4th SG13 Regional Workshop for Africa on “Future Networks for a better Africa: IMT-2020, Trust, Cloud Computing and Big Data” (Accra, Ghana, 14-15 March 2016) Big Data – The Analytic Vision Abdallah Ajlani Ph.D., Principal Consultant, ERICSSON Abdallah.ajlani@ericsson.com Byline Outline APPROACH TO BIG DATA EEA USE CASES INTEGRATION IN OOREDOO TN Telecom Big Data Unique Challenges Facebook Roaming Service Device Care Subscriber TetheringSpeedVoice IncidentChurnQoE Location Coverage VIP How to interpret GUMMEI BCCH IupsX2APkHz RNTI DRX RTT Ic/No PHICHNBAP QCI RAB ISUP ARQ HSDSCH-CC ICIC RACH/FACH RANAFQPSKCQI PSPRSCPS1APID How to correlate E2E OSS/BSS CONTENT RADIO CORE NETWORK REAL TIME INSIGHTS AND ACTIONS Telecom Data Models & Insights Networks, SOC, Care, Marketing Subscriber Analytics application ? Human “data Big Data scientist” Store Closed loop action, e.g. Self Care ? Query Real-Time ? Raw data Stream analysis TELECOM DATA MODELS AND INSIGHTS Real-Time Closed loop action, e.g. Policy Control • Subscriber Profile • OTT usage • Service Usage • Service Profile & Sessions • Lifecycle, CE index, CLV • Location, mobility, roaming… Netwo Network • Service Level Index • Quality Indicators (SKPIs)… Device rk • Incidents and Root Cause • Device Profile • Traffic, Load and Forecast • Device usage • Performance ACTIONABLE INSIGHTS FROM COMBINATION OF OFFLINE AND ONLINE DATA Service • Radio, Core, Data UNIQUE, END-TO-END Center… • Tethering • Replacement patterns… EXPERTISE THAT EMPOWERS REAL WORLD USE CASES UNIQUE data modeling Technical Network Performance Resource-KPIs & Counters Session QoE per User Service-KPIs & E2E session record Took too long to start Accessibility Video didn’t start Video Accessibility Video stopped Video Access Time Video/App crashed Retainability Buffer time is too long Video freezes often Integrity Video quality is bad Video Freeze Rate Part of the video is lost Video Quality User Overall Perception of Service Quality User Satisfaction Service Level Index (SLI) Net Promoter Score (NPS) Video Retainability Resolve & Retain Drive ARPU The Service kpi concept How to measure service experience? A/R/I For Video Accessibility e.g. End-users Video Accessibility Video Access Time Retainability e.g. Video Retainability Integrity e.g. Video Freeze Rate Video Quality Services & Apps Identify Network Understand Act Understand the Why of issues • Determine the most probable cause when an S-KPI has a threshold violation using the Correlate Subscriber QoE with Device, Radio and Core network Resource KPIs (R-KPIs) for each session Device “Individual Customer” “Holistic” “Real Time” End-to-End Session Record (ESR) User Servic e QoE Device Handovers Cell load QoE Interferen ce Coverage Video freeze CPU load Radio bearer Drop/outa ge Page delays Radio Core Transp ort … PDP connectivi ty UE type Topology Subscriptio n Content type Service providers Attach Roaming Service Catalog CRM data Service quality Voice quality QoS class Load Charging Profile Settings Throughpu Web experience t App usage Locati on Subscriptio n Coverage Usage Probes ….. 2G OSS/BSS CONTENT NETWORK Terminals ENB MME Congestion SGW PDN GW Internet Service Level Index (SLI) VoLTE Sessions Voice / SMS Sessions Service Level Index (SLI) Web Browsing Sessions Video Streaming Sessions Subjective Weight – Early Adopter Weight – Family Subjective Subjective Weight – Student Subjective Weight – Professional Objective Service-KPIs: Video Streaming Video Start Time Video Image Quality Video Freeze Video Retainability X Service X Incident Significance Repetition X Memory Fading = Predictor of Customer Satisfaction Available for Every User, Continuously Updated Correlates with Net Promoter Score SLI Correlation with NPS Outline APPROACH TO BIG DATA EEA USE CASES INTEGRATION Expert analytics themes OPERATIONAL & BUSINESS EFFICIENCY 3 2 PROACTIVE SERVICE OPERATIONS 1 EMPOWERED CUSTOMER CARE 8 CUSTOMER EXPERIENCE & MARKETING 4 5 EXPERIENCE BASED MARKETING BUSINESS INNOVATION 6 CUSTOMER JOURNEY OPTIMIZATION ANALYTICS PLATFORM & INSIGHTS DATA MONETIZATION 7 INDUSTRY & SOCIETY (IoT) 1 ANALYTICS PLATFORM & insights FLEXIBILITY REDUCED COST Flexible KPI Management Reduced TCO (HW footprint and 3PP license cost) Data Source Ingestion Micro-service architecture CAPABILITY Data Center Deployment Hadoop 2.0 ADVANCED MODELS Self Learning Rules Streaming Analytics to provide near real-time insights Nita Actuation Logic CLI - Cell Level Index Cell 567340 CLI: 8.4 Video CLI: 8.1 Web CLI: 9.3 CEI - Customer Journey 2 Proactive service operations IMPROVED EFFICENCY IN OPERATIONS • Customer-centric analysis and added drill-down capability • Workflows and functionalities (e.g. roaming, CSP Incidents, Drill-down to Incident Summary/List) SESSION BROWSER SERVICE QUALITY MANAGEMENT • Proactive / (Near) real-time monitoring of Service Quality / SLA’s • Reduce handling time with root cause and Impact analysis VIP AND ENTERPRISE CUSTOMERS • Rapid identification of root cause • Assure strict Enterprise level Service SLAs • Deep flow cross-protocol analysis • E2E Actuations to optimize experience of certain VIP and Enterprise customers. • Per subscriber control & user plane event browser 3 EMPOWERED CUSTOMER CARE TIER 1 & TIER 2 IMPROVEMENTS • Further Improve Efficiency of Tier-1 Customer Care Agents and Expand Use to Tier-2+ Agents • Improve out-of-box Most Probable Causes and next Best Actions ROAMING ANALYTICS • Proactive monitoring of Inbound Roaming to avoid Revenue Leakage • Experience on Inbound/ Outbound Roaming • Including input from UDM INTELLIGENT SELF-CARE • Reduce cost with self care capabilities to manage complaints • Intelligent selfservice recommendations CHARGING & BILLING INSIGHTS • Enrich with per transaction data from call & account history • Find correlation between charging & experience 4 Experience based marketing Loss Service Level Index • Calibration Enhancements • Improved SLI Root Cause drilldown • SLI Validation improvements with statistics and tracing PRE-PAID RETENTION Profit SLI ENHANCEMENTS • Identify customers holding back on their spending • Trigger action when user demand increases Expand Customer Lifecycle Value (CLV) MOBILE BROADBAND BOOSTER Profit CLV • Identify pre-paid customers to keep and start monitor • Trigger action when Extend Customer Lifecycleusers at a critical stage of their lifecycle CLV Loss Loss Profit ARPU OPTIMIZATION CLV Expand Customer Lifecycle Value (CLV) • Identify users at a critical stage of their lifecycle • Convert ad hoc data users to a reoccurring data package 6 Data MONETIZATION RETAIL CAMPAIGNS OPTIMIZATION • • • Allow operators to monetize anonymized subsciber insights based on their interests. Footfall analysis Lead generation LOCATION BASED NETWORK ANALYTICS • SPECIAL EVENTS Combine network quality and location data and improve network customer experience • • • Demograpic analysis Traffic patterns and footfall analysis Upsell / Crossell CONTEXTUAL MARKETING • Real time context, interests and behavioural patterns to allow B2B monetization Analytics use cases SoC: Proactive Service Operations SoC: VoLTE Analytics CC: Improve call resolution in customer ca Marketing Analytics: SoC: Real-Time Enterprise SLA Assurance Marketing Analytics: Dynamic Experience Management Customer Retention & Drive Spend Outline APPROACH TO BIG DATA EEA USE CASES INTEGRATION ANALYTICS ARCHITECTURE JMS actuation** Trouble ticket request** Operator’s Map server* 3PP Exposure and Analytics tools * Web service (Api’s) Exposure and Visualization Layer Customer Care Views Service Operations Views Marketing Analytics Dynamic QoE Management Views Executive Dashboard Web application server SNCD loader Processing Layer Correlation Layer Connection Layer In memory data grids Online repository (GREENPLUM) Storage Layer Incident generator Reference data CRM Cell info IMEI TAC Billing Esr data loader Aggregator Stream analytics Extended session records Data loaders Offline repository (HADOOP) Real time Actuation PCRF Subscriber Profiler and Sli calculation Real time event stream Real Time Correlator GTP External Generic RNC / E/// GPEH E/// CTR E/// EBM Generic SGSN / E/// ctum Generic KPI Interface Probes adapters adapters adapters MME Interface adapters eNodeB If Subscriber data CDR Adapters IMS Probes 3G UE rnc O S S sgsn sgsn / MME 4G UE enb ggsn MSC EMM IM S Integration with OSS/BSS SUITE CRM API for CRM GUI 3PP Exposure and Analytics tools TroubleTicketin g Self Care Reference data CRM Customer info SLA / SLI Cell topology (NIM) ETSI IMEI / TAC db EEA SMS Notifications Customer Surveys DB NPS billing expert analytics summary Award Winning Solution Productized Applications Winner in the Mobile CEM category CTIA 2015, Las Vegas Planning Sales&Marketing I&S Operations Customer Care Advanced Business Logic Networ k Pre-packaged logic: - Data Models - Service KPIs - Patented Data Correlation - Machine Learning SERVICE LEVEL INDEX Resolve & Retain Unloc k ARPU Closed Loop Actions Real-Time Insights Network Device Service Customer Subscrib er E2E Actuation