2. Introduction
Customer service team for Retail Banking handles customer queries or
concerns raised through multiple channels – emails, calls, walk-ins, letters,
and faxes.
Barring high priority cases flagged by persistent or irate customers, most
cases aggregated on a periodic basis ranging from daily, weekly, monthly
or annually
The customer service team attempts to identify any patterns such as spikes
or drops in calls based on historical data from the reports
Inferences drawn highly dependent on the team’s experience and
knowledge of the prevailing conditions such as festive season, tax filing
period and so on
3. Challenge
Contact centers receive large streams of data - combination of audio calls and text communication.
Making sense of such largely unstructured data and taking real time action is a major challenge
Traditional analytics reveal trends about data such as calls received, average hold time, average call
duration, resolution rate, inquiry type etc.
Reports are mostly reactive - essentially giving a view of what has already occurred
4. Need
A mechanism that will enable the customer service team to
adopt a proactive approach - alerting it to incidents that might occur in foreseeable future
take pre-emptive measures to tackle any such situations
Resulting in both rapid turnaround time and better decision making capabilities
5. Use Case Scenarios
Industries are using Emerging Customer Service Analytics to
Isolate revenue-related calls or other forms of communication
Identify agent best practices
Identify areas of gaps in knowledge of contact center personnel
Identify cases for personalized agent coaching and training
Predict root cause of customer dissatisfaction
Identify what characteristics of a contact lead to costly repeat communications
Identify other causes of customer churn e.g. better products and services of competitors
Improved operational efficiency - Optimize call handling and first contact resolution
Personalized cross selling and up selling
Source: http://www.cio.com/article/2396132/customer-relationship-management/big-data-analytics-gold-for-the-call-center.html
7. Starting Point on the Analytics Journey
Analyze customer interactions
Email requests and
conversations
Audio calls - search
by keywords
CRM Data
Analyze customer transactions
Relationship data
Portfolio data
Transaction data
9. Big Data powered Analytics Platform
Audio
Calls
IVR
SMS
IVR
Email
Original Customer Data
Audio to Text Mining
Data Mining, Storage and Analysis
Big Data
Audio Mining Platform
• Linguistic Analysis
• Intention Analysis
• Dependency Analysis
• Trend Analysis
• Text Mining
Departmental
Regulatory
Management
Reports
Stakeholders
Audio mining platform to convert audio to text.
Big Data Analytics solution
Uses transactions and interactions data to derive correlations and dependencies
Reveals trends and patterns to alert team and direct focus on potential situation(s).