Lead scoring + auto-assignment CRM Odoo: predictive vs rule-based (guida 6 di 15)

Settantanovesima guida operativa Odoo 19 per PMI italiane. Sesta della serie 15 guide CRM Odoo. Senza lead scoring, ogni lead riceve eguale attenzione — ma non sono uguali. Un CFO di Fortune 500 e uno studente curioso producono lead diversi. Lead scoring permette di prioritizzare automaticamente, focalizzando sales team sui prospect ad alto potenziale. Auto-assignment chiude il cerchio: lead “hot” al senior AE, lead “warm” al SDR per qualifying, lead “cold” al marketing nurture. In questa guida vediamo come implementare lead scoring + auto-assignment in Odoo CRM.

Vediamo: cosa è lead scoring, scoring rule-based vs predictive, criteri scoring tipici PMI, configurazione Odoo Enterprise, auto-assignment routing, casi pratici PMI italiane.

Cosa è lead scoring

Definizione

• Assegnare punteggio numerico ad ogni lead
• Basato su criteri demografici + comportamentali
• Higher score = higher probability di conversion
• Prioritizzazione automatica per sales

Tipi di scoring

• Demographic: company size, industry, role
• Firmographic: revenue, employee count, location
• Behavioral: visits, downloads, email opens
• Predictive: ML-based su pattern history

Beneficio business

• Sales time focused on best opportunities
• Conversion rate +25-40% tipico
• Lead response time -50% per high-score
• Marketing-sales alignment migliore

Rule-based scoring

Approccio più semplice: criteri esplicit definiti da te. “Se company > 500 dipendenti +30 punti, se settore = target +20, se ha downloaded whitepaper +15”. Trasparente e gestibile senza ML expertise:

• Define criteria semplici esplicit
• Assegna punteggi positivi/negativi
• Compute score automatic
• Threshold per classification (Hot/Warm/Cold)
• Easy to explain a sales team
• Easy to adjust over time

Esempio scoring rules B2B SaaS

• Company > 100 dipendenti: +20
• Industry target: +15
• Decision maker title (CIO, CFO): +25
• Italian/EU company: +10 (target market)
• Downloaded pricing page: +15
• Free email (gmail.com): -10 (probabilmente non business)
• Studente/Studi: -20 (low fit)

Predictive lead scoring (Enterprise)

Odoo Enterprise include predictive scoring basato su ML. Analizza pattern di lead chiusi vs persi storicamente, identifica feature predittive automaticamente, score nuove lead con accuracy alta. Setup richiede 12-18 mesi storico minimum:

• ML model trained su historical data
• Auto-feature engineering
• Score 0-100 automatic
• Re-train periodically
• Better accuracy vs rule-based
• Explainability: top contributors per score

Auto-assignment rules

Score lead = primo passo. Auto-assignment = chiusura cerchio. Lead high-score → senior AE. Medium → SDR. Low → marketing nurture. Configurabile per team/territorio/skill:

• Score-based routing
• Geography matching
• Industry vertical specialist
• Round-robin within team
• Workload balancing
• Skip out-of-office

Team-based assignment

Multi-team organization (vedi guida #84) si combina con scoring. Team Nord per Italia Nord, Team Enterprise per high-score, Team SMB per low-score. Configurazione per ogni team:

• Team definition per geography
• Team capacity considered
• Team-level scoring criteria
• Cross-team rules per spillover
• Dashboard team performance

Priority levels + visual cues

Lead score si traduce in priority level visible nella pipeline. Star icon per high-priority, color coding per stage. Sales team vede immediatamente cosa lavorare prima:

• Star icon (0-3) per priority
• Color coding per stage
• Sort by priority default
• Filter “High priority only”
• Manager view aggregate
• Alert for stale high-priority

Criteri scoring tipici per settore

SaaS B2B

• Company size + decision maker title
• Use case fit
• Tech stack compatibility
• Geography (within service area)
• Behavior: trial signup, demo request

Manifattura B2B

• Industry vertical
• Order volume potential
• Existing supplier landscape
• Geographic proximity
• Behavior: spec sheet download, sample request

Servizi professionali

• Company size
• Project complexity
• Budget signals
• Decision timeline urgency
• Behavior: consultation request, RFP

E-commerce B2B

• Account age
• Purchase history
• Cart abandonment
• Engagement frequency
• Average order value

Lead scoring lifecycle

Initial scoring at creation

• Demographic data scored immediately
• Source-based bonus
• Initial classification (Hot/Warm/Cold)
• Routing decision

Score updates over time

• Behavioral data added (website visits)
• Email engagement scored
• Re-classification if threshold cross
• Routing re-evaluation

Score decay

• No activity 30 giorni: -10
• No engagement 60 giorni: -20
• Prevents stale leads inflation
• Force re-engagement or close

Scoring + assignment automation

On lead creation

• Auto-compute initial score
• Auto-assign based on rules
• Auto-create initial activity
• Auto-send welcome email

On score change

• Re-classify if threshold cross
• Re-route to different rep if needed
• Update priority visual
• Alert manager if significant change

On behavioral event

• Website visit: +5
• Pricing page: +10
• Demo request: +25
• Whitepaper download: +5
• Multiple emails opened: +10

Errori comuni lead scoring

“Scoring senza data validation”

Problema: rules basate su intuition, no historical validation.
Soluzione: analyze conversion per criteria, calibrate punteggi.

“Mai update scoring rules”

Problema: rules ferme da 2 anni, processo evolved.
Soluzione: quarterly review + adjustment.

“Solo demographic, no behavioral”

Problema: scoring incomplete, missing intent signals.
Soluzione: combine demographic + behavioral.

“Score irrelevant per sales”

Problema: sales ignores score, lavora per propria intuition.
Soluzione: training + show conversion data per score bucket.

“Auto-assign senza overflow”

Problema: high-score lead to vacation AE, lead stalls.
Soluzione: fallback rules per out-of-office, capacity check.

Casi pratici PMI italiane

Caso 1 — SaaS: predictive scoring

• Odoo Enterprise predictive
• 18 mesi history train model
• Conversion: +38% high-score lead
• Sales productivity: +25%

Caso 2 — Manifattura: rule-based

• 12 scoring rules simple
• Industry + size weights heavy
• Auto-assign per area geografica
• Response time: 2h → 20min

Caso 3 — Servizi: hybrid approach

• Initial rule-based score
• Predictive overlay post 6 mesi data
• Manual override possible
• Best of both worlds

Caso 4 — E-commerce B2B: behavioral-heavy

• Website tracking deep
• Cart + product views scored
• Real-time scoring updates
• Auto-trigger sales call high-score

FAQ

Predictive vs Rule-based: quale scegliere?

Start rule-based per setup veloce. Migrate predictive quando hai 12+ mesi data. Hybrid optimal: rule-based + predictive overlay. Predictive necessita Enterprise.

Quanti criteri scoring?

10-20 criteri tipico. Sotto 10: rough. Sopra 25: complex senza accuracy boost. Focus su top 5-10 differenziatori.

Score decay è importante?

Sì critico. Senza decay, vecchi lead alti score inflate prioritization. Implementa anche se simple (es. -10 per ogni 30gg no activity).

Sales può vedere logica scoring?

Trasparenza essential. Sales che capisce perché lead è alto score: trusts data, focus right. Hidden scoring = ignored scoring.

Posso scoring lead manuale?

Sì, manual override sempre disponibile. Sales può aumentare score se reasoning specifico (es. “knows prospect personalmente”). Track override per analytics.

Prossimi passi

Nella prossima guida (7/15 serie CRM) vedremo forecasting + probability management: weighted forecast, sales-led vs marketing-led, accuracy improvement, dashboard executive.

Vuoi lead scoring nella tua PMI?
G Tech Group implementa lead scoring + auto-assignment CRM Odoo per PMI italiane: rule-based + predictive, routing intelligente, dashboard analytics. Track record di conversion +30% post-implementation.
Richiedi un preventivo gratuito oppure prova la nostra demo Odoo 19 live. Oppure prova Odoo direttamente su odoo.com (link partner Brentasoft).