How We Evaluate AI-First Business Managers

What This Role Is (and Isn’t)

Where to Find Candidates

Entry Level (L1–L3)

• BPO top performers — look for people who’ve already adopted AI tools on their own, not because they were told to
• Freelance VAs — high-rated Upwork/OnlineJobs.ph talent who want a real career path, not gig work
• Fresh graduates — strong English, hungry, tech-comfortable. Especially business, communications, or hospitality programs in the Philippines and Taiwan
• Customer support teams — people at Zendesk-heavy companies who’ve started automating their own workflows

Mid Level (L4–L5)

• SDRs at SaaS companies — people already doing outbound who want to grow into closing and leadership
• Sales trainers at BPOs — experienced enough to teach but stuck in a system without upward mobility
• Community managers — strong communicators with audience-building skills who want a revenue path

Senior Level (L6+)

• Account executives at staffing/services firms — relationship sellers, not transactional sellers
• Executive assistants to founders — high-trust, high-context operators who already do half of what L8 describes
• Chiefs of staff at startups — people doing L9-level work without the title or compensation

Screening Criteria

Level Placement Guide

Most candidates enter at L1. Lateral entry at higher levels is rare and requires demonstrated equivalent experience.

The L6 filter is real. Closing requires personal qualities — rejection tolerance, comfort with money, real-time improvisation, trust-building speed — that can’t be trained. Screen for these explicitly at L6+ interviews. Many excellent L5s will never be L6s, and that’s fine.

Interview Process

Step 1: Resume Screen (5 min)

• Evidence of progressive responsibility (not lateral moves between equivalent roles)
• AI tool usage — mentioned specifically, not generically
• Strong English writing (cover letter or email sample)
• For L4+: quantified sales or pipeline metrics
• For L6+: named client relationships and revenue numbers

Step 2: Screening Interview (30 min)

Adapt questions to the target level:

L1–L3 (Foundation):

• “Show me how you use AI in your daily work. Not what you could do — what you actually did this week.”
• “Tell me about a time you caught your own mistake before anyone else noticed.”
• “How do you stay organized when you have 15 tasks from 3 different people?”

L4–L5 (Pipeline & Training):

• “Walk me through your outbound process. What’s your response rate and how do you know?”
• “You have a mentee who’s been at L2 for 8 months and isn’t improving. What do you do?”
• “Show me a playbook or SOP you wrote. Why did you write it? Did anyone else use it?”

L6+ (Closing & Leadership):

• “Tell me about a deal you lost after investing months. What happened next — that day, not that week?”
• “A client says your price is 40% higher than the competitor. Walk me through the next 60 seconds.”
• “Describe a relationship with a client that started transactional and became strategic. What changed?”

Step 3: Trial Project (2–4 weeks, paid)

L1–L3: Execute a set of defined tasks with AI tools. Measured on accuracy, speed, AI integration, and communication quality.

L4–L5: Run a mock outbound campaign. Research real prospects, draft personalized outreach, maintain a CRM pipeline. Measured on research depth, personalization quality, and process discipline.

L6+: Handle a real client scenario (with supervision). Run a discovery call, prepare a proposal, present pricing. Measured on relationship quality, deal judgment, and objection handling.

Red Flags

• “I don’t really use AI” — at any level, this is disqualifying. AI fluency is the multiplier for the entire track.
• Can’t quantify anything — if an L4+ candidate can’t tell you their response rates, pipeline value, or close rate, they weren’t really doing the job.
• Blames others for lost deals — at L6+, ownership is everything. “The prospect wasn’t serious” is a red flag. “I misread the urgency” is a green flag.
• No evidence of developing others — from L4 onward, mentorship is mandatory. If they’ve never taught, coached, or developed anyone, they’re missing a core requirement.
• Job-hops without progression — lateral moves between equivalent VA/support roles signal someone collecting experience, not building it.
• Uncomfortable discussing money — for L6+ candidates, probe directly. If they can’t talk about pricing in an interview, they can’t do it with a client.

Compensation Benchmarks

Common Mistakes

• Over-indexing on English fluency at the expense of people skills. Perfect grammar doesn’t mean they can build trust or read a room. Test both.
• Placing someone at L4+ because of years of experience. Years don’t equal levels. Someone with 5 years of VA work is still L1 if they never built pipeline or qualified leads. Place on demonstrated capability, not tenure.
• Skipping the L6 personal qualities screen. Technical sales skills can be assessed on a resume. Rejection tolerance, improvisation, and comfort with money can only be assessed live. Don’t skip it.
• Assuming L8–L9 candidates look like salespeople. The best executive assistants and chiefs of staff don’t present as “sales.” They present as calm, trusted, invisible operators. The closing skills are there — they just stopped calling it sales.