Loading...
Help Center / Getting Started / Mango GTM App Overview

Mango GTM App Overview

Apps Getting Started Updated April 30, 2026 8 views

Mango GTM App Overview

The revenue-side of the MangoApps stack — Product Advisor chat on the marketing site, demo request lifecycle with HubSpot sync and geo-routing, deterministic lead scoring with LLM rationale, and a Prospecting Agent that drafts outbound grounded in Mango IQ battle cards.

What is the Mango GTM App?

Mango GTM owns the revenue-side lifecycle for the tenant business — anonymous visitor on the marketing site → Product Advisor chat → Book-a-Demo request → scored lead → outbound follow-up grounded in competitive context. It’s the action layer on top of Mango IQ (which observes and analyzes the competitive market): Mango IQ sees the market, Mango GTM acts on it.

Distinct from AI Front Office (B2C front desk for the tenant business’s own customers) — Mango GTM is the B2B SaaS sales funnel for the tenant business itself.

Core Value Proposition:

  • 💬 Product Advisor → Pipeline — Anonymous visitor chat extracts CI context (competitor, content gap, feature comparison, market theme) and converts buying-intent into demo requests automatically.
  • 🎯 Deterministic Lead Scoring — Reproducible 0–100 score from advisor signals + visitor journey + CI context, paired with LLM-generated prose rationale.
  • 📝 CI-Grounded Outreach — Prospecting Agent drafts outbound email referencing matching battle cards, feature comparisons, and content briefs from Mango IQ.
  • 🧭 Unified Activity Timeline — Advisor conversations + demo requests in one timeline, filterable by competitor, content gap, market theme, or buying intent.

At a Glance

⏱️ Setup Time 🔗 Integrations 🤖 AI Agent 📊 Score Method
~30 minutes HubSpot, ZoomInfo, LinkedIn Ads, Google Ads Prospecting Agent + Product Advisor Deterministic + LLM rationale

Perfect For:

  • 🏢 B2B SaaS go-to-market teams — Replace the disconnected stack of website chat + form router + scoring tool + outbound drafter with one app.
  • 📈 SDRs / AEs working a queue — My Leads view ranks by deterministic score with prose rationale; CI context attaches to every record.
  • 🧠 Marketing leaders running competitive plays — Filter the GTM Activity timeline by competitor to refresh battle cards from real visitor behavior.

How It Works

┌─────────────────────────────────────────────────────────────────────────┐
│             ANONYMOUS VISITOR → CLOSED REVENUE                          │
├─────────────────────────────────────────────────────────────────────────┤
│   ┌──────────────┐      ┌──────────────┐      ┌──────────────┐         │
│   │  PRODUCT     │─────▶│  CI CONTEXT  │─────▶│  DEMO        │         │
│   │  ADVISOR     │      │  EXTRACTOR   │      │  REQUEST     │         │
│   │  (chat)      │      │  (LLM nano)  │      │ +HubSpot sync│         │
│   └──────────────┘      └──────────────┘      └──────┬───────┘         │
│                                                       │                 │
│                                                       ▼                 │
│   ┌──────────────┐      ┌──────────────┐      ┌──────────────┐         │
│   │  PROSPECTING │◀─────│   LEAD       │◀─────│   ROUTING    │         │
│   │  AGENT       │      │   SCORING    │      │   RULES      │         │
│   │  (drafts)    │      │  (det + LLM) │      │ (geo/RR/AB)  │         │
│   └──────────────┘      └──────────────┘      └──────────────┘         │
│                                                                         │
│   GTM ACTIVITY TIMELINE  ◀─── unified view across all stages           │
└─────────────────────────────────────────────────────────────────────────┘

Where Mango GTM Sits in the Revenue Stack

                ┌────────────────────┐
                │      MANGO IQ      │   ← observes market, competitors, content
                │ (CI canonical data)│     (provides battle cards, feature matrices)
                └─────────┬──────────┘
                          │ reads canonical data
                          ▼
                ┌────────────────────┐
                │     MANGO GTM      │   ← B2B sales funnel
                │ (revenue lifecycle)│     advisor → demo → score → outreach
                └─────────┬──────────┘
                          │ won proposal
                          ▼
                ┌────────────────────┐
                │   MANGO BILLING    │   ← AR / dunning / RevRec / commissions
                └─────────┬──────────┘
                          │
                          ▼
                ┌────────────────────┐
                │     MANGO CS       │   ← post-sale customer success
                └────────────────────┘

Key Features

💬 Product Advisor (Inbound Intelligence)

The visitor-facing chat agent on the marketing site that turns anonymous traffic into qualified pipeline.

Feature Description
Visitor-facing chat Lives on marketing pages; persists conversations cross-session
ZoomInfo enrichment Visitor → company match for B2B context
CI context extraction CiContextExtractor (nano-tier LLM) extracts competitor, content gap, feature comparison, market theme
Buying-intent detection Conversation patterns flag high-intent visitors
Topic gate Nano-tier LLM filters off-topic / jailbreak attempts before RAG cost
Heuristic fallback If LLM is unavailable, deterministic heuristics still classify
Brand voice configurable advisor_brand_voice admin setting injects extra guidance into the system prompt
Welcome message override Custom greeting per business
Required-link nudges Admin lists links + “when to include” — advisor prefers them as [RESOURCE:...] links
Topics to avoid Admin lists topics the advisor will politely decline
Escalation phrases Strong buying-intent triggers; advisor emits [SCHEDULE_DEMO] (server-side safety net appends it if missed)

💡 Pro Tip: Use the advisor_required_links setting to ensure your highest-converting case studies and pricing pages land in advisor responses when the conversation hint matches — no more linking back to the homepage.

📋 Demo Request Management

The full lifecycle from public Book-a-Demo form through closed-won.

Feature Description
Public Book-a-Demo form With honeypot + rate-limit protection
HubSpot Form API sync Portal-configurable; retry + failure notifications built in
Geo resolution IP → country / continent, cached 7 days
Lifecycle stages new → contacted → qualified → demo → proposal → closed
Auto-assignment Priority-ordered routing rules (round-robin, least-busy, or specific-admin)
High-priority flag Auto-set when Product Advisor detects buying intent
CSV bulk import With field validation
My Leads view Per-rep queue ranked by score

💡 Pro Tip: Routing rules are priority-ordered. Put your highest-value rules (e.g. “Enterprise + North America → Sara”) first; the first match wins.

🧮 Lead Scoring & Prioritization

Deterministic score + LLM rationale — reproducible across reruns, not just a vibes vote.

Feature Description
Numeric score 0–100 with tiered banding (deterministic, reproducible)
Signal sources Buying intent, visitor profile depth, CI context depth, company size, geo, routing match
LLM rationale Medium-tier LLM generates prose explaining the score (OpenAI via RubyLLM)
Similar-prospect search pgvector over visitor extracted data — find lookalikes of recent wins
Bulk re-scoring RescoreAllAccountsJob recomputes when scoring weights change

Mango::Gtm::LeadScoringService is the single source of truth.

🤖 Prospecting Agent (Outbound)

Natural-language chat interface for GTM workflows that drafts outbound grounded in CI data.

Capability Description
Outbound email drafting Grounded in CiCompetitorProfile battle cards from Mango IQ
Follow-up drafting References visitor journey (pages viewed, questions asked, competitor mentioned)
Hook types feature_gap, competitor_win, content_topic
Outreach channels Email, LinkedIn
Audit trail Every draft persists with full generation metadata (model, prompt, rationale)
Internal “Mark Published” flag v1 — direct HubSpot Engagement write-back is on the roadmap

Agents::MangoGtmAgent ships with the app. Read-only intent for sensitive operations; tools allowlist for write paths.

🎯 Target Accounts & ICP

Feature Description
ICP profiles Define the ideal customer profile per business
Target Account list Curated accounts for ABM motion
Account brief generator LLM-drafted account brief grounded in ICP signals
Similar-prospect company search SimilarProspectCompanyService — find companies that look like a won deal
External prospect provider seam ExternalProspectProvider interface — Apollo / Clay / Clearbit ready (integration deferred)

🧭 GTM Activity Timeline

The unified surface that correlates everything.

Filter Surfaces
Competitor Every visitor + demo request that mentioned this competitor
Content gap Cases where the advisor flagged missing content
Feature comparison Visitors comparing specific features
Market theme Higher-level themes (e.g. “frontline ops”)
Buying intent High-intent visitors first
High priority Demo requests flagged by routing or advisor

📣 Plays & Cadences

Structured outbound motions you build once and run repeatedly.

Feature Description
Plays Reusable templates of outbound steps
Play enrollments A target account or contact joining a play
Play runs Per-enrollment execution record
Play scheduler PlaySchedulerJob triggers next-step actions on cadence
Reply detection CadenceReplyDetector halts cadences when a prospect replies
Auto-reply filter Don’t count auto-responders as engagement
Play analytics Per-play conversion + reply rates

📰 Narrative Pillars + Mango Scoop

Feature Description
Narrative pillars The positioning canon — admin authors, advisor + agent reference
Embeddings GenerateNarrativePillarEmbeddingJob builds vectors for retrieval
Mango Scoop newsletter Internal newsletter subscriber surface (only enabled for the tenant that owns the global subscriber list)

📺 Advertising (LinkedIn + Google Ads)

Optional capability behind paid_media_enabled — the tenant connects LinkedIn or Google Ads accounts and runs ABM campaigns from inside Mango GTM.

Feature Description
LinkedIn Ads integration OAuth-connect; push campaigns; reporting sync
Google Ads integration OAuth-connect; push campaigns; reporting sync
AI ad creative generation AdVariantGenerator + AdCreativeRenderer produce campaign-ready creative variants
Audience CSV builder Build PII-safe contact list audiences from advisor + demo data
Approval gate ad_require_approval_above_cents — campaigns above threshold require admin approval (default $500)
Monthly budget cap ad_monthly_budget_cap_cents — hard cap across the tenant
Dry-run mode Log intended platform API calls without executing — dev / test
LinkedIn Insight Tag Partner ID setting + auto-injection into the marketing site
LinkedIn organic posting Member + company page posts (separate linkedin_organic_enabled flag)

🤝 Sales Proposals

The proposal lifecycle that hands off to Mango Billing on close.

Feature Description
Proposal authoring Tenant-branded proposals built off ICP signals
Proposal expiration Billing::ProposalExpirationJob marks expired
Mark won → Mango Billing Billing::ProposalService#mark_won fires CommissionEarning creation
Renewal proposals RenewalProposalGenerator for recurring deals

📊 Analytics & Marketing Funnel

Mango::Gtm::AnalyticsService powers the analytics tab.

Standard Analytics:

  • Lead funnel (new → contacted → qualified → demo → proposal → closed)
  • Score distribution and conversion by score band
  • Per-rep lead volume and conversion
  • Per-competitor pipeline volume
  • HubSpot sync health (success / failure rates)

Marketing Funnel (only for the tenant that owns the global PageVisit table):

  • Visitor + Marketing views
  • Page-level conversion
  • Source attribution

🔗 Integrations

Integration Purpose
HubSpot Form API Demo request sync (portal-configurable)
ZoomInfo Visitor → company enrichment
LinkedIn Ads + Marketing Paid media + organic posting
Google Ads Paid media
Mango IQ CI canonical data (CiCompetitorProfile, CiFeatureComparison, CiContentGap, market themes)
Anthropic Claude + OpenAI LLMs via RubyLLM / LlmConnectionManager
External prospect providers Apollo / Clay / Clearbit seam (deferred)

⏰ Background Jobs

Job What it does
MangoGtm::ScanIntentSignalsJob Periodic scan for buying-intent signals across visitor data
MangoGtm::ExtractCiContextJob Extract CI context from advisor conversations
MangoGtm::GenerateAccountBriefJob Create AI account briefs
MangoGtm::EnrichProspectCompanyJob Run ZoomInfo / external enrichment
MangoGtm::Generate*EmbeddingJob Build pgvector embeddings for advisor + demo + narrative pillars
MangoGtm::DiscoverExternalProspectsJob Run external provider discovery (when integration is enabled)
MangoGtm::SlaMonitorJob Monitor demo-request response SLA
MangoGtm::CampaignWatcherJob Sync ad campaign state
MangoGtm::PushCampaignToLinkedInJob Push campaigns out to LinkedIn
MangoGtm::Sync*ReportingJob Pull LinkedIn / Google Ads reporting data
MangoGtm::PlaySchedulerJob + RunPlayJob Execute play cadences
MangoGtm::Refresh*TokenJob Keep LinkedIn / Google Ads OAuth tokens fresh
MangoGtm::RescoreAllAccountsJob Bulk re-score after weight changes
MangoGtm::ScanOutreachEngagementJob Detect replies / clicks on outbound

📧 Notifications

Mailer When
MangoGtm::OutboundOutreachMailer Outbound prospect emails (when not using HubSpot)

Plus:

  • Admin email notifications when advisor_buying_intent_recipients is set (admin emails get a buying-intent ping)
  • HubSpot sync failure notifications

Cost & Budget Guardrails

Setting Default
Monthly LLM budget $25 (monthly_llm_budget)
Monthly ad spend cap Unlimited by default (ad_monthly_budget_cap_cents)
Approval threshold $500 (ad_require_approval_above_cents)
Required reviewers per creative 1
Dry-run mode Off (recommended ON during dev / test)

User Roles & Permissions

Role Capabilities
Member No access by default
Manager Access to My Leads, Demo Requests, GTM Activity timeline, scoring views
Admin / Super Admin Everything Manager can do, plus routing rules, ICP / target accounts, advisor configuration, prospecting outreach generation, ad campaigns + budget management, narrative pillars, settings

Default allowed_roles = [manager, admin, super_admin]. Reps see their assigned leads (My Leads); routing rules and outreach generation are admin-gated.

How We Compare

Mango GTM sits in a category with enterprise GTM stacks (HubSpot Sales Hub, Salesforce + Pardot, Apollo.io) — except they don’t ground outbound in CI data, and they don’t bundle the website Product Advisor.

Feature MangoApps Workforce HubSpot Sales Hub Apollo.io Salesforce + Pardot
Product Advisor chat with CI context extraction ✅ ❌ ❌ ❌
Deterministic + LLM-rationale lead scoring ✅ ⚡ ⚡ ⚡
CI-grounded outbound drafting (battle cards) ✅ ❌ ⚡ ❌
HubSpot Form API sync ✅ n/a ⚡ ❌
Unified advisor + demo timeline ✅ ❌ ❌ ❌
LinkedIn + Google Ads integration in-app ✅ ⚡ ⚡ ✅
Native to a workforce platform (HR + Ops + Finance) ✅ ❌ ❌ ❌
AI-native, not bolted on ✅ ⚡ ✅ ⚡
Legend: ✅ Included ❌ Not Available ⚡ Limited / partial

Why MangoApps Workforce?

  • 🔗 Unified Platform — Mango GTM reads Mango IQ canonical data directly. Won proposals flow into Mango Billing. CS handoff lands in Mango CS. No integration glue.
  • 🤖 CI-Grounded by Default — Every advisor conversation extracts competitor, content gap, feature comparison, and market theme. Every outbound draft references those signals. No other GTM tool bundles this.
  • 🛡️ Cost Caps Built-In — Monthly LLM budget + ad spend cap + approval thresholds + dry-run mode are first-class settings, not feature requests.

Getting Started

For Administrators

  1. Enable Mango IQ first (recommended) — Mango GTM runs without Mango IQ, but loses most of its value. Battle cards, feature matrices, and content briefs all come from Mango IQ.
  2. Enable the app — Admin → Apps Marketplace → Mango GTM. Toggle app_enabled.
  3. Connect HubSpot — Configure hubspot_sync_enabled + portal credentials. Demo requests will sync.
  4. Author advisor configuration — Brand voice, welcome message, topics-to-avoid, required links, escalation phrases. The advisor is only as good as the canon you give it.
  5. Build routing rules — Priority-ordered. Round-robin, least-busy, or specific admin assignment.
  6. Define your ICP — Mango GTM → ICP Profiles. The lead scoring service uses this; the Prospecting Agent grounds drafts in it.
  7. (Optional) Enable advertising — paid_media_enabled + connect LinkedIn / Google Ads accounts. Set the monthly budget cap and approval threshold.
  8. Pilot the Prospecting Agent — Generate a few outbound drafts, review them, refine the narrative pillars based on what reads well.

For SDRs / AEs

  1. Open My Leads — Your queue, ranked by score, with prose rationale per lead.
  2. Read the rationale — Why did this lead score high? Which competitors did they mention? Which content did they read?
  3. Use the Prospecting Agent — Ask it to draft a follow-up grounded in the visitor’s advisor conversation. Edit, send.
  4. Filter the GTM Activity timeline — Find similar leads when you’re prospecting expansion.

Best Practices

  • ✅ Start with Mango IQ canon — Battle cards, feature matrices, content briefs are the fuel for advisor + agent. Skip them and the agent has nothing to ground in.
  • ✅ Keep advisor required-links current — Stale links = stale advisor. Refresh quarterly with the latest case studies.
  • ✅ Use deterministic scoring as the queue ranker, not a forecast — The LLM rationale is the why; the number is the order.
  • ✅ Run paid media in dry-run mode for the first week — Verify campaigns, audiences, and budgets before any platform API actually executes.
  • ✅ Set the monthly LLM budget conservatively — Default $25 covers a normal week of advisor + agent traffic; raise after observing usage.
  • ✅ Refresh narrative pillars when positioning shifts — They’re embedded for retrieval; advisor + agent both read them.
  • ✅ Don’t disable HubSpot sync mid-flight — Demo requests in flight will leave orphan records. Disable, then drain, then re-enable.

Frequently Asked Questions

Q: How is Mango GTM different from Mango IQ?
A: Mango IQ observes and analyzes the competitive market (competitors, feature matrix, content gaps, keyword opportunities). Mango GTM acts on it — owning the revenue-side lifecycle from anonymous visitor chat through demo request, scoring, and outbound prospecting.

Q: Does installing Mango GTM change anything on the public website?
A: No. The Product Advisor widget and the Book-a-Demo form on the marketing site are unchanged, along with all submission endpoints. Mango GTM only adds admin-side surfaces and agent tools.

Q: Does Mango GTM require Mango IQ to be installed?
A: Mango GTM runs without Mango IQ, but loses most of its value without competitive context. The Prospecting Agent, CI context filtering, and score signals all pull from Mango IQ’s canonical services.

Q: How are outbound emails sent?
A: v1 persists all drafts as CiProspectOutreach records with an internal “Mark Published” flag. Direct HubSpot Engagement write-back and automated sending are on the roadmap.

Q: Can Mango GTM find new prospects via external data providers?
A: The interface seam (ExternalProspectProvider) is in place for Apollo / Clay / Clearbit-style integrations. Integration work is deferred until v1 dogfooding validates the inbound-first workflow.

Q: How is this different from AI Front Office?
A: AI Front Office is the B2C front desk for the tenant business’s own customers (homeowner calling a plumbing co, salon client booking an appointment). Mango GTM is the B2B SaaS sales funnel for the tenant business itself — anonymous visitor on the marketing site → demo request → revenue. Different audience, different stage, different agent.

Q: Who can access Mango GTM?
A: Access is role-gated; default allowed_roles = [manager, admin, super_admin]. Super admins implicitly have access. Non-admin reps can see their assigned leads (My Leads) but not routing rules or outreach generation.

  • Apps Overview — Browse the full marketplace catalog
  • Mango IQ — Competitive intelligence canon that fuels GTM
  • Mango Billing — AR / dunning / RevRec / commissions (where won proposals land)
  • Mango CS — Post-sale customer success

Mango IQ sees the market. Mango GTM acts on it. From anonymous visitor through closed deal, grounded in CI data at every step.

Related Articles

Mango Billing App Overview Mango CS App Overview Mango IQ App Overview Apps & Extensions Overview

Was this article helpful?

Article ID: mango_gtm_app_overview
Ask AI Product Advisor

Hi! I'm the MangoApps Product Advisor. I can help you with:

  • Understanding our 40+ workplace apps
  • Finding the right solution for your needs
  • Answering questions about pricing and features
  • Pointing you to free tools you can try right now

What would you like to know?

AI-generated responses may not be 100% accurate