Business Answers International (BAINTL)

🚦 State of Play - 2026-01-26 Campaign Ready for Launch

  • Mood: 🟢 Full Speed Ahead
  • Phase: Phase 3 - Campaign Launch
  • Blocker: Dan needs to add his sender account to Instantly
  • Next Critical Action: Import 188 Plastics Extrusion leads + activate campaign
  • Decision Point:DECIDED: Instantly.ai for email campaigns (not LOXO native)
  • Target Launch: Tuesday/Wednesday (Jan 27-28) - 9-11am CT

📊 Project Overview

Business Model: Plastics industry recruiting agency Database Size: ~10,000 legacy contacts (fresh Bullhorn → Loxo migration) Project Goal: Replace manual offshore data enrichment with AI automation + launch automated email campaigns for hiring managers Success Metrics: 90% data completion, <5% bounce rate, 15%+ response rate within 90 days


📝 Activity Log - UPDATED

2026-01-26 - Weekly Recap Call: Platform & Taxonomy Alignment

Attendees: Alex, Aaron, Dan

Key Decisions:

  • Global Status: Confirmed migration to native Loxo Global Status; custom “Hierarchy 23” field is now obsolete.
  • Workflow Pruning: Cleaned up Loxo progression activities to reduce noise (e.g., disabled “Note” and “Voicemail” triggers for stage moves).
  • Marketing Strategy: Active Clients will receive less frequent, strategic emails, while Prospects receive more active outreach.
  • Reactivation: Added “No Activity” status to trigger automated reactivation campaigns for stalled leads.
  • Data Hygiene: Identified company entity duplication issues (internet scraping vs. manual entry) and planned a cleanup pass with Kelsey.

Campaign Adjustments:

  • Removed active clients (King Plastic, Pexco, Commerlin) from the initial Plastics Extrusion trial batch.
  • Aaron to rewrite the third sequence email to be more relationship-focused rather than transactional.

2026-02-01 - Outreach Routing Logic (Instantly)

Action: Defined the automated routing logic for Instantly.ai campaigns. Created the Outreach Routing SOP.

Strategic Insight: Implemented an “Owner-Routing” pattern with a Shared Inbox fallback. This ensures personalized outreach from Aaron/Dan while maintaining a safety net for unassigned leads.

2026-02-01 - Deal Pipeline Mapping

Action: Documented Aaron and Dan’s personal Deal Pipelines and win probabilities. Created the Deal Pipeline Map.

Strategic Insight: Defined the 50% probability threshold for Job Orders as the handoff point between Sales and Delivery.

2026-02-01 - Workflow Automation Mapping

Action: Mapped Loxo activity-to-stage triggers to clarify automatic candidate movement. Created the Workflow Automation Map.

Strategic Insight: Identified the “Outbound” catch-all risk and recommended leveraging Global Status for better visibility into engagement vs. prospecting.

2026-02-01 - Global Status Migration

Decision: All automation triggers will now leverage the standard Global Status field instead of the custom “Hierarchy 23” field.

Rationale: Simplifies maintenance and aligns with Loxo standard architecture. Ensures better compatibility with native Loxo features and external n8n workflows.

2026-01-26 - Weekly Recap Call: Campaign Launch Brief

Attendees: Alex, Aaron, Dan

Campaign Lists Status:

  • Plastics Extrusion: ~188 contacts (READY TO LAUNCH)
  • Sheet/Rod/Tube Distribution: 200-400 contacts (next campaign)
  • Film & Sheet (Plastics News): ~500 NEW contacts incoming from Kelsey - need industry categorization

Email Sender Accounts:

  • ✅ Aaron connected his Gmail to Instantly
  • ⏳ Dan needs to add his account (was unavailable during call)
  • ❌ Alex’s account excluded from campaigns
  • Emails will alternate between Aaron & Dan for deliverability

Automation Logic (Loxo → n8n → Instantly):

  • Rule-based assignment: Industry + Hiring Function + Global Status = campaign
  • Owner field drives dynamic signature/from address
  • Kelsey bulk-adds contacts - need special handling for ‘from’ field
  • Aaron/Dan share response handling (ownership doesn’t matter for replies)

Campaign Reuse Strategy:

  • Same contacts can be re-emailed in 2-3 months with different MPC candidate
  • “Not now, keep me posted” responses → update status → fall back into future campaigns
  • Create new campaign with same criteria, different message each time

Deliverability Settings:

  • ~20% bad emails expected in lists
  • Instantly auto-pauses if too many bounces (built-in fail-safe)
  • Bounce webhook → n8n → Loxo update (NOT BUILT YET - next week)
  • 100/100 deliverability score on test
  • No tracking pixels, plain text, inconsistent send patterns

Optimal Send Schedule:

  • Best days: Monday AM, Tuesday, Thursday
  • Avoid: Fridays
  • Best hours: 9-11am
  • Target launch: Tuesday or Wednesday this week (Jan 27-28)

Industry Classification Complexity:

  • Companies often span multiple industries
  • Blow molders → categorize as Packaging
  • Aaron writing classification rules
  • Contacts can have multiple industry tags

Future Roadmap:

  • AI chatbot for Aaron/Dan to query campaigns from phone
  • Bright Data integration for lead enrichment (Aaron needs account)
  • Handoff documentation for business continuity

Instantly Campaign Created:

  • Campaign ID: 24dd9a95-0578-4564-a617-cb5a81f13958
  • Name: BAINTL - Plastics Extrusion MPC Outreach
  • Sequence: 3 emails (Day 0, Day 4, Day 9)
  • Settings: 30/day limit, 10min gaps, 9-5 CT, no tracking, stop on reply
  • Status: Draft (awaiting lead import + activation)

Technical Debt

  • Company entity duplication (internet scraping vs. manual entry)
  • Bounce webhook not built (Instantly → n8n → Loxo)
  • AI Agent rules still pending from Aaron

Campaign Ready

First Instantly campaign created with Aaron’s MPC email style. Ready for 188 lead import and Tuesday/Wednesday launch.


2026-01-24 - Instantly.ai Decision Finalized + Weekend Setup Plan

Decision: Going with Instantly.ai for all email campaigns.

Weekend Plan (Jan 25-26):

  • Set up entire Instantly.ai environment
  • Configure domains and warm-up
  • Build n8n integration workflows

Integration Architecture:

  • Instantly.ai MCP Server → n8n workflow
  • Loxon8nInstantly.ai programmatic communication
  • Full automation layer between all three platforms

Target Launch: Monday or Tuesday (Jan 27-28) for first campaign

Platform Decision Made

After evaluating LOXO native campaigns vs. Instantly.ai, the decision is final: Instantly.ai provides the deliverability, automation, and flexibility needed for this project.


2026-01-19 - Status Workflow Refinement

Context: Review of conversation with Aaron regarding status complexity. Decision: Implemented “Suffix Strategy” to distinguish System vs. Manual statuses. Changes:

  • Renamed automation-heavy statuses to include [Auto] suffix (Enrich, Prospect, Responded).
  • Defined “The Big 3” for manual entry: Outreach (Cold), Nurture (Warm), Active (Current).
  • Updated Status SOP with “Manual Entry Cheat Sheet”.
  • Created CRM Taxonomy Insights with deep research on industry best practices.
  • 2026-01-22: Pivoted from Loxo Native to Instantly.ai for outreach due to Loxo’s sender-locking and mandatory tracking limitations.
  • Created Architecture Proposal evaluating Loxo vs. Instantly.ai stack.

2026-01-13 - Project Reset Call: AI Enrichment Near Complete, Email Campaigns Target Next Week

Attendees: Alex, Aaron, Dan

Key Updates:

Project Status After Break:

  • Alex was out 4-5 weeks, now back and ready to execute
  • Timeline: ~1.5 weeks behind schedule but recoverable
  • Goal: Email campaigns ready by next week (week of 2026-01-20)

AI Enrichment Progress:

  • Status: Almost complete, just needs final testing on additional contacts
  • Confidence scoring implemented: 70% threshold - data below 70% confidence is NOT pushed to Loxo
  • Email discovery:
    • Searches LinkedIn and multiple databases for correct email
    • Fail-safe: If no email found anywhere, will guess pattern (similar to Loxo native behavior)
    • Discovery: Loxo native enrichment also guesses email patterns when can’t find real email
  • Pattern validation: Improved since Polymer Shapes test - now does broader research before making pattern assumptions
  • Next step: Once enrichment verified, proceed to email campaign setup

Database Status:

  • Current total: ~11,000 contacts
  • Recent adds: ~190 plastics extrusion contacts (core industry, mostly sales hiring function + some CEOs/presidents)
  • Import velocity: 100-200 contacts every couple days (~30/day via Kelsey)
  • Extrusion list ready for testing: 190 contacts, majority sales function, core industry with strong candidate pipeline

LOXO Platform Limitations (Confirmed):

  • No native automation support: LOXO lacks intuitive automation capabilities beyond basic pipeline triggers
  • Pipeline automations not feasible inside LOXO: Must build externally via n8n for status updates, workflow triggers, etc.
  • Positive: LOXO has decent API documentation for building external integrations
  • Note: Some automations Aaron mentioned (e.g., “if X then move to Y pipeline”) cannot be done natively in LOXO

External Automations Plan:

  • All workflow automations (status updates based on campaign actions, pipeline movements) will run outside LOXO via n8n
  • Scalable and maintainable approach
  • Alex can demo how these work in future calls
  • Not a problem, just more time-consuming than expected

New Feature Proposed: AI Candidate Finder Agent:

  • Functionality: Voice-activated agent accessible from phone/anywhere
  • Usage: “Hey, find me 100 people in [industry] at [company] with [criteria] and add them to database today”
  • Cost: ~$0.20-0.30 per lead added (includes enrichment)
  • Benefit: Could eventually replace Kelsey and offshore team
  • Decision: Aaron approved - “Let’s do it”
  • Concerns noted: Aaron suspects Kelsey isn’t working claimed hours; decent job but until replacement ready, keeping them

Email Campaign Infrastructure:

  • Subdomain setup: Scott set up mail.baintl.com (~2 weeks ago)
  • Next step: Alex to contact Christy at LOXO (or tech support) to configure domain for sending
  • Expected timeline: Email campaigns ready to deploy next week
  • Target list: 190 extrusion contacts (core industry where BAINTL has deep candidate bench)
  • Campaign approach: Start conservative - Aaron wants careful control since it’s core industry

Status Field Cleanup Required:

Universal Status Decision (2026-01-13):

  • Decision: Use the standard Global Status field as the universal status for both hiring partners (Type = Contact) AND candidates (Type = Candidate)
  • Rationale: Simplifies workflow - same status progression applies to both types with context-appropriate interpretation
  • Implementation: Use existing hiring manager statuses for all people in database
  • No separate candidate status needed: All contacts can be both candidate AND hiring partner (very common in plastics distribution)

Status Simplification:

  • Merge redundant statuses: “Active”, “Current Client”, “Customer” → all become just “Active”
  • Reasoning: Once job order is open and actively working, they’re a client. Distinctions were redundant
  • Note: Some old statuses may have existed pre-Alex and weren’t cleaned up to avoid retroactive issues

Mass Database Updates Coming:

  • Before launching email campaigns, Alex needs to perform mass status/type updates
  • Ensure right people are in correct statuses for workflow triggers
  • Critical: Status updates must happen as contacts move through campaign (e.g., if they respond, opt out, etc.)
  • Status update automations will run OUTSIDE Loxo (another external n8n automation)

Edge Case Discussion - Branch Managers:

  • Scenario: Many branch managers are BOTH candidates AND hiring managers
  • Complication: Some work at active client companies
  • Problem: Don’t want to send candidate recruitment emails to people at client companies, BUT still want them as hiring manager prospects AND potential candidates for future
  • Attempted solutions discussed:
    1. Create exclusion list by company name (but only blocks hiring manager emails too)
    2. Create two separate contacts (rejected - too messy)
    3. Add custom checkbox field to manually exclude from candidate campaigns (manual, not ideal)
  • Aaron’s key point: In plastics distribution, can’t accidentally send 50+ branch managers at biggest client a “hey, want to talk about a job?” email
  • Current thinking: May need to manually review campaign lists before sending to remove active client employees
  • Campaign size: Only 200-500 max per send, so manual review of client names is feasible

LOXO Sentiment:

  • Alex diplomatically expressed concerns about LOXO’s limitations
  • Aaron confirmed locked into 1-2 year contract, not switching again
  • Alex committed to making it work: “We’ll make it work… I’m gonna make sure we can produce some more revenue for you guys”
  • Acknowledged LOXO feels “early in development” but has roadmap items coming

Loxo Self-Updating CRM Agent:

  • Currently running on database
  • Concern: Updates data but unclear WHAT gets updated
  • Risk: Could overwrite good data (from resumes) with bad data (guessed info)
  • No documentation: LOXO provides zero transparency on backend logic
  • Alex recommendation: May turn this feature off due to lack of control/visibility
  • Aaron’s preference: If candidate submitted resume with email/phone, use THAT data - not LOXO’s guessed version

Hiring Manager vs Candidate Distinctions:

  • Terminology confirmed: “Contact” = Hiring Manager in LOXO type field
  • Reality: Almost ALL contacts could be candidates too (branch managers, VPs, etc.)
  • Practical rule: If checked as “Contact”, they’re hiring manager. If only “Candidate”, they’re candidate-only
  • Status determines campaign inclusion: Active status = in campaigns; other statuses = excluded
  • Both types common: Many extrusion branch managers are prime candidates BAINTL places, but also hire salespeople

Three Prerequisites for Email Campaign Launch:

  1. ✅ AI enrichment working correctly (nearly done)
  2. ⏳ Mass status cleanup in database (in progress - Alex doing this week)
  3. ⏳ Email automations configured for campaign responses (status updates, tracking, etc.)

Action Items Assigned:

  • Aaron: Send Alex the “rules” for AI agent (industry + hiring function parameters/definitions)
  • Aaron/Dan: Continue pumping contacts into database (priority: 100-200/couple days)
  • Aaron/Dan: Decide on fallback sender preference (didn’t finalize but acknowledged need)
  • Aaron/Dan: Think about workflow pain points in LOXO - send to Alex anytime “holy shit this takes too long” moments occur
  • Alex: Email Christy at LOXO (CC Aaron/Dan) about domain configuration for mail.baintl.com
  • Alex: Create candidate status values + SOP document (mirroring hiring manager statuses)
  • Alex: Finish AI enrichment testing (couple more contacts)
  • Alex: Mass status updates in database this week
  • Alex: Set up email campaign automations outside LOXO
  • Alex: Continue building AI candidate finder agent
  • Alex: Investigate LOXO email campaign rule capabilities (can we exclude by company list?)

Timeline:

  • This week (2026-01-13 to 2026-01-19): Finish enrichment, mass updates, campaign setup
  • Next week (2026-01-20+): Deploy first email campaigns to extrusion list (190 contacts)
  • Future: Scale to candidate campaigns once hiring manager campaigns proven

LOXO Automation Limitations

Confirmed: LOXO cannot handle the automation workflows Aaron originally envisioned. Most automation (status updates, pipeline triggers, campaign response handling) must be built externally via n8n. This is maintainable and scalable but requires more development time than anticipated.

AI Enrichment Nearly Ready

Confidence scoring (70% threshold) prevents bad data from entering database. Pattern-guessing only used as last resort, similar to LOXO’s native behavior. Ready for final validation before campaign deployment.

Tasks

Action items updated in BAINTL - Action Items.

2025-12-23 - Call Recap: Database Build-Out, Self-Updating CRM Results, and January Launch Plan

Attendees: Alex, Aaron

Key Updates / Decisions:

  • Self-Updating CRM is running: Enabled on 2025-12-18.
    • Observed: ~187 updates since enablement.
    • Limitation: Loxo logs are not great for auditing who was updated and what fields changed.
    • Action: we need a practical validation pass (spot-check recent updates) + clarification from Loxo on scope/cadence.
  • Contacts need volume before campaigns:
    • Current priority is building out hiring manager contacts (Loxo “Contact” type) to 100+ per industry + hiring function segment.
    • Kelsey is doing Excel-based collection/import; Aaron plans to prioritize imports after this week.
  • Accuracy > volume (for now):
    • BAINTL prefers quality contact data even if lists are smaller per segment.
    • Scale later by expanding into additional industries (not only bigger lists).
  • Terminology standard (important):
    • “Contact” means hiring manager.
    • “Candidate” means candidate.
    • Use “People” only when speaking about both together.
    • “Contractor” type is not used.
  • Industry taxonomy guidance:
    • Prefer sub-industry selections rather than only broad industries (except for mega-company targets like Dow/DuPont).
  • Campaign expectations:
    • Start with small, tightly targeted batches in January 2026.
    • Aaron expects strong response rates on the first touch (based on proven manual MPC process), then drop-off across follow-ups.
  • Email deliverability infrastructure:
    • Plan to create a dedicated sending subdomain outreach.baintl.com before scaling volume (coordinate with Scott, BAINTL IT).
  • Workflow + pipeline automation (nice-to-have):
    • Review Loxo workflow “progression triggers” (some may be misconfigured).
    • Nice-to have: when a candidate hits First-time interview in a job pipeline, auto-create a deal in the BAI placement pipeline.
      • Not critical path; only pursue if it’s a short-effort win.
  • Candidate statuses are messy:
    • Candidate global statuses are legacy (Bullhorn-era). Resetting is acceptable later as long as it doesn’t disrupt active interviewing/placement pipeline.

Next Steps:

  • Aaron to send AI enrichment rules/examples (industry + hiring function mapping).
  • Aaron/Dan to import backlog hiring manager contacts (Kelsey’s Excel lists) to create sufficient campaign volume.
  • Alex to validate Self-Updating CRM outcomes and ask Loxo for scope/cadence + better audit trail guidance.
  • Alex to propose subdomain + deliverability setup plan (and coordinate with Scott as needed).

Tasks

Action items are tracked in BAINTL - Action Items.

2025-12-19 - Call Recap: LOXO Enrichment Limitations + First-Pass AI Enrichment Plan

Attendees: Alex, Aaron

Key Updates / Decisions:

  • LOXO limitation confirmed (again): No way to automatically enrich brand-new incoming candidates/contacts at point of ingestion.
    • LOXO enrichment appears to be tied to pipeline / manual “Fetch Now” / bulk fetch rather than true “auto-enrich on create”.
  • Aaron enabled 2 LOXO enrichment features today (costs in credits):
    • Self-Updating CRM: ~0.02 credits per update
    • Work Email: ~0.2 credits
    • Open question: LOXO provides limited transparency on run cadence/frequency; we need to monitor credit burn and observe whether it backfills/updates older records automatically.
  • Proposed enrichment strategy (new incoming leads):
    • First pass: Custom AI enrichment (OpenAI-driven) using lead-data sources optimized for enrichment (e.g., Apollo-style datasets) + optional email verification passes to reduce bounce risk.
    • Ongoing/nurture: LOXO self-updating + manual fetch as needed once the record exists and/or relationships progress.
  • Email conflict handling:
    • Plan to add/maintain a secondary email field as a fallback if custom enrichment and LOXO produce different work emails.
    • Aaron will test and determine which enrichment path is more accurate and iterate workflow accordingly.
  • Credits / cost note: LOXO credits are bundled with seats (not clearly priced as a per-credit $ figure), so we’ll compare effective cost and accuracy vs custom enrichment.
  • Lead generation note: Aaron mentioned testing LOXO’s in-house sourcing as another way to add leads beyond LinkedIn.

Next Steps:

  • Alex to start building/testing the custom enrichment workflow and share an update later tonight or tomorrow morning.
  • Monitor LOXO credit usage over the next few days to infer self-updating cadence/backfill behavior.

Unknown enrichment cadence

The LOXO “self-updating CRM” feature does not clearly disclose frequency/run conditions. We should treat it as unbounded until verified by observed credit usage and/or LOXO support confirmation.


2025-12-16 - Call Recap: Enrichment Agents, Pilot Batch, Email Pattern Validation

Attendees: Alex, Aaron

Key Updates / Decisions:

  • Dependencies (Aaron):
    • Create OpenAI API account (business email), add 10 credit, add Alex as admin
    • Send 2-3 example MPC emails + hiring function/industry mapping examples
  • LOXO enrichment agent settings need clarification (avoid surprises):
    • Contact Information Finding Agent settings show Personal Email = ON and Work Email = OFF
    • Self-Updating CRM Agent may auto-run and could burn credits at scale
    • Alex to confirm exact behavior + guardrails/caps with LOXO support/Artemis before enabling
  • Pilot dataset for enrichment/testing:
    • In LOXO, ~550 / ~11,000 records are Type = Contact (likely Bullhorn hiring managers)
    • Use older/pre-LOXO contacts for initial batches (target ~200-300); avoid contacts created in the last ~30 days (active relationships / data likely correct)
  • Import workflow constraint:
    • LOXO Chrome extension can set Type (Contact vs Candidate) + tags, but cannot set Industry/Hiring Function at import time
    • Workaround: import into lists/tags, then mass-update fields inside LOXO
  • Early enrichment test result (Bright Data + Claude Sonnet):
    • Produced strong company/title/industry/hiring-function signals quickly
    • Observed work-email pattern mismatch for at least one company (model returned first.last@… while pattern appears first initial + last@…)
    • Action: add email-pattern detection/validation step + conservative confidence threshold before writing work emails
  • Campaign learnings (directional):
    • Dan: 2 job-related blasts to candidates (44 each) → ~4.5% response rate
    • Aaron: small, highly targeted MPC to hiring managers previously saw ~30% response rate (expect lower at scale)

MPC email examples received

Added 3 examples to the MPC Copy Library section below.

Credit burn risk

Do not enable any “auto-updating” enrichment agents on the full database until we confirm how often they run and how to cap usage.

Linked doc

Action items live in BAINTL - Action Items.

2025-12-08 - Weekly Sync Call: Enrichment Strategy & Timeline

Attendees: Alex, Aaron, Dan

Key Decisions:

  1. Weekly Sync Calls Established: Regular check-ins every week to maintain momentum toward January 1st launch target

  2. Enrichment Approach:

    • Test LOXO native enrichment first (validate email/phone accuracy)
    • Consider Bright Data for cross-reference/verification if LOXO quality insufficient
    • AI classification using GPT-4o (5.1) for hiring function/industry taxonomy mapping
    • Cost estimate: ~$0.20-0.30 per enrichment (worst case), likely lower
    • Pilot approach: Test with 10 → 50 → 100 contacts before full scale
  3. OpenAI Account Required:

    • Aaron to create OpenAI API account at openai.com (separate from personal ChatGPT)
    • Add minimum $5 credit
    • Add Alex as admin for integration work
  4. Campaign Platform - LOXO Native with Tracking Limitation:

    • CONFIRMED: LOXO cannot disable click/open tracking (platform limitation)
    • Decision: Proceed with LOXO native campaigns, accept tracking limitation
    • Alternative Considered: MailChimp integration for more flexibility
    • Rationale: Keep everything in one platform for simplicity, tracking pixels are industry standard
    • Still need to verify unsubscribe compliance requirements for bulk sends (100+ contacts)
  5. Classification Rules Needed:

    • Aaron to provide examples for AI classification (e.g., “VP of Sales = Sales function”)
    • Need to handle nuances like “Branch Manager” classification
    • AI will map job titles to BAINTL’s specific taxonomy (not generic categories)
  6. Timeline Confirmed:

    • Week 1 (Current): Enrichment testing, external sourcing evaluation
    • Week 2: Campaign development and testing
    • Week 3: Scaling enrichment (bulk processing)
    • Week 4: Launch (targeting first week of January, worst case second week)
  7. Database Strategy:

    • Kelsey (offshore researcher) to build hiring manager contact lists
    • Import via Excel with industry/hiring function pre-filled where possible
    • Available LOXO credits: 1,350/month (can enrich 6,750 contacts at 0.2 credits each)
    • Need to consider cleaning up old contacts (2011 Bullhorn migration data with no activity)
    • Current database: 10,800 contacts, likely 20%+ are outdated/deceased
  8. Workflow Review:

    • Alex to step back and review all workflows before proceeding
    • Create visual mermaid diagrams for hiring manager/candidate/deal pipelines
    • Ensure workflows are efficient and scalable before automation
    • Low priority: Consider automating “First-time interview → Auto-create deal in BAI Placement Pipeline”
  9. Deal Pipeline Automation (Low Priority):

    • Interest in automating: First-time interview status → Auto-create deal in BAI Placement Pipeline
    • Currently manual process: Add deal, enter name/value, link to job, link to contact
    • Not critical path, but would save time as volume increases

Test Results Shared:

  • Dan sent bulk email to 44 branch managers about relocation job
  • Only 2 responses (4.5% response rate)
  • Likely killed by relocation mention in email
  • Next test: Send another batch without mentioning location to compare

Chrome Extension Question:

  • Aaron uses LOXO Chrome extension to quickly add LinkedIn profiles (one-by-one)
  • Current limitation: Can’t add industry/hiring function during import
  • Must go back into LOXO to add those fields manually
  • Alex to ask LOXO support if extension can be customized

Action Items Created:

  • Aaron: Create OpenAI account, provide classification examples, provide MPC email samples
  • Alex: Test enrichment options, ask LOXO about Chrome extension, test campaign sender behavior, review workflows
  • Decision needed: LOXO vs MailChimp for campaigns

Project Momentum

Team aligned on January launch target. All infrastructure complete, now moving into execution phase with enrichment testing and campaign development.

Cost Management

OpenAI API pay-as-you-go model. No prepayment option, but tiered pricing ($2.50 per 1M tokens). GPT-4o (5.1) is token-efficient and excellent at rule adherence, making it ideal for taxonomy classification.


2025-12-04 - Kristi Confirms: Tracking Cannot Be Disabled ⚠️

Decision Made: Build campaigns externally via n8n + Gmail API

Kristi confirmed that Loxo product team CANNOT disable click/open tracking in Outreach campaigns. This is a platform limitation, not a configuration option.

UPDATE (2025-12-05)

After further consideration, decision was made to accept the tracking limitation and proceed with LOXO native campaigns. The integrated platform benefits outweigh tracking pixel concerns, and tracking pixels are industry standard in modern email platforms.

Original Impact Analysis: Proceeding with Loxo-native campaigns would risk deliverability during reputation-building phase due to aggressive tracking pixel flagging by modern email clients (Gmail, Outlook).

Original Path Considered: Build email campaigns externally using n8n + Gmail/Outlook API for full control over tracking settings. This would allow:

  • Clean emails without tracking pixels during initial outreach
  • Better deliverability during 2-4 week reputation building period
  • Ability to enable tracking later once domain reputation is established

Trade-offs Evaluated:

  • Pros (External): Full control, better deliverability, cleaner emails
  • Cons (External): External system to maintain, slightly more complex setup, analytics split between n8n and Loxo

Final Decision (2025-12-05): Proceed with LOXO native campaigns, accepting tracking limitation as industry standard.

2025-12-03 - Loxo Meeting with Kristi (Critical Findings)

Attendees: Alex, Kristi (Loxo CSM)

Critical Discovery #1 - No Dynamic Sender: ❌ Loxo Outreach campaigns are sender-locked to creator. “Owner” field is ignored. Workaround: Duplicate campaigns for Aaron/Dan with manual contact segmentation.

Critical Discovery #2 - Click/Open Tracking Cannot Be Disabled: ⚠️ Kristi confirmed tracking is mandatory (currently). Escalated to Loxo team for formal confirmation.

Discovery #3 - LoxoSource May Be Sufficient: Aaron hasn’t explored LoxoSource (Loxo’s native 1M+ profile database). Enrichment costs:

  • Work email: 0.2 credits
  • Personal email: 1 credit
  • Phone: 1 credit
  • Boolean search: Free
  • Natural language search: 0.04 credits

Current Credits: 1,300+ available (~6,000-7,000 enrichments possible)

Recommendation: Test LoxoSource on 50 contacts before investing in Apollo/Clay/BrightData.

Other Notes:

  • Subdomain decision: Correctly recommended AGAINST (only needed for mass untargeted campaigns)
  • Inaccurate data reporting: Aaron to use “black flag” feature for bad phone numbers (credit refund)
  • Phone system side project: Call recording + AI transcription interest (Twilio/RingCentral), but porting = 3 weeks downtime

2025-12-02 - DNS Records Fixed

Issue: 11 DNS errors + UCEPROTECTL3 blacklist warning in MXToolbox Root Cause:

  • DKIM Error: Cryptographic key placed on root domain (@) instead of google._domainkey selector
  • DMARC/SPF Collision: _dmarc record contained SPF data instead of DMARC policy
  • Blacklist: False alarm - GoDaddy shared hosting IP (208.109.16.142) neighborhood blocking. Irrelevant since Google Workspace handles mail.

Resolution:

  • Moved DKIM to google._domainkey with 2048-bit key
  • Replaced _dmarc with monitoring policy: v=DMARC1; p=none;
  • Retained SPF: v=spf1 include:_spf.google.com ~all

Status: ✅ DNS correctly configured. Awaiting 48hr propagation.

Next Step: Send test email to personal Gmail → View “Show Original” → Verify PASS on SPF/DKIM/DMARC

Strategy: Starting with p=none (monitoring) for 2-4 weeks before tightening to p=reject. Protects against blocking legitimate Loxo emails during reputation building.

2025-11-28 - Field Structure Redesign

Issue Identified: Original “Prospect Status” field mixed relationship history with workflow stages (“Current Client”, “Previous Client”, “YES”, “NO”).

Solution Implemented: Renamed to “Hiring Manager Status” with workflow values:

  1. Enrich - Missing data; AI enrichment needed
  2. Prospect - Data complete; ready for review
  3. Outreach - Active in MPC email campaign
  4. Responded - Replied to email
  5. Qualified - Confirmed hiring need
  6. Active - Job order open
  7. Customer - Placement made
  8. Nurture - Long-term relationship building
  9. Inactive - No engagement 12+ months
  10. Do Not Contact - Unsubscribed/bounced

Dual-Status Architecture: Created TWO separate status fields (contacts can be both):

  • Hiring Manager Status (Phase 1 focus - business development)
  • Candidate Status (Phase 2 - job seekers)

Rationale: Prevents workflow collision when same person is both hiring manager AND candidate.

2025-11-27 - Taxonomy Review

Industry Field (Approved):

  • Core Plastics: Resin Distribution, Extrusion, Injection Molding, Thermoforming, Packaging, Sheet/Rod/Tube Distribution
  • Adjacent Markets: Chemical Distribution, Metals, Industrial Distribution

Hiring Function Field (Pending Approval): Recommended removal of “ALL (Talent Acquisition/HR)” catch-all category to force data specificity.

Revised List:

  • Branch Management
  • Finance/Accounting
  • Human Resources
  • Mfg/Engineering/Operations
  • Purchasing/Procurement
  • Sales

Rationale: Catch-all categories prevent personalized campaigns and encourage lazy data entry.

2025-11-25 - Project Kickoff

  • Gained Loxo admin access
  • Explored Loxo API capabilities
  • Audited current CRM fields and statuses
  • Confirmed fresh Bullhorn → Loxo migration (~10k contacts)

🛠 Architecture & Tech Stack

Current Stack

  • CRM: Loxo (Recruiting ATS/CRM)
  • Email: Google Workspace (baintl.com)
  • Hosting: GoDaddy (website only)
  • Enrichment (Pending Decision):
    • Option A: LoxoSource (native, cheaper, untested quality)
    • Option B: Apollo/Clay/BrightData (proven, $0.05-0.15 per contact)

Proposed Automation Architecture

LOXO-Native Campaign Architecture (CONFIRMED APPROACH)

graph TD
    A[Contact enters Enrich status] --> B[LoxoSource/Bright Data API]
    B --> C[AI Classification GPT-4o 5.1]
    C --> D[Map to BAINTL Taxonomy]
    D --> E[Update Loxo Record]
    E --> F[Move to Prospect Status]
    F --> G[Human Approval]
    G --> H[Move to Outreach Status]
    H --> I[LOXO Campaign Trigger with Tracking]
    I --> J[Monitor Replies via LOXO]
    J --> K[Update Status in LOXO]

Tracking Limitation Accepted

LOXO confirmed click/open tracking CANNOT be disabled (platform limitation). Decision made to proceed with LOXO native campaigns as integrated platform benefits outweigh tracking pixel concerns. Tracking pixels are industry standard in most email platforms.


📧 Email Campaign Strategy

Campaign 1: Conservative MPC (Core Plastics)

  • Target: Resin Distribution, Extrusion, Injection Molding, Thermoforming, Packaging
  • Trigger: Hiring Manager Status = “Outreach” + Core Plastics Industry
  • Cadence:
    • Email 1 (Day 0): Personalized intro referencing specific niche
    • Email 2 (Day 7): Value-add follow-up
    • Email 3: TBD
  • Example: “I noticed XYZ Company distributes engineering resins for automotive…”

Campaign 2: Aggressive MPC (Adjacent Markets)

  • Target: Chemical Distribution, Metals, Industrial Distribution
  • Trigger: Hiring Manager Status = “Outreach” + Adjacent Industry
  • Cadence:
    • Email 1 (Day 0): Broader value prop
    • Email 2 (Day 4): Social proof
    • Email 3 (Day 8): Direct CTA
    • Email 4: TBD

Campaign 3: Customer Maintenance

  • Target: Hiring Manager Status = “Customer”
  • Frequency: Every 60 days
  • Message: “Hope [Placement Name] is working out. Any new hiring needs?”

MPC Copy Library

MPC Email Example #1

Target Persona: Branch Manager / General Manager / Profit Center Manager Industry: Plastics sheet/rod/tube distribution Hiring Function: Sales

Hi John,
 
I have a really great outside sales candidate in Dallas that has spent the past 10 years in sheet distribution. He’s currently exploring new opportunities and has extensive experience selling sheet products to OEM’s, machine shops, sign manufacturing, and other segments.
 
Does this sound like someone that might be of interest to you?

MPC Email Example #2

Target Persona: Regional or District Sales Manager Industry: Plastics resin distribution Hiring Function: Sales

Hi Mike,
 
I just completely a recruiting search in the Minneapolis area for a client, and interviewed a couple of excellent candidates that are currently in the engineered resins distribution industry that are quietly exploring new opportunities.
 
One of them is selling a full line of products mostly into injection molding customer accounts, and the other one is doing a lot of sales to medical device OEM’s and their suppliers.
 
Do you have a need for anyone like that right now?

MPC Email Example #3

Target Persona: VP of Manufacturing Industry: Plastics color & additive compounding Hiring Function: Mfg/Eng/Ops

Hi Jane,
 
I just got off the phone with a Plant Manager I know who is with another color compounder that manufactures color concentrate and masterbatch products. He’s open to relocation and wants to explore opportunities with other color or additive compounders.
 
He has extensive experience with compound extrusion production and also has a mechanical engineering background, so he’s a highly hands on plant manager. Not an office guy type.
 
Do any of your facilities need a plant manager, or are there other hands on manufacturing management positions you have that could use someone like this?

🤖 AI Enrichment Strategy

The Workflow

  1. Trigger: Contact enters “Enrich” status (missing email/industry/function)
  2. Enrichment: AI calls vendor API to find email, phone, LinkedIn
  3. Classification: LLM (GPT-4) reads job title + company description
  4. Mapping: AI maps to BAINTL’s specific taxonomies (not generic terms)
  5. Update: Push data back to Loxo
  6. Advance: Move to “Prospect” status
  7. Safety Valve: Aaron/Dan manually approve before “Outreach”

The LLM Classification Prompt

You are a data classifier for Business Answers International, a plastics recruiting firm.

I will provide a Job Title and Company Description.

You must categorize the contact into ONE Industry and ONE Hiring Function from the allowed lists below.

Allowed Industries: [Resin Distribution, Extrusion, Injection Molding...]
Allowed Functions: [Sales, Purchasing, Operations...]

Return ONLY the exact string from the allowed lists.

Value Prop: Standard enrichment outputs “Plastics” or “Manufacturing”. LLM maps to BAINTL’s SPECIFIC buckets (“Resin Distribution”, “Extrusion”).

Cost Comparison:

  • Manual (Offshore Team): 5 min/contact, $2-3 cost
  • AI Automation: 30 sec/contact, $0.10 cost
  • ROI: 10x faster, 20x cheaper

⏳ Timeline & Milestones

Timeline update (2025-12-23)

We are still in Phase 2 build/testing. The schedule below reflects the updated timeline.

Week 1 (Dec 8-15, 2025)

  • DNS records fixed ✅
  • Loxo onboarding call completed ✅
  • Test DNS propagation (send test email) ✅
  • LOXO tracking limitation confirmed ✅
  • Weekly sync call established ✅

Week 2 (Dec 16-22, 2025)

  • Aaron creates OpenAI API account ✅ 2025-12-18
  • Self-Updating CRM enabled/observed ✅ 2025-12-18
  • Test LOXO native enrichment (10-50 contacts)
  • Test Bright Data enrichment (cross-reference verification)

Week 3 (Dec 23-29, 2025 - Current)

  • Build AI classification automation using GPT-4o (5.1)
  • Build n8n enrichment workflow
  • Create campaign copy (Conservative, Aggressive, Customer)
  • Test campaigns in LOXO native platform
  • Start/continue importing hiring-manager contact lists (Excel backlog) to reach 100+ per industry/function segment

Week 4 (Dec 30, 2025 - Jan 5, 2026)

  • Complete pilot enrichment + classification on 25-100 contacts
  • Finalize enrichment approach (vendor decision)
  • Confirm sending subdomain plan: outreach.baintl.com

Week 5 (Jan 6-12, 2026)

  • Database cleanup (remove old/inactive contacts)
  • Bulk enrichment phase (scale to thousands)
  • Manual review of “Prospect” bucket

Week 6 (Jan 13-19, 2026)

  • Launch first campaigns (200+ hiring managers)
  • Monitor deliverability metrics
  • Adjust DMARC policy from p=quarantine to p=reject if metrics are clean

Target Go-Live: Week of January 13, 2026 (stretch: week of January 6)


🛳 Current Blockers

✅ RESOLVED

  1. Click/Open Tracking: LOXO confirmed tracking CANNOT be disabled (platform limitation - CONFIRMED 2025-12-08).

    • Decision: Accept limitation, proceed with LOXO native campaigns
    • Rationale: Integrated platform benefits + tracking pixels are industry standard
    • Status: ✅ Resolved - Moving forward with native LOXO campaigns
  2. DNS Configuration: All records fixed, propagation complete, domain reputation excellent

    • Status: ✅ Complete
  3. Status Structure: SOP created, all 10 statuses implemented in LOXO

    • Status: ✅ Complete

Phase 2 Priorities (Current Focus)

  1. Data Enrichment Testing: Test LoxoSource on sample batch

    • Action: Enrich 10 → 50 → 100 contacts to validate quality before scaling
    • Decision Criteria: If 80%+ accurate, use LoxoSource (saves $500-1,500)
  2. Campaign Build-Out: Create first campaign sequences in Loxo native platform

    • Conservative MPC: 3-email sequence for core plastics
    • Aggressive MPC: 4-email sequence for adjacent markets
    • Customer Maintenance: Quarterly check-in template
  3. Automation Preparation: Prep database for automated workflows

    • AI Classification: Build LLM mapping for industry/function taxonomy using GPT-4o (5.1)
    • Webhook Setup: Configure triggers for status transitions
    • Testing: Validate full enrichment → campaign pipeline

Low Priority

  1. Manual Campaign Segmentation: LOXO limitation requires Aaron/Dan to duplicate campaigns and manually add their contacts (no auto-assignment by Owner field).
    • Mitigation: SOP to be created for campaign duplication workflow

📋 Awaiting Approvals

From Aaron/Dan

  1. OpenAI Setup:

    • Create OpenAI API account and add Alex as admin
  2. Classification Rules:

    • Provide hiring function classification examples (VP Sales, Branch Manager, etc.)
  3. Email Campaigns:

    • Approve Conservative MPC strategy (core plastics)
    • Approve Aggressive MPC strategy (adjacent markets)
    • Approve Customer Maintenance cadence (60 days)
    • Provide 2-3 example MPC emails for tone/style
  4. AI Enrichment:

    • Approve approach after pilot testing (10 → 50 → 100 contacts)
    • Confirm manual approval preference before “Outreach” status
  5. Database Strategy:

    • Confirm database cleanup approach (old contacts from 2011 migration)
    • Confirm readiness for bulk enrichment

💰 Budget & Credits

Loxo Credits (Current)

  • Available: 1,300+ credits
  • Capacity: ~6,000-7,000 enrichments
  • Burn Rate: Not a concern (plenty available)

External Enrichment (If Needed)

  • Apollo: ~$0.10-0.15 per contact
  • Clay: ~$0.08-0.12 per contact
  • BrightData: ~$0.05-0.10 per contact
  • 10k contacts: 1,500 one-time cost

OpenAI API (AI Classification)

  • Model: GPT-4o (5.1)
  • Pricing: ~$2.50 per 1M tokens (pay-as-you-go)
  • Estimated Cost: $0.20-0.30 per contact enrichment (worst case), likely lower
  • Setup: Aaron to create account, add $5 minimum credit, add Alex as admin

🎯 Technical Notes

DNS Configuration (Current)

  • Domain: baintl.com
  • Registrar/Host: GoDaddy
  • Mail Provider: Google Workspace
  • A Record: 208.109.16.142 (GoDaddy shared hosting - website only)
  • MX Records: Google Workspace servers
  • SPF: v=spf1 include:_spf.google.com ~all
  • DKIM: google._domainkey with 2048-bit key ✅
  • DMARC: v=DMARC1; p=none; (monitoring mode) ✅
  • Blacklist: UCEPROTECTL3 (GoDaddy IP) - irrelevant for email ℹ️

Loxo API Capabilities

  • RESTful API with OAuth 2.0
  • Webhook support for real-time updates
  • Custom field mapping available
  • Rate limits: TBD (to be confirmed with Kristi)

Fresh Migration Advantage

Context: Just migrated from Bullhorn (~10k contacts) Window: 60 days before bad habits form Opportunities:

  • No legacy reporting dependencies → Can delete old fields
  • No team muscle memory → Can enforce new standards from day one
  • No active campaigns → Can make structural changes without risk
  • Offshore team can be trained on NEW system only (no unlearning)

🎓 Key Learnings & Decisions

Strategic Insights

  1. Reputation ≠ Domain Age: Email reputation is built on current authentication (DMARC/DKIM), not historical longevity. Without valid auth, a classic domain is indistinguishable from spam.

  2. Monitoring Before Enforcement: Starting with DMARC p=none allows verification that legitimate emails (including Loxo’s) aren’t being blocked before tightening to p=reject.

  3. Taxonomy Precision = Campaign Personalization: Mapping to BAINTL’s specific industry buckets (“Resin Distribution” vs generic “Plastics”) enables hyper-personalized email openers.

  4. LLM Classification > Human Judgment (at scale): AI can map job titles + company descriptions to specific taxonomies faster and cheaper than offshore teams, with consistent accuracy.

Platform Limitations (Loxo)

  1. No Dynamic Sender: Campaign sender is locked to creator, not contact Owner. Requires manual campaign duplication per sender.
  2. Mandatory Tracking (CONFIRMED): Click/open tracking cannot be disabled - this is a permanent platform limitation. Decision: Accepted as industry standard.
  3. No Auto-Assignment: Cannot automatically add contacts to campaigns based on Owner field.

Risk Mitigation

  1. Vendor Testing: Test LoxoSource before committing to external enrichment vendors (Bright Data/Apollo/Clay).
  2. Manual Approval Gate: Human review before contacts enter “Outreach” prevents bulk mistakes.
  3. Phased Rollout: Test with 10 → 50 → 100 contacts before scaling to thousands.
  4. Database Cleanup: Filter old/inactive contacts before enrichment to avoid wasted costs.

Project Reminders

  • Fresh Migration Window: 60 days to enforce standards before inertia sets in
  • DMARC Strategy: Starting with p=none for 2-4 weeks, then tighten to p=reject
  • Manual Approval Gate: Review “Prospect” bucket before advancing to “Outreach”
  • A/B Testing: Once campaigns launch, test subject lines across duplicate campaigns
  • Black Flag Feature: Use in Loxo to report bad data (credit refund)
  • Daily Review: Check for vendor responses, monitor enrichment progress, flag new blockers

📞 Contacts

Primary

  • Aaron: [Contact details TBD]
  • Dan: [Contact details TBD]

Support

  • Kristi: Loxo Customer Success Manager
    • Escalated click/open tracking request to product team
    • Demonstrated LoxoSource capabilities
    • Contact: [Via Loxo platform]

Project Lead

  • Alex (You): Primary automation engineer and project manager

🔗 Resources & References

Internal

Note

The Loxo resource file is directly tied to BAINTL’s CRM migration, automation, and campaign setup. See cross-linked sections for project-specific learnings.

New to Plastics Recruiting?

Read the Plastics Industry Knowledge Base first - it covers key terminology, industry segments, business models, and BAINTL-specific insights that will help you understand the project context.

Quick Access Tools

External


📌 Next Actions (Priority Order)

  1. [Aaron] Create OpenAI account, add $5 credit, add Alex as admin
  2. [Aaron] Provide hiring function classification rules/examples
  3. [Aaron/Dan] Provide 2-3 example MPC emails for tone/style reference
  4. [Alex] Test LOXO native enrichment on 10-50 sample contacts
  5. [Alex] Test Bright Data enrichment for verification (Alex’s budget)
  6. [Alex] Ask LOXO support: Can Chrome extension include industry/hiring function fields?
  7. [Alex] Test LOXO campaign sender behavior (creator vs “from” field)
  8. [Alex] Review all workflows and create visual mermaid diagrams
  9. [All] Build campaigns in chosen platform
  10. [All] Build AI enrichment automation using chosen vendor