Come migliorare l’engagement tramite analisi dei dati, algoritmi predittivi e Customer Clustering
- 1. Master version 0.0.3 COME MIGLIORARE L’ENGAGEMENT TRAMITE ANALISI DEI DATI, ALGORITMI PREDITTIVI E CUSTOMER CLUSTERING Federica Gandolfi – Head of Data Analyst Stefano Lieto – Head of Account Management 21 NOVEMBRE 2019
- 2. 2 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner TECNOLOGIA E SERVIZI Customer Engagement Audit Benchmark Competition Deliverability Calcoliamo l’impatto delle campagne digitali su tutti i canali di vendita. Analizziamo i comportamenti degli utenti e il livello di engagement del database identificando i customer journey più adatti, utilizzando modelli predittivi. Analisi della UX di specifici processi, includendo benchmark con un numero definito di competitor. Lo scopo è verificare le best practice e i trend di mercato e confrontarli con i punti di forza e debolezza del brand, e fornire recommendation per evidenziarne le criticità e migliorarne l’usabilità. Grazie ad un team di risorse certificate siamo in grado di gestire campagne di email marketing anche su piattaforme di terze parti quali : Salesforce – Adobe – Oracle... Best-practice, monitoraggio continuo con strumenti avanzati, indagine e verifica della consegna delle email, evidenziando potenziali criticità e opportunità di miglioramento della email reputation e deliverability. Multi-Platform Campaign Management
- 3. 3 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner I DATI PER CONOSCERE, ANTICIPARE, MIGLIORARE Algoritmi per prevedere modelli di comportamento Audience Look Alike per colpire target simili Customer Clustering basati su pattern comportamentali + CONVERSIONI
- 4. 4 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner Una rappresentazione olistica del cliente che integra tutti i dati e gli eventi, e ti consente di arrivare ad visione completa dei suoi comportamenti indipendentemente dai canali utilizzati (sito web, social, app, ecommerce, negozio, customer care, ecc.) SINGLE CUSTOMER VIEW
- 5. 5 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner «Una CDP è un sistema di marketing che unifica i dati dei clienti di un’azienda provenienti dal marketing e da altri canali.» (Gartner.com) «Una CDP è un sistema gestito dal marketer che crea un database dei clienti persistente e unificato accessibile ad altri sistemi.» (David Raab – CDP Institute Founder) ALLA BASE UNA «CDP»: CUSTOMER DATA PLATFORM
- 6. 6 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner I 4 ALGORITMI APPLICATI Attribution model Email Engagement Cluster RFM Self-Organizing Map (SOM) Email Intelligence & Customer Engagement1 Customer Purchase Insights2
- 8. 8 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner OBIETTIVI ED EVIDENZE Obiettivo: misurare l’impatto della comunicazione direct sul comportamento d’acquisto della customer base Evidenze Fino a 9 newsletter a settimana per singolo user Personalizzazione dei contenuti solo per gender Poca personalizzazione nei messaggi
- 9. 9 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner ANALISI E RISULTATI RICEVENTI 1.997.555 Con ordini impattati dall’email 399.511 - 20% Senza ordini impattati dall’email 1.598.044 - 80%
- 10. 10 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner ANALISI E RISULTATI RICEVENTI 1.997.555 Con ordini impattati dall’email 399.511 - 20% 60% degli uomini che effettuano acquisti comprano prodotti da donna 15% delle donne che effettuano acquisti comprano prodotti da uomo Senza ordini impattati dall’email 1.598.044 - 80%
- 11. 11 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner ANALISI E RISULTATI RICEVENTI 1.997.555 Con ordini impattati dall’email 399.511 - 20% 60% degli uomini che effettuano acquisti comprano prodotti da donna 15% delle donne che effettuano acquisti comprano prodotti da uomo Senza ordini impattati dall’email 1.598.044 - 80% Incremento scontrino medio: +5% (3 mesi)
- 12. 12 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner ANALISI E RISULTATI RICEVENTI 1.997.555 Con ordini impattati dall’email 399.511 - 20% Senza ordini impattati dall’email 1.598.044 - 80% INTERESTED 40% INACTIVE 37% DORMANT 15% ENGAGED 6% Top 4 email engagement cluster (98% of users)
- 13. 13 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner ANALISI E RISULTATI RICEVENTI 1.997.555 Con ordini impattati dall’email 399.511 - 20% Senza ordini impattati dall’email 1.598.044 - 80% INTERESTED 40% INACTIVE 37% DORMANT 15% ENGAGED 6% Top 4 email engagement cluster (98% of users) INTERESTED 47% INACTIVE 30% DORMANT 12% ENGAGED 9% Cluster distribution post new contact strategy
- 14. 14 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner ANALISI E RISULTATI RICEVENTI 1.997.555 Con ordini impattati dall’email 399.511 - 20% Senza ordini impattati dall’email 1.598.044 - 80% INTERESTED 40% INACTIVE 37% DORMANT 15% ENGAGED 6% Top 4 email engagement cluster (98% of users) INTERESTED 47% INACTIVE 30% DORMANT 12% ENGAGED 9% Cluster distribution post new contact strategy Conversione in clienti post apertura email 47.941 - 3%
- 15. 15 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner ANALISI E RISULTATI RICEVENTI 1.997.555 Con ordini impattati dall’email 399.511 - 20% Senza ordini impattati dall’email 1.598.044 - 80% INTERESTED 40% INACTIVE 37% DORMANT 15% ENGAGED 6% Top 4 email engagement cluster (98% of users) INTERESTED 47% INACTIVE 30% DORMANT 12% ENGAGED 9% Cluster distribution post new contact strategy 60% degli uomini che effettuano acquisti comprano prodotti da donna 15% delle donne che effettuano acquisti comprano prodotti da uomo Conversione in clienti post apertura email 3% Incremento scontrino medio +5%
- 17. 17 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner OBIETTIVI Aumentare la conoscenza dei propri clienti, del loro comportamento, delle loro preferenze riuscendo a mappare le informazioni in modo che siano facilmente disponibili e consultabili Sfruttare le informazioni dei clienti migliori per attivare delle tecniche di lookalike
- 18. 18 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner ANALISI E RISULTATI Algoritmo RFM Customer Segment Activity Champions Bought recently, buy often and spend the most! Loyal Spend good money with us often. Responsive to promotions. Potential Loyalist Recent customers, but spent a good amount and bought more than once. New Customers Bought most recently, but not often. Promising Recent shoppers, but haven’t spent much. Need Attention Above average recency, frequency and monetary values. May not have bought very recently though. About To Sleep Below average recency, frequency and monetary values. Will lose them if not reactivated. At Risk Spent big money and purchased often. But long time ago. Need to bring them back! Cannot Lose Them Made biggest purchases, and often. But haven’t returned for a long time. Hibernating custom Last purchase was long back, low spenders and low number of orders. Lost customers Lowest recency, frequency and monetary scores.
- 19. 19 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner ANALISI E RISULTATI Algoritmo RFM Customer Segment Activity Champions Bought recently, buy often and spend the most! Loyal Spend good money with us often. Responsive to promotions. Potential Loyalist Recent customers, but spent a good amount and bought more than once. New Customers Bought most recently, but not often. Promising Recent shoppers, but haven’t spent much. Need Attention Above average recency, frequency and monetary values. May not have bought very recently though. About To Sleep Below average recency, frequency and monetary values. Will lose them if not reactivated. At Risk Spent big money and purchased often. But long time ago. Need to bring them back! Cannot Lose Them Made biggest purchases, and often. But haven’t returned for a long time. Hibernating custom Last purchase was long back, low spenders and low number of orders. Lost customers Lowest recency, frequency and monetary scores. Strategie di retention Clienti rischio churn Strategia di upsellClienti potenziali CaringClienti migliori
- 20. 20 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner ANALISI E RISULTATI Algoritmo RFM Customer Segment Activity Champions Bought recently, buy often and spend the most! Loyal Spend good money with us often. Responsive to promotions. Potential Loyalist Recent customers, but spent a good amount and bought more than once. New Customers Bought most recently, but not often. Promising Recent shoppers, but haven’t spent much. Need Attention Above average recency, frequency and monetary values. May not have bought very recently though. About To Sleep Below average recency, frequency and monetary values. Will lose them if not reactivated. At Risk Spent big money and purchased often. But long time ago. Need to bring them back! Cannot Lose Them Made biggest purchases, and often. But haven’t returned for a long time. Hibernating custom Last purchase was long back, low spenders and low number of orders. Lost customers Lowest recency, frequency and monetary scores. 7% dei clienti ha fatto upgrade Strategie di retention Clienti rischio churn Strategia di upsellClienti potenziali CaringClienti migliori
- 21. 21 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner ANALISI E RISULTATI Algoritmo SOM
- 22. 22 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner ANALISI E RISULTATI Algoritmo SOM Clienti del fresco Esclusivisti alcolici Clienti da spesa completa
- 23. 23 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner ANALISI E RISULTATI Algoritmo SOM Clienti del fresco Esclusivisti alcolici Clienti da spesa completa • Upsell / Cross sell reparti • Nuova distribuzione dello spazio nel supermercato
- 24. 24 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner ANALISI E RISULTATI Algoritmo SOM Clienti del fresco Esclusivisti alcolici Clienti da spesa completa • Upsell / Cross sell reparti • Nuova distribuzione dello spazio nel supermercato Strategia in fase di definizione
- 25. 25 © Copyright 2017-2020 Contactlab This document may not be modified, organized or reutilized in any way without the express written permission of the rightful owner ANALISI E RISULTATI Algoritmo RFM Algoritmo SOM Customer Segment Activity Champions Bought recently, buy often and spend the most! Loyal Spend good money with us often. Responsive to promotions. Potential Loyalist Recent customers, but spent a good amount and bought more than once. New Customers Bought most recently, but not often. Promising Recent shoppers, but haven’t spent much. Need Attention Above average recency, frequency and monetary values. May not have bought very recently though. About To Sleep Below average recency, frequency and monetary values. Will lose them if not reactivated. At Risk Spent big money and purchased often. But long time ago. Need to bring them back! Cannot Lose Them Made biggest purchases, and often. But haven’t returned for a long time. Hibernating custom Last purchase was long back, low spenders and low number of orders. Lost customers Lowest recency, frequency and monetary scores. Clienti del fresco Esclusivisti alcolici Clienti da spesa completa • Upsell / Cross sell reparti • Nuova distribuzione dello spazio nel supermercato Strategie di retention Clienti rischio churn Strategia di upsellClienti potenziali CaringClienti migliori Audience Lookalike
- 26. Grazie! Via Natale Battaglia, 12 | 20127 Milano explore.contactlab.com | explore@contactlab.com +39 02 28 31 181