Post-Show Follow-up
SkillDocs & knowledgeDraft post-show email sequences from qualified lead tiers and supplied conversation notes.
Use Post-Show Follow-up in Claude, ChatGPT or Ahel Desktop
Free. Sign in, add Post-Show Follow-up and connect your AI. About a minute.
Also: Claude Code · Cursor · Codex
Then ask your AI: use the Post-Show Follow-up skill
Details
Instructions available. Your AI can read the instructions. Execution depends on the setup they require.
Account requirements not reviewed. Check the skill instructions before use; Ahel provides instructions and does not run this skill.
No other account needed.
Add Ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.
What this skill tells your AI
The instructions your AI receives, as published by lensmorofficial/trade-show-skills in post-show-followup/SKILL.md and read by Ahel’s review.
Generate tiered follow-up email sequences that convert trade show conversations into pipeline — sent within the critical 48-hour window when you're still fresh in their memory.
Why This Matters
Many trade show leads never receive timely, relevant follow-up. Of those that do, many get a generic "Great meeting you!" email that goes nowhere. This skill creates targeted sequences based on how warm the lead actually is.
When this skill triggers:
- Use it in the 24-48 hour window after the show, once leads are tiered or at least roughly segmented
- Use it after
badge-qualifierif you want tier logic grounded in actual booth notes - Do not use it to qualify raw leads from scratch; do that first in
badge-qualifier
Example Requests
- Write follow-up emails for my trade show leads
- 帮我写展后跟进邮件
- Messe-Nachfassung schreiben
- 展示会後フォローアップメール
- emails de seguimiento post-feria
Workflow
Step 1: Understand the Context
Extract from the user's request:
Required:
- Show name (just completed or about to end)
- What they were showcasing / selling
Helpful:
- Lead tiers — does the user already have a system? (e.g., hot/warm/cold, or A/B/C)
- Typical deal cycle — quick transactional vs. 6-month enterprise
- CRM they use (affects formatting and merge tags)
- Any specific conversations they want to reference
If the user just says "help me follow up after MEDICA", generate three sequence templates with visible placeholders for missing facts. Do not invent lead counts, conversation details, or qualified tiers.
Step 2: Define Lead Tiers
Preserve supplied qualified tiers. For raw notes, count confirmed buying involvement, explicit need, and a stated timeline: three signals = Hot, two = Warm, zero or one = Cold. A title or resource request alone does not establish buying involvement. Use badge-qualifier when available for a fuller lead card; this skill can apply the same rules without requiring another installed skill.
Tier 1 — Hot (three confirmed qualification signals)
- Reference a specific action item only when recorded
- Follow-up within 24 hours
Tier 2 — Warm (two confirmed qualification signals)
- Tailor the message to the confirmed need or timing; keep missing signals unknown
- Follow-up within 48 hours
Tier 3 — Cold (zero or one signal, including badge-only contacts)
- Use a light introduction without implying a conversation that was not recorded
- Follow-up within 1 week
If the user qualified leads with badge-qualifier, its Hot / Warm / Cold output maps directly: Hot → Tier 1, Warm → Tier 2, Cold → Tier 3. The lead cards can be pasted in as input.
Do not promote a lead because of title, company size, or a resource request. The timing above is a suggested plan, adjusted to the user's deal cycle and consent to follow up. Never treat badge capture alone as a meaningful conversation.
Step 3: Write the Sequences
Ground every message in the supplied booth notes. Never invent conversation details, customer names, case studies, resource URLs, certifications, quantified results, pricing, or agreed next steps. If an asset or proof point was not provided, use a visible placeholder such as [verified case study] or [resource link] and tell the sender to replace it before sending.
For each tier, create a 2-3 email sequence.
Tier 1 — Hot Lead Sequence
Email 1 (Day 1): Personal recap + specific next step
Subject: [Action item from your conversation] — following up from [Show]
Hi [Name],
[Reference something specific from the conversation — a problem they mentioned, a question they asked, a joke you shared. This is what separates you from the 50 other "great meeting you" emails they'll get.]
[Restate the next step you agreed on and make it concrete — attach the pricing sheet, propose 3 meeting times, send the case study they asked about.]
[One-line CTA]
Email 2 (Day 4): Value-add if no reply
- Don't just "check in" — share something useful (relevant case study, data point, article)
- Reference the show context again briefly
Email 3 (Day 10): Last touch with lower-commitment CTA
- Shorter, more casual
- Offer an alternative next step (async demo, webinar, intro to a colleague)
Primary CTA: book the agreed next step or lock in a concrete meeting time Best asset to send: the exact thing requested in the conversation (pricing sheet, case study, meeting link, technical follow-up)
Tier 2 — Warm Lead Sequence
Email 1 (Day 2): Connect + educate
Subject: [Specific thing they'd care about] — from [Show]
Hi [Name],
[Brief, genuine opening — reference the show experience, not just "we met at..."]
[1-2 sentences about what you do, angled toward THEIR use case based on what you discussed]
[Offer something low-commitment: a relevant resource, a 15-min call, a recorded demo]
Email 2 (Day 7): Different angle
- Come from a different direction — industry insight, customer story, or comparison guide
- Don't repeat Email 1's pitch
Primary CTA: low-commitment meeting or resource review Best asset to send: one relevant proof point, customer example, or short recorded demo
Tier 3 — Cold / Badge Scan Sequence
Email 1 (Day 3-5): Soft intro + resource
Subject: [Industry-relevant hook] — we were at [Show] too
Hi [Name],
[Don't pretend you had a deep conversation if you didn't. "We connected briefly at [Show]" is honest. "It was great chatting with you" when you just scanned their badge is not.]
[Quick value proposition — one sentence]
[Link to a genuinely useful resource — not a sales deck]
Email 2 (Day 14): One more try
- Very short, different hook
- If no engagement, let it go — don't spam
Primary CTA: soft opt-in to receive one useful resource or a short intro call Best asset to send: neutral educational content, not a heavy sales deck
Step 4: Format and Personalization
Mark all personalization fields with [brackets]:
[Name],[Company],[specific detail from conversation][product/feature they asked about],[resource link]
If the user mentions a CRM, use appropriate merge tags:
- HubSpot:
{{contact.firstname}} - Salesforce:
{!Contact.FirstName} - Generic:
[First Name]
Add a Sequence Handoff Summary at the end:
- Tier 1 owner and send window
- Tier 2 batch owner and send window
- Tier 3 nurture owner and send window
- First asset or proof point each tier should receive
Step 5: Timing and Tips
Include a recommended send schedule:
Tier 1: Day 1 → Day 4 → Day 10
Tier 2: Day 2 → Day 7
Tier 3: Day 3-5 → Day 14
Tips:
- Send from the person who actually had the conversation, not a marketing alias
- Use the recipient's timezone for scheduling; test send windows against the team's own results rather than claiming a universal best open-rate hour
- If you collected business cards, photograph them and add to CRM before the flight home
- Don't attach large files — link to them instead
- A/B test subject lines for Tier 3 — this is your largest group, so even a small open rate improvement matters. Suggest two subject line variants and recommend splitting the list 50/50.
- Signature format: Keep it simple — name, title, company, phone. Include a scheduling link (Calendly/HubSpot meetings) so the recipient can book a call without email ping-pong. Skip the logo and social icons in follow-up emails — they scream "mass email." Legal disclaimers: remove them if your industry permits; if you're in pharma, medical devices, financial services, or any other regulated sector, keep required disclaimers and do not truncate them.
- For large lead volumes (100+): recommend processing Tier 1 first (within hours of landing), then batch Tier 2 and 3. Delayed follow-up on hot leads is one of the clearest avoidable conversion risks.
- For company enrichment requested by the user, keep returned company evidence separate from the conversation-based lead tier
Quality Checks
Before delivering results:
- Do not give the same CTA to all tiers; hotter leads should receive a stronger, more concrete next step
- Never imply a detailed conversation for Tier 3 / badge-scan-only leads
- Assets or proof points mentioned in the emails must be plausible and consistent with the user's offer
- Never fabricate conversation details, promised next steps, resources, customers, case studies, or metrics; use a visible placeholder when verified material was not supplied
- The Sequence Handoff Summary should make ownership and timing obvious enough to execute without reinterpretation
Signals
- GitHub stars
- 48
- Forks
- 6
- Last commit
- Oct 2026
Advanced
- Item type
- skill
- Key
post-show-followup- Source
- github.com/lensmorofficial/trade-show-skills
github.com/lensmorofficial/trade-show-skills
Related picks
Skill · builderio
The pick for Hubspotsalesforce-marketing-cloud-automation
Skill · composio-community
The pick for Salesforcesalesforce-connector
Skill · a5c-ai
The pick for Salesforcegws-gmail
Skill · googleworkspace
The pick for Gmailopenakita/skills@gmail-automation
Skill · openakita
The pick for Gmailoutlook_calendar
Skill · win4r
The pick for Outlook