What is an Insight?
An insight is a valuable understanding gained from analyzing customer conversations, feedback, and behavior. It involves identifying patterns or recurring themes in customer inquiries or requests. These insights can reveal customer preferences, pain points, or expectations, enabling the business to improve service, streamline processes, or develop targeted solutions that enhance the overall customer experience.
Convirza’s insights can help companies keep track of agent and team performance, identify potential areas of improvement, and provide a solid foundation for coaching. These insights can be used to filter conversations, conversation scoring, and reports and can also be used as part of scorecards. Convirza provides four types of insights.
- AI
- Ultimate AI
- Keyword
- Manual
AI Insights:
AI Insights leverage Convirza’s advanced analysis of over a billion calls, providing a robust, data-driven understanding of customer interactions. These insights are built on extensive language models and patterns identified across diverse call scenarios, automatically enabling the system to detect critical phrases and behavioral trends. With this vast dataset, AI Insights continuously improves, delivering accurate, actionable information that helps businesses understand customer needs, monitor performance, and uncover opportunities for growth and optimization.
Convirza's AI-powered insights analyze conversations to provide specific, actionable feedback. Each insight is evaluated with two outcomes: "Pass" or "Fail." When the predetermined criteria for an insight are met during the conversation, it is marked as "Pass." Conversely, if the criteria are unsatisfied, the insight is marked as "Fail." This streamlined process ensures accurate and efficient performance assessment.
Here is a list of Convirza’s AI and classification criteria with descriptions and examples.
"Classification" refers to sorting or categorizing conversations of similar characteristics based on predefined criteria to help businesses better organize, analyze, and act upon various customer interactions.
More information can be found here: Classifications.
Ultimate AI Insights: Transform Your Business with Customized Insights
What is Ultimate AI?
Ultimate AI is a fully customizable, state-of-the-art insights tool that leverages cutting-edge technology to help you achieve a deeper understanding of your data. Designed with flexibility, Ultimate AI can be tailored to meet your business needs, ensuring you gain the precise insights you seek.
How Does Ultimate AI Work?
An Ultimate Insight is a deep understanding of vast customer conversation data analyzed using advanced AI models like OpenAI. Large Language Models (LLMs) can process and interpret complex, unstructured language data from phone calls, extracting nuanced information, such as sentiment, intent, and key themes, at scale. By applying Ultimate Insights, businesses can uncover insights into customer emotions, emerging trends, and specific needs that may go unnoticed. This empowers companies to anticipate issues, personalize interactions, and continuously refine products or services based on real-time, data-driven insights from customer conversations.
Using advanced OpenAI technology, Ultimate AI processes and analyzes conversations seamlessly, delivering meaningful insights that are both comprehensive and actionable. Its transparency sets it apart — every score provided comes wth a clear, easy-to-understand explanation, empowering you with data and context.
- Ultimate insights can be incorporated into a scorecard and combined with other insight types to create a fully customized assessment tool. Your scorecard can include a mix of AI, Ultimate, Manual, and Keyword insights for a comprehensive, tailored evaluation.
- Ultimate Insights can function as a standalone insight, independent of any scorecard, and can be applied as an insight filter across various pages like Conversations, Conversation Scoring, Reports, and more. As an individual insight, it will be scored separately and displayed under "Other Insights" on the Conversation Details page, as illustrated below.
You can tailor ultimate insights to suit your specific businss needs. This customization allows you to define metrics, identify key patterns, and generate valuable insights based on the unique aspects of your industry or customer interactions.
For example:
- Customer Satisfaction: You can customize insights to track specific words or phrases that indicate customer satisfaction levels. By prompting the system with phrases like "helpful," "satisfied," or "disappointed," you can generate insights that reveal overall customer sentiment trends.
- Sales Opportunities: Tailor insights to identify language that suggests a buying interest, such as "pricing," "trial," or "package options." This allows your team to spot potential leads or upsell opportunities more accurately.
- Issue Resolution: Set prompts to capture words like "problem," "help needed," or "support," allowing your team to focus on areas where customers face challenges. This way, you can identify patterns in service requests and address recurring issues efficiently.
- Give-a-Shit Indicator: The Give-a-Shit Indicator is a unique metric to measure a team member's genuine engagement and care during a customer call. This indicator assesses the entire conversation to determine whether the team member displayed an attentive, empathetic, and proactive attitude toward the customer’s needs and concerns. Using advanced analytics, the Give-a-Shit Indicator evaluates factors such as tone, responsiveness, active listening, problem-solving approach, and overall enthusiasm, providing an objective score that reflects how much the team member “gave a shit” about delivering a high-quality, customer-focused experience. This insight helps businesses understand and improve team commitment to customer care, ensuring that calls meet service standards and resonate with customers on a personal level.
These tailored insights will give you a deeper understanding of customer needs, allowing you to respond more effectively and continuously refine your approach.
You can also refine your settings and check the insight on a few calls to see how the ultimate insights perform.
During the creation phase, we collaborate closely with you to understand your requirements, develop insight, and enhance performance by:
- Refining Keywords and Phrases: Adjust specific keywords or phrases that trigger insights. Testing these on a small batch of calls lets you see if the insights accurately capture relevant conversations and refine any key details that are too broad or missing.
- Testing Scoring Criteria: If your insights are linked to performance metrics, you can adjust the scoring criteria to ensure they align with your goals. By analyzing a sample, you can confirm if scores reflect the actual call quality or need further calibration.
- Evaluating Insight Accuracy: Run initial tests to confirm if your insights highlight the intended customer sentiments, issues, or behaviors. This helps identify unintended results early on and fine-tune them for better accuracy.
By optimizing settings through small-scale testing, we can enhance the accuracy and relevance of insights, ensuring that they provide valuable, actionable data when fully implemented across all conversations.
Why Choose Ultimate AI?
- Customizable Insights: Ultimate AI adapts to your requirements, offering insights that align with your goals.
- Advanced Technology: Powered by the latest OpenAI innovations, Ultimate AI handles complex data efficiently.
- Transparent Scoring: Understand the numbers and reasoning behind every score for more intelligent decision-making.
Get Started Today
Are you interested in building your own Ultimate AI insights? Our dedicated Convirza customer service team is here to assist you. Contact us at support@convirza.com, and we'll guide you through the process from start to finish.
Keyword Insights:
What is a Keyword Insight?
A keyword insight is an understanding gained from analyzing frequently used words or phrases in customer conversations, especially during phone calls. By identifying recurring keywords, businesses can detect trends in customer needs, common questions, or specific product interests. This type of insight helps companies recognize what matters most to customers, allowing them to tailor services, address frequent issues, and improve communication to better align with customer expectations.
The system automatically scores the Keyword insights when analyzing conversations. The calls over 30 seconds are sent to the system, and the insights are scored based on ‘Pass or Fail’ criteria.
To create a Keyword Insight, navigate to Insights–>Click Create in the right-hand corner.
Add an insight name, Keywords, or phrases to decide if the insight is Hit or Miss.
These insights can be edited, removed, or added. A field to select a speaker (Agent Caller or Both) and a field to enter the condition are available. The keyword Insight can be considered Hit/Miss, depending on these parameters.
“Importance“ will be excluded on this page as it's not a scorecard flow. Once the insights are created and active, they cannot be edited. However, the insight's status can be made inactive or active.
If an insight is part of one or more scorecards, the active/inactive toggle is greyed out, and a tooltip will appear with a message that the insight cannot be deactivated. There is a duplicate button to duplicate insights.
Negative keywords will permanently mark the insight as a hit if they are found in the conversation. For example, if three positive and one negative keywords are found in the conversation, the insight's result will always be marked as ‘Miss.’
Once the keyword insight is created, it will be available under insights filters on Conversations, Conversation Scoring, and the reports. You can also add a conversation action based on the keyword insight.
Manual Insights:
Convirza offers manual insights that are customizable as per your requirements. Manual insights can be built using the create/edit scorecard flow. We can use existing manual insights or develop new insights for the scorecard.
Manual insights can be created with the insight name Help text, which will help the scorer understand how to score the insight and the possible outcomes.
The following outcome types are available.
- Pass/Fail
- Scale 0-3
- Scale 0-5
- Scale 0-10
We can set up the importance from 1 to 10.
All the manual insights need to be scored manually. The clients can do the scoring themselves, or Convirza can provide a team to score the insights manually for them.
Please note:
- Except for manuals, all insights have outcome types of Pass/Fail and N/A (if applicable), which the system automatically determines.
- Manual insights have scale and Pass/Fail outcome types, which are scored manually by the scorers.
- AI insights are predefined and part of the standard scorecard.
- Keyword Insights are determined to be Pass/Fail depending upon the exact keyword/phrases said in the conversation.
- Ultimate AI insights are determined as pass/fail depending on how the insight is defined and how the conversation context matches the defined insight.
- You can delete Draft and Inactive insights except for the AI insights.
Once the insights, AI, Keyword, and Ultimate AI are created, they will be available as triggers for the conversation actions and insight filter on Conversations, Conversation actions, and reports.