Attendee follow up
@eventcrew/attendee-follow-up
Turn badge scans and booth notes into a follow-up plan within 48 hours: sort contacts into hot, warm, and nurture, draft a personal email for each hot lead and a template for the rest, and export a CSV ready for your CRM or email tool. Drafts only; never sends.
- Activity
- New: be the first to try it
- Installs
- 0 installs
- Likes
- 0 likes
- Clones
- 0 clones
How this works
- It is a skill: written instructions Claude reads when your request matches what the skill is for. You do not call it by name; just ask for the task.
- While it works, it reads and changes files in the places it lists, nowhere else.
- Nothing runs until you choose to use it, and you can remove it any time.
SKILL.md
Attendee follow-up
Leads go cold fast: reply rates drop sharply after the first two days. This kit turns a scan export and scribbled booth notes into a ranked follow-up plan and drafts, so the first emails go out while people still remember the conversation.
When to use
- The day after an event, ideally within 48 hours.
- "Follow up with everyone we met", "draft emails for the booth leads", "who are the hot ones?"
- Webinar attendees and no-shows too: treat attended as warm and registered-but-missed as nurture.
Inputs
- Contacts:
data/attendees.csv, exported from the lead retrieval app, registration list, or webinar tool. Useful columns: name, email, company, title, notes, rating or qualifier answers, scanned by, attended (yes/no). - Booth notes:
data/booth-notes.md, anything the team wrote down ("Dana, ops lead at a 40-clinic group, wants a pilot in Q1"). Match notes to contacts by name and company; list any you cannot match. - The offer for each tier: hot (book a call, pilot), warm (a resource, a demo video), nurture (recording, newsletter).
- Sender name and role for each tier, and the event name and date.
- Voice: if
brand.mdexists, follow its voice section.
Steps
- Clean the list. Remove exact duplicate emails (keep the row with the most notes) and obvious staff or vendor scans (your own domain, exhibitors). Report the counts.
Tier every contact, with the reason in one short phrase:
- Hot: a note with a concrete need, timeline, or ask for next steps, or a top rating in the scan app. Sales calls these.
- Warm: a real conversation or matching title and company, no stated need yet.
- Nurture: scanned for swag, no notes, or outside your audience. When unsure between two tiers, choose the lower one and say why.
- Draft hot emails one by one. Each: subject under 7 words, first line references the actual conversation from the notes, one sentence of value, one clear ask with two time options or a booking link placeholder
<booking link>. Under 120 words. No attachments. - Draft one template each for warm and nurture with merge fields (
{{first_name}},{{company}}). Warm: a useful resource plus a soft ask. Nurture: thanks plus the recording or a resource, no ask. - Set the sequence: hot day 1 email, day 3 call or LinkedIn note, day 7 short bump. Warm day 2, day 9. Nurture day 3, then into the newsletter.
- Write the outputs (below) and show the summary: counts by tier, the top 10 hot leads, and unmatched notes.
- Ask before anything leaves. The user reviews drafts and sends them from their own tools.
Output files
out/follow-up/plan.md: summary, tiers with reasons, sequence, unmatched notes.out/follow-up/hot-emails.md: one draft per hot contact.out/follow-up/templates.md: warm and nurture templates.out/follow-up/import.csv:email,first_name,last_name,company,title,tier,reason,owner,next_step,next_step_date,eventfor a CRM or email tool import.
# Follow-up plan: <Event>, <date>
212 contacts: 18 hot, 64 warm, 121 nurture, 9 removed (duplicates or staff).
## Hot (sales calls first)
| Name | Company | Why hot | Owner | Next step |
| Dana R. | Clinic group (40 sites) | Wants a Q1 pilot | AE 1 | Email day 1, call day 3 |
## Unmatched notes
- "Guy with the red jacket, wants pricing" (no match; ask the team)Guardrails
- Never send emails, connect to an email tool, or add anyone to a list. Drafts and a CSV only.
- Only email people who gave their details at the event or opted in. Mark contacts from regions with stricter consent rules (EU, UK, Canada) in the CSV so the user can check consent before a marketing send.
- Never invent details about a conversation. If notes are thin, the email stays general and says nothing it cannot back up.
- Keep contact data in
data/andout/follow-up/. Show names and companies in chat, not full email lists.
README.md
Attendee follow-up
The day after an event, hand over your badge scans and booth notes. You get every contact sorted into hot, warm, and nurture, a personal email draft for each hot lead, templates for the rest, and a CSV ready to import. Nothing is sent without you.
Who it is for: event marketers, SDRs, and founders who came home with 200 scans and a notebook of half-sentences.
Example
Input: data/attendees.csv (212 rows from the scan app) and data/booth-notes.md:
- Dana, ops lead, clinic group with 40 sites. Wants a Q1 pilot. Asked about SSO.
- Sam from a dental chain, just wanted socks.Output (out/follow-up/plan.md):
| Tier | Count | Who | First touch |
|---|---|---|---|
| Hot | 18 | Concrete need or timeline | Personal email day 1, call day 3 |
| Warm | 64 | Real conversation, no need yet | Resource email day 2 |
| Nurture | 121 | Scanned, no notes | Thanks plus recording day 3 |
And a hot draft:
Subject: Your Q1 pilot across 40 sites Hi Dana, thanks for walking me through how your front desks juggle reminders at 40 locations. You asked about SSO: yes, and I can show it in 20 minutes. Does Tuesday 10am or Wednesday 2pm work?
What it can access
- Reads
data/attendees.csv,data/booth-notes.md, andbrand.md(for voice) if present. - Writes only inside
out/follow-up/. - No scripts, no network, no email sending.
Setup
None. Export your scans as CSV into data/attendees.csv. Works in Claude Code, Codex, and Cursor.
Files
- README.md
- SKILL.md
- kit.json
Did it work?
0worked for them0did not work
Sign in to say whether it worked for you.
Nobody has said yet. Tried it? Tell the next person.
