AI-Powered Prospecting, Email, SMS & CRM Marketing Platform

AI-Powered Prospecting, Email, SMS & CRM Marketing Platform Project Overview I am looking for an experienced developer or development team to build a centralised AI-enhanced B2B prospecting, marketing and sales outreach platform. The platform will be used across multiple businesses and websites. The objective is to create one system that can: Identify target customer segments → build comprehensive prospect databases → research organisations and contacts → create campaigns → personalise outreach → send email/SMS campaigns → manage responses → update the CRM → measure sales outcomes → improve future campaigns using AI. This is not simply an email marketing application or web scraper. It should become the central prospect intelligence and outbound marketing platform across multiple businesses. 1. Multi-Business Architecture The system must support multiple businesses/websites from one central platform. Each business should have its own: Brand Website Products/services Target markets Email domains/accounts SMS identity/numbers Campaigns Templates Staff CRM pipelines Reporting Prospect lists and customer segments must be assignable to: One business Multiple businesses The same organisation should not need to exist multiple times simply because it is relevant to several businesses. 2. Master Organisation & Contact Database The platform should maintain one central prospect database. The hierarchy should broadly be: Business → Market → Segment → Prospect List → Organisation → Contact → Campaign → Conversation → Opportunity → Customer Organisation records should support information such as: Organisation name Website Industry Customer segment Address/location State Number of locations Approximate employee/company size Phone General email Relevant businesses Source URLs Research notes AI qualification information Last verification date Each organisation can contain multiple contacts, including: Name Position/title Email Mobile/phone Decision-making role Public professional profile/source Source URL Verification status Confidence score Last verified date Lists should effectively operate as organised/dynamic groups of records within this master database. 3. AI Customer Segment & List Idea Generator For each business, AI should be able to analyse what the business sells and suggest potential prospect lists. For example: Med Uniforms Potential lists could include: GP practices Dental practices Veterinary clinics Physiotherapists Aged-care providers NDIS providers Radiology clinics Pathology providers Medical specialists Allied-health organisations Champion Trophies & Awards Potential lists could include: Basketball clubs Football clubs Soccer clubs Cricket clubs Schools Dance schools Swimming clubs Martial arts organisations Community organisations Corporate organisations AI should continually be able to propose new market segments and campaign opportunities. 4. Comprehensive List Building The objective is to build complete market lists, not simply lists containing large organisations. Small organisations should NOT automatically be excluded. A small organisation today may become a significant customer later. For example, if building an Australian basketball club database, the objective should be to identify as many legitimate basketball clubs across Australia as reasonably possible rather than only the largest clubs. The system should support list-building by: Industry Organisation type Geography State City/region Business category Association membership Other defined characteristics AI/web agents should search multiple public internet sources to identify organisations. 5. AI Internet Research & Data Enrichment Once a prospect segment is selected, AI should research the internet and build/enrich the database. Research should attempt to identify: Organisation Website Locations General contact information Relevant decision makers Names Job titles Email addresses Business phone numbers Publicly available professional contact information Organisation size Number of locations Other commercially relevant information The system should record: Source Source URL Date discovered Date verified Confidence level The research system should be capable of revisiting records and updating outdated information. 6. Contact Verification & Database Hygiene Before outreach, contact information should be validated where possible. Required functionality includes: Email verification Domain verification Duplicate detection Contact-role validation Catch-all detection Bounce-risk assessment Invalid-email removal Data freshness monitoring Automatic re-verification Global suppression checking The system should minimise poor-quality data entering campaigns. 7. AI Lead / Opportunity Scoring AI should analyse each prospect and provide an opportunity or relevance score. Importantly, this score should NOT automatically exclude smaller organisations. It should help prioritise prospects and determine appropriate messaging. Factors could include: Organisation type Size Number of locations Potential product requirements Relevant services Existing supplier information where publicly available Growth signals Buying triggers Previous interaction Previous purchases Campaign engagement Users should be able to sort/filter by score without removing lower-scoring organisations from the database. 8. AI Campaign Generator Users should be able to enter a simple campaign concept. Example: Target basketball clubs before end-of-season presentations and promote trophies, medals and participation awards. AI should then develop the campaign, including: Campaign strategy Target audience Offer Messaging angle Subject lines Email copy SMS copy Follow-up messages Calls to action Recommended sequence Recommended timing Potential landing-page concept Response-handling rules Campaigns should be editable and require appropriate approval before launch. 9. Individual AI Personalisation The platform should be capable of generating individualised outreach rather than simply inserting: Hi First Name AI should use relevant organisation/contact research to tailor messaging. Personalisation should be based on commercially relevant, publicly available information and should avoid inappropriate or overly personal information. Each recipient can therefore receive slightly different messaging while remaining within the approved campaign framework. 10. Multi-Step & Multi-Channel Campaigns The system should initially support: Email SMS Staff follow-up tasks The architecture should allow additional channels to be added later. Campaigns should support sequences such as: Day 1: Initial personalised email Day 4: Follow-up email Day 8: SMS where legally appropriate Day 12: Create staff call/follow-up task Day 18: Final follow-up Sequences must stop or change automatically when recipients reply, unsubscribe, bounce, convert or otherwise meet predefined conditions. 11. Email Sending & Deliverability Infrastructure The platform needs professional outbound email management. Required functionality should include: Multiple sending domains Multiple mailboxes SPF/DKIM/DMARC support Domain/mailbox configuration Sending limits Sending schedules Mailbox rotation where appropriate Email reputation monitoring Bounce monitoring Spam complaint monitoring Automatic pausing when problems occur Domain/mailbox health dashboard Controlled ramp-up/warm-up capability where appropriate The architecture should protect primary business domains from damage caused by outbound prospecting. 12. SMS Infrastructure The system should integrate with an SMS provider/API and support: Campaign SMS Individualised SMS Two-way SMS Incoming responses Conversation history Unsubscribe handling CRM integration Staff assignment Email and SMS conversations should ideally appear within the same customer/prospect record. 13. AI Response Classification Incoming email and SMS responses should automatically be analysed. AI should classify responses such as: Interested Quote requested Question Call requested Not currently interested Contact later Wrong person Referral to another person Existing supplier Unsubscribe Negative response Out of office Bounce The appropriate workflow should then be triggered automatically. 14. AI-Assisted Response Drafting When a prospect responds, AI should be able to prepare a recommended reply for staff. The staff interface should show: Incoming message Relevant prospect/customer information Previous conversation AI recommended response Recommended next action Staff should be able to: Approve Edit Send Reassign Create task Create/update CRM opportunity Low-risk responses could potentially become automated later, but the system should initially support human approval. 15. CRM Integration The CRM should become the commercial source of truth. The platform should track: Prospect → Contacted → Engaged → Qualified → Opportunity → Quote → Won/Lost → Customer → Repeat Customer All outreach and responses should be recorded against the appropriate organisation/contact. The system should prevent inappropriate prospecting of: Existing customers Active opportunities Current quotations Suppressed contacts unless specifically authorised. 16. Staff Assignment & Workflow Responses and opportunities should automatically be routed to the appropriate: Business Department Staff member Salesperson Assignment rules should be configurable. Staff should have dashboards showing: New replies Leads requiring action Quote requests Follow-ups due AI-drafted responses awaiting approval Opportunities Tasks Overdue actions 17. Cross-Business Intelligence Because the platform supports multiple businesses, it should identify cross-selling opportunities. For example, an organisation purchasing uniforms may potentially require: Name badges Promotional merchandise Awards/trophies Other products supplied by another business within the portfolio The platform should identify these opportunities without unnecessarily duplicating organisations or contact records. 18. Campaign Analytics Campaign reporting should include: Emails sent Delivered Bounces Opens where technically available Replies Positive replies Negative replies Unsubscribes Qualified leads Opportunities Quotes Sales Revenue Gross profit where CRM/accounting data permits The system should focus particularly on meaningful commercial outcomes rather than simply open rates. 19. AI Campaign Learning Campaign results should feed back into the AI system. AI should analyse performance by: Business Industry Customer segment Organisation size Geography Decision-maker role Campaign Offer Subject line Message Sequence Channel Timing The AI should then recommend: Better segments Better campaign ideas Better messaging New tests Underperforming campaigns to stop Successful campaigns to expand The objective is for the system to become more effective as campaign and sales data accumulates. 20. Compliance & Suppression Management Compliance must be designed into the platform from the beginning, particularly for Australian email and SMS requirements. The system should support: Unsubscribe management SMS opt-outs Do-not-contact records Global suppression lists Business-specific suppression Consent/status records where required Contact/source records Identification requirements Automated suppression before campaigns Audit history If someone opts out, the system must ensure they are not accidentally rediscovered by the research system and contacted again. 21. AI Autonomy / Approval Controls Different activities should have configurable levels of AI autonomy. For example: Level 1 — AI recommends; human performs action Level 2 — AI prepares; human approves Level 3 — AI automatically executes approved rules Level 4 — AI optimises campaigns within predefined limits The initial implementation should favour human approval for important external communications while allowing automation to increase over time. 22. Administration & Permissions The system should include appropriate user permissions. Possible roles include: Administrator Business manager Marketing manager Salesperson Staff member Read-only/reporting user Access should be controllable by business and function. 23. Search, Filtering & List Management The prospect database needs powerful search/filtering. Users should be able to filter by combinations such as: Business Market Segment Industry State Location Organisation size Contact role Verification status Campaign status Customer status Last contacted Opportunity status Response type Users should also be able to create and save dynamic lists from these filters. 24. Duplicate & Organisation Matching The system needs intelligent entity matching. For example: ABC Medical Pty Ltd and ABC Medical should not automatically become separate organisations. AI/fuzzy matching should help detect: Duplicate organisations Duplicate contacts Different branches of the same organisation Parent/subsidiary relationships Multiple websites/domains belonging to the same organisation 25. Audit Trail Important automated actions should be logged. We need to know: What AI discovered Source When it was discovered What data AI changed What campaign contacted someone What was sent Which mailbox/number sent it Who approved it What the recipient replied What CRM action resulted 26. Dashboard The central dashboard should provide a portfolio-level overview with the ability to drill down into individual businesses. It should show metrics such as: Database Organisations | Contacts | Verified Contacts | New Prospects Outreach Campaigns | Emails | SMS | Replies | Positive Replies Sales Leads | Opportunities | Quotes | Sales | Revenue Actions Replies Awaiting Response | AI Drafts Awaiting Approval | Follow-Ups | Data Issues 27. Scalability The platform should be designed to handle: Multiple businesses Hundreds of customer segments/lists Large numbers of organisations Multiple contacts per organisation Multiple sending domains Multiple SMS numbers Multiple staff Large campaign histories Millions of research/data points over time The architecture should avoid building separate systems for every business. 28. Key Design Principle This should NOT be designed as: Web scraper + email sender. It should be designed as a: Central AI Prospect Intelligence, Marketing Automation and Outbound Sales Platform. The long-term value will come from the accumulated organisation database, contact intelligence, campaign history, response history and sales outcomes across multiple businesses. The platform should gradually develop a proprietary commercial database and learn: who to target, what to offer, how to approach them, when to contact them and which campaigns actually generate sales. Development Approach I am open to using proven third-party APIs/services for functions that should not be rebuilt from scratch, including: Email delivery Email verification SMS AI models Web search/data enrichment CRM Authentication Other infrastructure I do not want unnecessary custom development where a reliable API/service already solves the problem. However, the central database, business logic, AI orchestration, campaign intelligence, workflows and user interface should operate as one integrated platform. What I Need From Applicants Please explain: Your recommended architecture and technology stack. Similar AI/SaaS/CRM/outbound systems you have built. How you would approach AI-assisted internet research and large-scale list building. How you would prevent duplicate and poor-quality data. How you would implement email and SMS sending. How you would manage email deliverability and sending-domain reputation. How you would integrate AI personalisation without creating low-quality/spam-like messages. How you would manage inbound email and SMS responses. Your recommended CRM approach — custom CRM versus integration with an existing CRM. How you would address Australian email/SMS compliance and suppression. What third-party APIs/services you recommend. How you would structure the system so additional businesses can be added easily. How you would implement AI campaign learning and optimisation. Recommended development stages. Estimated timeframe and budget for each stage. Important I am not looking for someone to simply build a scraper or connect an email API. I am looking for someone capable of designing the underlying architecture for a platform that can eventually manage prospect discovery, data enrichment, campaign generation, individual personalisation, email/SMS outreach, response management, CRM workflows and campaign learning across a portfolio of businesses. I would prefer the project to be developed in stages so that a usable MVP can be operating relatively quickly, followed by additional automation and AI functionality.