optional: a sender's voice guide to load when they are the sender

---

When to use this page

An attendee list for {{EVENT_NAME}} (or an explicit "run pre-conference outreach on these people") needs sequences that drive a meeting before or during the conference, plus a side-door CTA: {{EXEC_SIDE_EVENT}} for the exec and sales-leader senders, {{PRACTITIONER_SIDE_EVENT}} for the practitioner sender.

This is a reusable play. Swap the event fill-ins when the next conference is the instance. Sequence everyone on the file. Score or scoring tier does not route or skip anyone.

Do not use this play when:

• They registered for or attended an event and the ask is generic warm-up or post-event follow-up with no side-door CTA. That is a generic event-registrant play (a separate play, not included here).
• The list is booth or badge scans after the conference. That is post-conference outreach.
• You already sat with them at an in-person dinner. That is a dinner follow-up (a separate play, not included here).
• {{CRM}} shows the account is a current customer or closed-won (including alias or sibling domains). Existing-customer hard stop. Do not sequence.

Approval is always queued for a human. There is no auto-send carve-out on this play.

---

Event fill-ins

Fill-in	Value
{{EVENT_NAME}}	e.g. GTM2026
{{EVENT_CITY}}	e.g. NYC
{{EVENT_DATES}}	e.g. September 28 - October 1
{{EXEC_SIDE_EVENT}}	e.g. an exec dinner on Tuesday night
{{PRACTITIONER_SIDE_EVENT}}	e.g. a skills hackathon during the conference

Phrasing that depends on timing and location is decided at send time, not hardcoded: "this week" vs "next week"; "hitting the conf?" if they are based in {{EVENT_CITY}} vs "Making the trip?" if they are traveling in.

---

HARD STOP - research every account on the upload, without exception

This is not optional and it is not only a first-party signal scan. Do not draft M1, assign a sender, or stage a sequence for a row until research has run on that domain.

Skipping research because "they're probably cold", "no CRM match", or "it's a list upload" is a failure.

For every row, read both:

1. Account memory for the company (the contact's domain, plus any alias or sibling domain already on the account).
2. {{CRM}} activity at the company level, not just this contact. Pull the company's engagements, notes, meetings, emails, and open or closed deals. Scan every associated contact, not only the person on the list.

If a prior deal exists (open, closed-won, or closed-lost), research the deal context the way your closed-lost re-engagement procedure does. Run it. Do not fork the procedure here. At minimum:

• {{CRM}} activity feed on the company and every associated contact (emails, meetings, notes, close reason, blockers named)
• Call transcripts if meetings appear
• Whether they ever trialed, how far it got, and any obvious setup gaps
• Capture churn or loss reasons verbatim. The close-reason field is usually a one-line summary that hides the full list.

Closed-won or customer (including alias domains) still hits the existing-customer hard stop: do not sequence.

If research surfaces material history (open deal, booked meeting, live thread, closed-lost with a stated "not now", prior trial, another owner working it), do not pretend the account is cold. Use that history as first-party context in M1 when it is still appropriate to sequence, or stop and flag when a standard pre-send gate trips (live unanswered inbound thread, account already engaged, booked meeting, active owner).

Empty memory plus a quiet {{CRM}} feed is still a completed research pass. Then fall through the M1 ladder below. Record what was read in the reasoning log:

research: [memory + CRM records actually read]; prior deal: none | [deal stage + one-line context]

If that line is missing, the research did not happen. Do not build.

This is in addition to your standard live-thread and account-engagement gates, not a substitute for them.

---

Sender and outcome (persona lens, not account tier)

This play overrides tier-based sender routing. Sender is the persona owner for the contact's actual role: judgment on the real job, not a fixed title list. Titles vary wildly company to company ("VP GTM Programs", "Global Head of Revenue Strategy, Ops & Enablement", "GTM Systems & Operations" appear on no list and are exactly the right people). One contact, one sender.

Use a verified employee count when available (a LinkedIn-verified count is the preferred source); otherwise the count already on the account record. State the count and the resulting sender in the reasoning log.

Sender	When	Mandate	Desired outcome
{{EXEC_SENDER}}	Only if org > {{EXEC_SENDER_MIN_ORG_SIZE}} employees AND the contact owns the problem or the budget (C-level or VP-equivalent who owns GTM systems and tooling decisions)	Authority lens, not hands-on-keyboard	Meeting before or during the conference, and/or {{EXEC_SIDE_EVENT}}
{{SALES_LEADER_SENDER}}	People who manage sellers (VP or Head of Sales, VP of Sales Development, Head of SDR/BDR; Director of Sales only if no department head). Also exec-mandate contacts at <= {{EXEC_SENDER_MIN_ORG_SIZE}} employees	Seller-management lens, plus small-org exec fallback	Meeting before or during the conference, and/or {{EXEC_SIDE_EVENT}}
{{PRACTITIONER_SENDER}}	Hands-on practitioners (GTM engineers, RevOps who run the stack, growth and funnel leads, director-level marketing ops or GTM systems owners). No employee-count gate.	Practitioner lens	Meeting before or during the conference, and/or {{PRACTITIONER_SIDE_EVENT}}

Exec size gate (hard): if employee count is {{EXEC_SENDER_MIN_ORG_SIZE}} or below, do not assign {{EXEC_SENDER}} even when the title is CRO, CMO, CEO, or VP RevOps. Route that contact to {{SALES_LEADER_SENDER}} and keep the exec side-door CTA (Scenario 1). If the count is unknown after a reasonable lookup, do not default to {{EXEC_SENDER}}. Use {{SALES_LEADER_SENDER}} and note the unknown count.

Hard exclusions:

• {{PRACTITIONER_SENDER}} never connects with salespeople or sales managers. AE, SDR/BDR, Account Executive, Sales Manager, Sales Director, Head of Sales, VP Sales, or any equivalent quota-carrying or seller-management title goes to {{SALES_LEADER_SENDER}} (Scenario 1).
• Prefer director-level operators over manager-level RevOps for {{PRACTITIONER_SENDER}}.
• Prefer VP or Head over Director of Sales for {{SALES_LEADER_SENDER}}.

If the upload already names a sender, still verify the mandate and the exec size gate; reassign if either fails.

If a row has no resolvable LinkedIn URL, skip the LinkedIn steps and flag the row. Do not invent a URL. If there is no work email after enrichment, fall back to LinkedIn-only rather than emailing a personal inbox.

Load {{VOICE_GUIDE}} when it exists for the chosen sender.

---

Two scenarios, same 4-step shape

Scenario 1 - Meetings + exec side event ({{EXEC_SENDER}} when over the size gate, otherwise {{SALES_LEADER_SENDER}})

Scenario 2 - Meetings + practitioner side event ({{PRACTITIONER_SENDER}}). M1 can still ask for a meeting. M2 is the only structural change.

Step	Timing	Channel	Content
1 / M1	Day 0	Email	Conference attendance is the primary hook, tied to a researched angle. Ask for a meeting. Max {{M1_CHAR_CAP}} characters.
2	Immediate (same day as M1)	LinkedIn connection request	Blank. No note. Auto-skips if already connected.
3 / M2	Day 3	Email	Side-door offer. Verbatim. No name in the message.
4 / M3	After M2 (next LinkedIn touch)	LinkedIn DM	Verbatim. No name in the message.

---

Scenario 1 - Meetings + exec side event

Who: {{EXEC_SENDER}} (over the size gate + authority-lens contact) or {{SALES_LEADER_SENDER}} (seller-management lens, or exec-mandate at or under the gate).

M1: researched email per the ladder below. Meeting ask.

Step 2: blank connection request.

M2 - email, Day 3, verbatim (no name):

We're also hosting {{EXEC_SIDE_EVENT}} if that's preferable - can shoot over the reg link if interested. Lmk!

Do not paste the registration link in the email. Offer to send it.

M3 - LinkedIn DM, verbatim (no name):

tried you on email - in town for {{EVENT_NAME}} this week?

Adjust "this week" / "next week" to the actual timing of the send vs the conference. Do not add a name.

---

Scenario 2 - Meetings + practitioner side event

Who: {{PRACTITIONER_SENDER}}. The angle should be relevant to technical personas (GTM engineers, RevOps, marketing ops, GTM systems, as examples). Do not use a seller-management angle on this path.

M1: researched email per the ladder below. Meeting ask is still allowed.

Step 2: blank connection request. {{PRACTITIONER_SENDER}} never connects with salespeople or sales managers, so those contacts should not be on this scenario.

M2 - email, Day 3, verbatim (no name):

We're also running {{PRACTITIONER_SIDE_EVENT}} if that's more your speed - can shoot over the details if interested. Lmk!

Do not paste the side-event link in the email. Offer to send details.

M3 - LinkedIn DM, verbatim (no name), same as Scenario 1:

tried you on email - in town for {{EVENT_NAME}} this week?

---

M1 - research ladder and copy

Conference attendance is the primary signal and may be named: it is a public, intentional act, the same logic that lets you name an event registration or a profile view. After the mandatory account research, pick the secondary angle with this ladder. Stop at the first that survives.

1. First-party. Material engagement found in the research: LinkedIn content engagement, website visits, prior replies, closed-lost or prior-deal context, {{CRM}} activity. If it exists, that is the secondary angle. Do not also weave in a third-party observation.
2. Else third-party. Load {{SIGNAL_LIBRARY}}. Run its verification protocol. Pick one verified angle and tie it to conference attendance. Role-to-angle mapping still applies. Scenario 2 uses technical angles only.
3. Else conference-only. Attendance plus a meeting ask. No invented signal.

Never mention that you pulled them from a registration or attendee list. Never dump product in M1.

Hard cap: {{M1_CHAR_CAP}} characters on the M1 email body (count the full message, including the name if used). If a draft exceeds the cap, cut before staging. Do not keep both a long observation and a long ask. Conference + one angle + one meeting ask is the whole message. Record the character count in the reasoning log.

M1 is allowed to ask for time ("How's your sched looking?", "Making the trip?"). This play is exempt from your cold-outreach no-mention, no-pitch, no-CTA rules. A persona-matched pitch check is not applicable on this short cadence: there is no pitch step.

Angle examples (reference only. Write fresh, stay under the cap):

First-party (LinkedIn engagement):

You've been engaging a lot with our content on LinkedIn. Given we're both in {{EVENT_CITY}} this week for {{EVENT_NAME}}, thought it might be a great time to connect. How's your sched looking?

Third-party (GTM scale-up):

Noticed you guys are in rapid GTM scale-up mode. So given we're both in {{EVENT_CITY}} this week for {{EVENT_NAME}}, thought it might be a great opportunity to connect re how AI agents are supporting the scale-up efforts. How's your sched looking?

Third-party (the prospect deploys AI in their own product). Traveling in:

Given the focus on deploying AI agents into call centers, was curious if agents are also driving your own GTM motion. Happy to chat on it live if you're at {{EVENT_NAME}} next week. Making the trip?

Local variant of the close: hitting the conf?

---

List-upload execution

• Sequence every row on the file unless a standard gate trips (customer, live thread, booked meeting, account already engaged, already-engaged owner).
• Research every row first. No exceptions.
• Check existing sequences per contact; read the full message history; edit, don't duplicate.
• Blank connection requests (no note). Plain-text URLs. No unfilled merge fields left in copy. No markdown-wrapped links.
• Stage queued for approval. Confirm: "Drafted a 4-step pre-conference sequence for [contact] at [company], staged for your approval. Sender: [name]. Hook: [one line]. Scenario: exec side event / practitioner side event."
• After staging: write the person-level attendee line to account memory and apply {{ATTENDEE_TAG}} on the company. Write it for every sequenced row and for connection-only rows. Do not duplicate an existing person+event line.

The reasoning log on every sequence must include:

• Sender mandate + why (quote the role, not just the title)
• Employee count + exec size gate verdict
• research: + prior deal: line (see HARD STOP)
• First-party vs third-party vs conference-only, and what was used
• Scenario 1 vs 2
• Event instance
• M1 character count
• A compact pre-flight line citing this play's sender override, not tier routing

---

Rules

• MUST research every account on the upload before drafting or staging. No exceptions.
• MUST run your closed-lost research procedure when a prior deal exists. Do not invent deal context.
• MUST assign sender from the persona mandate + exec size gate. Do not use tier routing.
• MUST keep M1 <= {{M1_CHAR_CAP}} characters.
• MUST use Scenario 1 (exec side event) for {{EXEC_SENDER}} and {{SALES_LEADER_SENDER}}, and Scenario 2 (practitioner side event) for {{PRACTITIONER_SENDER}}. Do not mix CTAs.
• MUST leave M2 and M3 verbatim and nameless (aside from this-week / next-week timing on M3).
• MUST NOT paste side-event registration links into M2. Offer to send them.
• MUST NOT mention the attendee list as the source.
• NEVER auto-send on this play.
• NEVER sequence a closed-won or customer account.
• MUST, after staging, write the attendee line to account memory and apply {{ATTENDEE_TAG}} on the company. One line per person per event.

---

What good looks like

Every row on the file has a research line before it has a draft. Sender was chosen by what the person actually does, and the exec size gate was applied even when the title said CRO. M1 reads like a 30-second note from someone who is also going to be in the room: the conference, one real angle, one ask for time, under the cap. M2 and M3 are word-for-word the verbatim lines, with no name added and no link pasted. Everything sits queued for approval with a reasoning log a colleague could audit.

Spot first: a row with no research line, an account under the size gate assigned to the exec sender, a salesperson routed to the practitioner sender, and an M1 over the cap.

Mediocre looks like: "saw you're attending, want to meet?" sent to everyone with no angle; a pitch paragraph in M1; a dinner link dropped in M2; a hackathon offer sent to a VP Sales; or a closed-lost account sequenced as if it were cold because nobody opened the deal history.