AI SDR — Lead Qualifier & Demo Booker
Companies pay sales development reps to do exactly this kind of triage — ask the same five questions, decide who’s worth a call. This workflow does that work in a Telegram thread, so your team only spends time on leads that are actually ready.
What it does
A guided five-question conversation that qualifies inbound leads in chat. AI scores each lead — hot, warm, or cold — and routes only the hot ones to your sales team with a booking link. Warm and cold leads get a friendly acknowledgment, so no one feels ignored and your sales team only sees prospects ready to buy.
- Five short questions, about ninety seconds, no form to fill in
- AI rates the lead and explains its reasoning in plain language
- Hot leads get an instant booking link AND your sales channel is notified
- Warm and cold leads get a polite response — zero human time required
What you get
- Lead starts with a single command → walks through five questions → gets a real-time fit assessment
- A hot lead receives a booking link in-chat and your sales team is notified instantly
- Warm and cold leads get a tailored, respectful response — nothing falls through the cracks
What you’ll need
A notification webhook URL
Where your sales team gets pinged for hot leads — a Slack incoming webhook, a Discord webhook, Microsoft Teams, or any service that accepts a JSON POST (Zapier, Make, your own endpoint).
A scheduling link
Your Calendly, Cal.com, SavvyCal, or any URL-based booking tool. The bot drops it in the message that hot leads receive.
Your scoring criteria (optional)
Edit the AI scoring step to match your ideal customer profile — team size, budget thresholds, target verticals. The default rubric is a reasonable starting point.
How it works
1. Lead kicks off the conversation
A simple command (like /demo) starts the flow. The bot welcomes the lead and sets expectations — five quick questions, then an honest fit assessment.
2. Five questions, asked conversationally
Company name, team size, use case, budget, timeline. Each answer is saved so the AI scorer can look at the full picture at the end. Numeric fields are gently validated to keep the data clean.
3. AI scores the lead
A "typing…" indicator appears while AI does its work. The model rates the lead as hot, warm, or cold and writes a short reasoning the lead can see — transparency builds trust, even with a "warm" rating.
4. Smart routing
Hot leads receive a personalized reply with the booking link AND your sales channel is notified with the full scorecard. Warm and cold leads get a polite acknowledgment and a promise of an email follow-up.
Why it works
- Conversational qualification feels less corporate than a form — leads give more honest answers, and AI sees the tone, not just the data.
- Your sales team’s time is reserved for prospects who actually fit your offering.
- Every lead is captured, even the ones you don’t book — so you can follow up over email or sync to a CRM later.
Build it with AI
Generate this workflow from a prompt.
Copy the prompt below and paste it into Eavur’s AI workflow builder. It builds the whole flow for you — then just drop in your API keys and go live.
Build a Telegram lead-qualification bot triggered by the /demo command. - Welcome the lead, then ask five questions one at a time. After each question, await the lead's text answer and save it to a session variable: company_name, team_size, use_case, budget, timeline. - After the last answer, show a typing indicator, then use an "Ask AI" node to score the lead. The AI must make the FIRST WORD of its reply HOT, WARM, or COLD using this rubric: HOT = team size >= 10 AND budget >= $200/month AND timeline is "this month", "next quarter", or "now"; WARM = at least 2 of those criteria; COLD = at most 1. Then 2-3 encouraging sentences of reasoning. Save the result to a session variable called lead_scorecard. - Use a text condition to check whether lead_scorecard contains "HOT". - If it does (HOT): reply with the scorecard and an inline URL button linking to a Calendly booking page, then send an HTTP POST request to a sales notification webhook (e.g. Slack) with the lead's name, username, and all five answers plus the scorecard as JSON. - If it does not (WARM/COLD): reply with a polite acknowledgment that includes the scorecard and promises an email follow-up. Use an OpenAI model for the AI node. I'll paste my Slack webhook URL into the HTTP request node and my Calendly link into the HOT reply afterwards.
More use cases
Voice-Driven Restaurant Order Bot
Customers order dinner by sending a voice note. The bot transcribes, prices the order, takes payment, and notifies the kitchen.
Property Manager — Tenant Support & Rent Bot
Tenants report maintenance issues, pay rent in-chat, and upload documents that are virus-scanned before storage.