Plai.io / Claude Performance Advertiser, Data Specialist, & AI / CRM Marketer
Worldwide
Read this part first We aren't your typical lead generation company. We are an AI-native marketing agency running paid lead generation for franchise groups, salon suite operators, and home service contractors. We do not run one-off ad campaigns. We build and operate a documented, repeatable 12-step lead generation system - ads to offers to landing pages to pre-qualification to 24/7 AI booking to nurture to scored pipelines to review and testimonial programs to reporting - and we run it across dozens of local accounts at once. That means this is not a "Facebook Ads expert" role and not a "GHL automation guy" role. It's both, plus the wiring between them. If you have only ever done one half of that, you are not a fit for this posting, and we'd rather you not apply than waste your connects. Everything below is drawn from our internal operations handbook. If a tool or feature is listed, we actually use it in production, and you will be asked about it. What you will own Provisioning new client sub-accounts end-to-end (target: at most 1 business day per account, sign-off gated) Building and launching Meta + Google campaigns through Plai, not through Ads Manager Wiring every lead source into GoHighLevel so it lands as an Opportunity in the correct pipeline's first stage, tagged by source, the moment it arrives - no orphan contacts, ever Standing up and monitoring the validation stack, lead scoring, and speed-to-lead tracking Working an exception queue daily (threshold breaches, missed-SLA tags, red dashboard cells) plus a rotating weekly deep pass on every account Logging every change - one variable at a time - in the tracking sheet and experiment registry Application-Specific Skillsets Rate yourself honestly. We will verify with a screen share. Required - you must be production-fluent GoHighLevel (expert level, agency view) Sub-account creation from snapshot; custom fields and field folders; tag taxonomy; pipelines and stages; workflows with If/Else branching, wait steps, webhook actions, and custom-code/HTTP steps; Surveys with conditional logic; Funnels and Websites incl. tracking-code fields; Chat Widget; Calendars incl. round-robin; Phone Numbers and Trust Center; Email Services; Conversations; native dashboards; snapshot maintenance. You should know the difference between funnel-level and account-level tracking code placement without looking it up. Plai.ai (or provable ability to learn a campaign-management layer fast) Agency workspace and sub-account structure; connecting Meta ad accounts and Google customer IDs inside the sub-account; campaign templates and cross-account cloning with location tokens; creative library management with hook-coded naming; AI budget optimization and recommendation review; the Simplified Targeting tab as a client-facing surface. If you have not used Plai specifically, tell us the closest equivalent you have run at portfolio scale and how quickly you ramped. Meta Ads (wiring and diagnostics, not daily buying) Business Manager structure and asset assignment; domain verification; Pixel + Conversions API with event deduplication and Event Match Quality; Special Ad Category identification; exclusion audiences; customer-list custom audiences and lookalikes; Instant Forms and the Lead Ads Testing Tool; Advantage+ Creative enhancement settings; account restriction appeals. Google Ads (wiring and diagnostics) MCC linking; conversion actions (Primary vs. Secondary); Enhanced Conversions and Enhanced Conversions for Leads; GCLID capture and offline conversion import; negative and brand-exclusion lists; disabling auto-applied recommendations; Lead Form assets delivered by webhook; Local Services Ads including lead disputes. Google Tag Manager, GA4, and pixel deployment One GTM container per client as the tag home; head/body snippet placement in both Webflow and GHL; page-view and form-submit triggers on thank-you steps; cross-domain measurement when the funnel and main site sit on different domains; GA4 key events; verification with Tag Assistant, Pixel Helper, and Test Events. Mailgun Dedicated sending subdomain setup; SPF, DKIM, DMARC records and alignment; MX and tracking CNAME; connecting the domain to a GHL sub-account; warm-up scheduling; monitoring bounce (under 2%) and complaint (under 0.1%) rates. You should be able to read a message header and confirm all three auth results. A2P 10DLC registration through GHL Trust Center Brand and campaign submission; matching legal name to IRS/EIN records; opt-in evidence and consent screenshots; compliant sample messages; privacy policy SMS clause; rejection triage and resubmission. Strongly preferred n8n (self-hosted) - scheduled jobs, webhooks, API calls, Slack alerting. We do not use Zapier; if Zapier is your only automation tool, say so up front. Canva - Brand Kits, multi-ratio master templates (1:1, 4:5, 9:16), and Bulk Create for localized variant sets across franchise locations. Looker Studio - with Windsor.ai or Porter Metrics connectors for Meta and Google. Webflow - site settings, custom code injection, publishing (you will not be designing; you will be installing and troubleshooting). Apollo API for inline per-lead enrichment inside a GHL workflow. Claude and/or Manus for research, ad copy, batch creative, and reporting narratives. Visitor identification tools - RB2B, Customers.ai, Pearl Diver, or Factors.ai. Note: for consumer/local clients like salon suites and home services, Customers.ai is our default, and knowing why B2C identification differs from B2B is a real differentiator here. Feature-Specific Skillsets These are the specific builds you will be doing. Tell us which of these you have personally shipped, and for whom. Lead capture and routing Mapping a Meta Instant Form question-by-question to GHL contact fields, including creating dropdown/radio custom fields whose option values match the Meta answers exactly Re-checking mappings after every form edit (and knowing why unmapped questions fail silently) Google Lead Form assets via Inbound Webhook trigger with payload field mapping Chat widget install and routing, including SMS/text-chat mode that captures a phone number up front Missed-chat and missed-call auto-reply branches Honeypot fields and spam-blocker workflows on every capture surface Qualification and scoring Multi-step quiz funnels with conditional logic, hidden UTM/fbclid/gclid fields, and one question per slide Answer-to-points scoring written to a lead_score field plus tier tags, with tier-driven routing branches A three-component scoring model (fit / engagement / validity) and back-testing it quarterly against actual stage conversion Feeding qualified-lead events back to Meta CAPI and Google offline conversion import so bidding optimizes to quality, not raw volume Validation stack Email and phone validation workflows writing valid/risky/invalid tags A four-rule post-validation routing matrix: both pass, both fail, email fail + phone pass, email pass + phone fail - each with different client notification and kick-starter campaign behavior Single-field "update your email / update your phone" micro-forms that re-run validation and lift suppression on pass Speed-to-lead Instant SMS + email at 0 minutes, gated by validation tags Call Connect style bridging that dials the rep first with a whisper and a press-1 gate before connecting the lead (so voicemail pickups don't count as connections) Timestamp fields for lead creation and first human-grade response, a nightly delta job computing elapsed seconds, and met/missed tagging against tiered SLA targets Slack notifications via Incoming Webhook, native - not through a third-party connector AI voice and booking Voice AI agent persona prompts, knowledge base loading, escalation rules Calendar booking actions connected to specific GHL calendars with buffer rules and round-robin routing Assigning the agent as overflow (ring human 20 seconds, then routes to agent) vs. after-hours vs. primary Transcript review as the actual training loop - 10+ test calls, then weekly review for the first month Campaign operations Learning-phase math: knowing when an ad set has enough weekly conversion volume to optimize on qualified events vs. raw leads, and consolidating rather than splitting below that floor Structure sizing from monthly spend (campaigns / ad sets / creatives per ad set) Scaling in +20-30% increments every 3-4 days; target CPA/ROAS moves in 10-15% steps Creative refresh triggered by frequency and CTR decay thresholds, not by calendar habit Portfolio experimentation: one variable tested across 5-10 comparable local accounts simultaneously, pooled to significance, winner cloned system-wide Compliance TCPA-aware send windows and stricter state rules (e.g., Florida's tighter window and contact caps) STOP/HELP handling, touch caps, and an opt-out/complaint-rate kill switch Call recording disclosure in all-party consent states Substantiated-claims review on creative Reporting GHL native dashboards: pipeline stage counts, leads by source tag, score tier distribution, median first-response time, appointments and show rate, call volume, CPQL trend Looker Studio builds with franchisee/campaign/date filters, comparison time periods, creative tables with hook rate and hold rate, and video funnel metrics (3-second plays, 25/50/75/100%, ThruPlays) Reconciling platform-reported numbers against CRM truth and investigating any discrepancy over 10% Vertical Experience - Home Service Contractors and Salon Suites This is weighted heavily. We would take a strong operator with real experience in these two categories over a technically stronger one without it. Home service contractors (roofing, HVAC, plumbing, electrical, remodeling, exterior/court and surface installation, landscaping, restoration) Radius and service-area targeting per location, and negative geo discipline Local Services Ads: onboarding, badge requirements, budget management, and the weekly lead dispute window Google Business Profile as a lead source alongside paid Seasonality and weather-driven demand swings, and how to pace budget through them Call-heavy lead flow - most of these leads phone in, so tracking numbers, whisper messages, missed-call text back, and call attribution matter more than form volume Estimate/quote-request funnels and the realistic timeline from lead to signed job Multi-location or multi-territory contractor groups where one template gets cloned across markets Salon suites, booth rental, and beauty-professional recruitment This is a tenant/operator acquisition motion, not a consumer-services motion. The buyer is a licensed stylist, barber, esthetician, nail tech, or lash artist deciding to leave a commission salon and rent their own suite. Marketing to that person is nothing like marketing haircuts. Qualification criteria that actually predict a signed lease: existing client book, license status, current employment situation, budget for weekly rent, move-in timeline Tour-booking as the primary conversion event, and the tour-to-lease follow-up sequence Beauty-professional interest targeting, competitor-salon and professional-group audience proxies, and where first-party seeds come from when a new location has no customer list yet Awareness of employment/housing-adjacent language risk in ad copy - suite rental creative can trip Special Ad Category review if written carelessly Franchise context: multiple franchisee-owned locations under one brand, each with its own territory, budget, and maturity level, all reporting up to a franchisor If you have run either of these categories at multi-location scale, lead with that in your first line. How we work The system is documented. You will be handed a 40+ section operations handbook with numbered SOPs and Loom walkthroughs. We expect you to follow it, and to propose changes to it in writing rather than freelancing around it. One variable at a time. No stacked changes, no untracked edits, no "I tweaked a few things." Humans approve, AI recommends. AI never executes against an ad account. Launches, pauses, and budget moves over 20% require a named human approval. Buy before build. No new tools without a written gap statement. Everything gets logged. Provisioning sheet, tracking sheet, experiment registry. If it isn't logged, it didn't happen. Exceptions first, rotation second. Nobody checks every account every day. Automated monitoring watches everything; you work the breaches and rotate deep passes. Screening questions - answer these in your proposal Applications without answers to these will not be reviewed. Short, specific answers beat long ones. Walk me through provisioning a brand-new GHL sub-account for a local service business from snapshot to first live lead. Name the order of operations and which step you'd start on day one and why. An A2P 10DLC campaign submission just got rejected. Name the three most common causes and how you'd triage each. A Meta Instant Form has been running for a week. The client says leads are arriving but the "what's your timeline" answer is blank on every contact record. What happened, and what do you check first? An ad set has been running 4 days with a CPL 22% above target. What do you do? A client wants to optimize campaigns toward qualified leads instead of raw leads. They spend $2,000/month. What's your answer, and why? Describe a home service contractor OR salon suite / booth rental account you've personally run. Client type, monthly spend, what you were measured on, and what the result was. Which of these have you personally configured in production: GTM container + GA4 + Meta Pixel with CAPI dedupe / Mailgun dedicated subdomain with full DNS auth / a Voice AI agent with live calendar booking / a webhook-triggered workflow calling an external enrichment API. List only the ones you've actually shipped. What's your experience with Plai specifically? If none, what's the closest platform and how many accounts did you manage in it at once? How to apply Start your proposal with the word SUITE so we know you read the posting. Include: Your answers to the screening questions above A screen-recorded Loom (3-5 minutes max) walking through one GHL workflow you built, with your camera off and your screen on - we want to see real work, not a portfolio PDF Your available hours per week and your timezone
- More than 30 hrs/weekHourly
- 6+ monthsDuration
- ExpertExperience Level
$13.00
-
$45.00
Hourly- Remote Job
- Ongoing projectProject Type
Skills and Expertise
Activity on this job
- Proposals:15 to 20
- Last viewed by client:2 weeks ago
- Interviewing:10
- Invites sent:14
- Unanswered invites:2
About the client
- United StatesKansas City2:32 PM
- $19K total spent12 hires, 1 active
- 13 hours
- EducationSmall company (2-9 people)
Explore similar jobs on Upwork
How it works
Create your free profileHighlight your skills and experience, show your portfolio, and set your ideal pay rate.
Work the way you wantApply for jobs, create easy-to-by projects, or access exclusive opportunities that come to you.
Get paid securelyFrom contract to payment, we help you work safely and get paid securely.
About Upwork
- 4.9/5(Average rating of clients by professionals)
- G2 2021#1 freelance platform
- 49,000+Signed contract every week
- $2.3BFreelancers earned on Upwork in 2020
Find the best freelance jobs
Growing your career is as easy as creating a free profile and finding work like this that fits your skills.
Trusted by