The Outbound Playbook: Running AI Cold Calling Campaigns That Convert

AI cold calling isn't just about automating what humans already do — it's about doing what wasn't possible before. With AI, you can call 500 qualified leads in the time it takes a human rep to call 20. But volume without strategy produces noise, not results.

This playbook covers the complete setup: list selection, script writing, compliance, call flow design, and the feedback loops that separate effective campaigns from wasted ones.

Step 1 — Define a precise target list

The quality of your lead list determines everything else. Broad lists produce low conversion and high unsubscribe rates. Focused lists with clear intent signals outperform every time.

• Use CRM filters or export tools to pull contacts by industry, company size, or last activity date
• Prioritise warm leads — contacts who visited your site, opened a past email, or engaged on social
• For net new outbound, build lists from data sources with verified phone numbers (LinkedIn Sales Navigator, Clay, Apollo)
• Remove disconnected numbers and known DNC registrants before uploading
• Aim for lists of 100–500 contacts for your first test campaign — tight enough to iterate fast

Step 2 — Write a call script that works

AI call scripts are different from traditional scripts. The AI can handle branching logic, objections, and natural back-and-forth — but you need to define the skeleton. Here's the structure that converts:

• Opening (5 seconds): Name, company, one-line context — why you're calling, why now
• Permission ask (5 seconds): "Is now a bad time?" or "Can I take 30 seconds?"
• Value hook (15 seconds): One specific outcome your prospect cares about — not features
• Discovery question (open): Something about their current situation that sets up your offer
• CTA (10 seconds): One ask — a calendar link, a yes/no, or a transfer to a human

The mistake most teams make is treating the AI call like a pitch deck. The best scripts are short, curious, and focused on a single next step. Less than 60 seconds if they engage.

Step 3 — Configure call timing and frequency

When you call matters as much as what you say. Our data across SEA markets shows clear patterns:

• Tuesday–Thursday outperform Monday and Friday for decision-maker availability
• Late morning (10–11am) and post-lunch (2–4pm) show highest answer rates
• Avoid calling during public holidays — SEA has many, and they vary by country
• Maximum 2 attempts per lead before resting for 7 days
• Space attempts at least 48 hours apart

Step 4 — Define your objection paths

Good campaigns anticipate the most common objections and give the AI graceful paths through them. You don't need to handle every edge case — focus on the top 3-4:

• "Not interested" → Acknowledge, thank them, ask if there's a better time later in the quarter
• "We already have a solution" → Validate, ask what they're using, explore if gaps exist
• "Email me instead" → Agree, confirm their email address while on the call, log it
• "Who is this again?" → Repeat company name and give a clearer one-line context

Step 5 — Measure what matters

Most teams focus on call volume. The metrics that actually indicate campaign health are answer rate, engagement rate (conversations lasting 30s+), and meeting conversion rate. Use Soara's analytics dashboard to track these by day, by list segment, and by script variant.

Compliance in SEA markets

Outbound calling regulations vary by country. Always check local rules before launching. Key points for common SEA markets:

• Malaysia: PDPA requires a lawful basis for processing contact data; include opt-out in every call
• Singapore: PDPC guidelines apply; DO NOT registry exists and must be respected
• Indonesia: UU PDP (Personal Data Protection Law) applies to B2C outreach
• Thailand: PDPA compliance required; consent records should be maintained