AI-Powered Business-Development Platform Development

AdVentures Americas is seeking an experienced developer to build an AI-powered business-development platform that systematically identifies, qualifies, and initiates contact with local advertisers around every FBO in the combined Private Jet Media Americas and JetSet Media network. The platform should invert the usual approach to national ad sales, which starts with an advertiser and asks which markets to recommend. Instead, it should start with each FBO and ask which local businesses in that market should be advertising there, then move qualified prospects into controlled, automated outreach.
The commercial logic is straightforward: with around 100 combined FBO locations and roughly ten advertiser categories per market, the addressable research task runs to well over a thousand local market exercises, repeated continuously as markets change. Performing that manually would require significant additional headcount. The objective of this build is to make that scale achievable through automation rather than hiring.
Scope of Work
The platform has six components, each performing a defined role and passing information through a shared database and commercial graph.
1. Combined FBO and inventory database
A structured database of the combined FBO footprint, covering location data (airport, FBO, address, coordinates, metro area, commercial catchment), network ownership (Private Jet Media Americas, JetSet Media, or both), inventory (screen count, positions, share of voice, creative duration, pricing, availability), audience data, existing advertisers and category conflicts, and sales assets.
2. Local market research
For each FBO, the system should automatically research the surrounding market against defined criteria (geographic or drive-time catchment, approved advertiser categories, qualification rules, local market characteristics) and surface businesses that plausibly represent good local advertisers — luxury real estate, automotive, jewellery, wealth management, private banking, and similar categories, adapted to what is actually relevant in each local market rather than a fixed national list.
3. Prospect qualification and opportunity scoring
Each business found should be scored against factors including audience fit, geographic fit, product or service value, brand positioning, company scale, marketing activity, growth signals, and private aviation affinity, and classified into tiers (exceptional / strong / possible / reject) so commercial effort concentrates on the strongest prospects.
4. Commercial graph
Rather than a flat list of businesses, the system should model relationships between FBOs, networks, markets, businesses, brands, decision-makers, outreach, campaigns, and sales history, so it understands not just that a company exists but which FBOs it is relevant to, which network has inventory there, and why it represents an opportunity. This should also allow the platform to detect when an apparently local advertiser is actually a regional or national opportunity (present at several FBOs in one region, or twenty-plus locations nationally) and flag it for human review rather than approaching it as a routine local lead.
5. Apollo.io integration
Once a business is qualified, Apollo.io should identify the appropriate decision-maker by category and company scale (for example CMO/Marketing Director/VP Marketing for automotive; Owner/President/Marketing Director for jewellery and watches), enrich contact information, and hand off to outreach.
6. Automated, personalised outreach with human-in-the-loop controls
Pre-approved outreach sequences (initial introduction, follow-up, further follow-up) should exist for each advertiser category, with AI personalisation based on the specific FBO, relevant inventory, business category, and reason for relevance — enough to avoid generic mass email without needing a human to draft each message. Routine, qualified local prospects should be able to enter an approved sequence automatically; major brands, national financial institutions, significant luxury groups, existing agency relationships, and multi-FBO accounts should be escalated for human review before contact.
7. Activity tracking and reporting
The system should maintain a persistent record of business-development activity per prospect — FBO association, network/inventory match, decision-maker identified, outreach status and sequence, response, and outcome — so this does not end up scattered across inboxes and spreadsheets, and should support reporting by FBO, network, market, and advertiser category.
Technical Environment
The platform needs to integrate with Apollo.io, which is already in use for cold email sequencing and AI-assisted personalisation. Candidates should note in their proposal whether Zoho CRM (the current pipeline system) and the existing Apollo→Zoho Zapier sync should serve as the system of record downstream, or whether this platform should maintain its own database as the source of truth with Zoho/Apollo as execution layers.
Deliverables
A working platform covering the six components above; documentation sufficient for AdVentures Americas to operate, maintain, and extend it; and a technical handover covering architecture, data model, and any third-party API dependencies and costs (hosting, AI/API usage, Apollo, and any additional data services).
Ideal Candidate
Experience building AI-assisted data pipelines or sales/marketing automation platforms; comfortable working with LLM APIs for research, qualification, and personalisation tasks; experience with CRM and sales-engagement API integrations (Apollo.io experience a strong plus); able to design a relational or graph-style data model rather than a flat spreadsheet-style database.
What to Include in a Proposal
● Relevant past projects, ideally sales automation, lead qualification, or data-enrichment platforms
● A proposed technical approach and architecture
● A phased build plan, if the full scope is not sensible as one phase
● A cost estimate and the reasoning behind it
● Estimated timeline to a working first version