Insight Narrator is a key component of the sandsiv+ AI suite, utilizing advanced Large Language Model technologies to transform individual feedback into actionable insights on specific topics and trends. It provides rapid decision intelligence, enabling organizations to respond swiftly to customer and employee feedback. For CX consultants, it enhances internal capabilities and streamlines client project management, boosting both efficiency and effectiveness.
As a sandsiv+ feature, Insight Narrator works with VoC Mine queries, which interface now contains Insight Narrator-related elements.
In the screenshot of a query interface, two new elements can be found: Query context in the query's Info section and Insight Narrator button in the query's output section.
Query Context is a generalized text about feedback records of the query. Query Context allows Insight Narrator's LLM model to provide more accurate results. Following the example above, the query contains feedback records that were received from the theme park visitors after their visit, which was reflected in the Query context field. There are no strict limitations for the Query Context, meaning that it could vary from "Swiss insurance company" to "Feedback records received from users of the banking application after a Call Center interaction.", as both these contexts may be useful depending on what type of insights or analysis is being conducted.
To access Insight Narrator, please click on the Insight Narrator button in the query's output section.
When accessing Insight Narrator for the first time in the query, you will see analysis sections Text Analytics, Frequency Analysis, Co-occurrence Analysis, and Correlation Analysis that are empty.
To run Insight Narrator and let it populate the aforementioned analysis sections, click the "Prompt" button on the bottom-left. In the window that will be opened after a click, it will be possible to set or change the Query Context, select the language to get insights in, and choose what type of analysis the Insight Narrator should perform.
Please see the explanation of every kind of analysis below.
-
Text Analytics
-
Semantic Analysis
Semantic Analysis is the process of understanding the meaning and context of text by analyzing language patterns and structures. In Customer Experience, semantic analysis helps identify trends, sentiments, and key themes in customer feedback, providing deeper insights into customer needs and preferences. This allows businesses to improve products, services, and overall customer satisfaction by addressing specific concerns and enhancing positive experiences. -
Clustering Analysis
Clustering Analysis is a technique that groups similar data points together based on their characteristics. In Customer Experience, clustering analysis helps segment customer feedback into distinct groups, revealing patterns and common issues. This allows businesses to tailor their responses and improvements to specific customer segments, enhancing personalized experiences and addressing the unique needs of different customer groups. -
Topic Detection and Sentiment Analysis
Topic Detection and Sentiment Analysis involve identifying key subjects in text data and assessing the emotional tone behind the words. In Customer Experience, these techniques help businesses understand what customers are talking about and how they feel about it. This provides actionable insights into customer concerns and sentiments, enabling companies to address issues promptly, enhance positive experiences, and improve overall customer satisfaction. -
Customer Journey Analysis
Customer Journey Analysis is the examination of a customer's complete experience with a brand, from initial contact to post-purchase interactions. In Customer Experience, this analysis helps identify critical touchpoints and pain points throughout the customer lifecycle. By understanding the customer's journey, businesses can optimize each stage, enhance customer satisfaction, improve retention rates, and create a more seamless and enjoyable experience. -
Pros & Cons
Listing the top pros and cons mentioned by clients involves systematically identifying and summarizing the most frequently praised and criticized aspects of a product or service. In Customer Experience, this process helps businesses understand what customers value most and what issues need urgent attention. By focusing on these key areas, companies can enhance strengths, address weaknesses, and ultimately improve customer satisfaction and loyalty. -
Improvements suggestions
Listing improvement suggestions based on the difficulty of implementation involves categorizing feedback into actionable items that range from easy to hard to execute. In Customer Experience, this approach helps prioritize efforts by addressing simpler, quick-win improvements first while planning for more complex, long-term changes. This ensures efficient resource allocation and steady progress in enhancing customer satisfaction and overall service quality. -
SWOT Analysis
SWOT Analysis is a strategic planning tool used to identify and evaluate the Strengths, Weaknesses, Opportunities, and Threats related to a business or project. In Customer Experience, conducting a SWOT analysis helps businesses understand their internal capabilities and external environment. Strengths and weaknesses focus on internal factors, such as service quality and customer support, while opportunities and threats address external factors like market trends and competitive actions. This comprehensive view aids in developing strategies to leverage strengths, address weaknesses, seize opportunities, and mitigate threats, ultimately enhancing the customer experience. - TOWS Analysis TOWS is a strategic planning tool closely related to SWOT analysis, used to identify and develop strategies by matching internal strengths and weaknesses with external opportunities and threats. The TOWS matrix takes a more action-oriented approach by focusing on strategic options and combinations rather than listing factors, as the SWOT analysis does.
- Taxonomy Taxonomy provides a detailed overview of the topic classification structure, organized in two distinct levels: Categories (1st Level), for broad themes and domains, each branching into more specific Topics (2nd Level). Categories cover the main areas of focus relevant to the feedback (e.g., “Product Quality,” “Customer Service,” “User Experience”). Under each Category, Topics are identified that refine these areas into specific subjects (e.g., within “Product Quality,” topics might include “Durability,” “Materials,” and “Design”). The resulting structure provides a detailed structure used to help build TopicSets or classification models feedback by main themes, with a secondary layer to highlight particular areas of interest within each theme.
-
Semantic Analysis
-
Charts Analytics
-
Frequency Analysis
Frequency Analysis involves examining how often specific themes, words, or phrases appear in customer feedback. In Customer Experience, this analysis helps identify the most common issues, concerns, and praises from customers. By understanding the frequency of these mentions, businesses can prioritize areas that require attention, make data-driven decisions, and focus on the most impactful improvements to enhance customer satisfaction and loyalty. -
Co-occurrence Analysis
Co-occurrence analysis examines how often specific words or phrases appear together in customer feedback. In Customer Experience, this analysis helps identify patterns and relationships between different aspects of the customer experience. By understanding these co-occurrences, businesses can uncover hidden insights, such as common issues linked to particular products or services, and discover underlying causes of customer sentiments. This enables more targeted and effective improvements to enhance overall customer satisfaction and address critical areas of concern. -
Correlation Analysis
Correlation Analysis involves examining the statistical relationships between different variables in customer feedback. In Customer Experience, this analysis helps identify how various aspects of customer feedback are related. For example, it can reveal how product quality changes correlate with changes in customer satisfaction. By understanding these correlations, businesses can pinpoint key drivers of positive and negative experiences, enabling them to make informed decisions to improve products, services, and overall customer satisfaction.
-
Frequency Analysis
When the needed analyses are selected, the "Run Narrator" button will become yellow, which means that Insight Narrator is ready to generate insights for the query. After the button is pressed, the analysis sections will be populated with insights that were selected.
Chart narration for “Frequency” “Co-Occurence” and “Correlation”
When Insight Narrator is activated, the user may also see additional insights related to Frequence, Co-occurence and Correlation tabs. Green button tab is shown when opening those tabs and when clicked it gives the enriched narration of what is shown in the chart.
In Frequency page the user gets a report that analyzes frequently mentioned words, key themes and actionable insights from the data provided.
In Co-occurrence page the user gets an analysis report, which primary objective is to identify significant term co-occurrences, assess their relationships, evaluate sentiment associations, and derive actionable insights.
In Correlation page the user gets a graph that reveals central nodes that represent core aspects of an analysed data.
Query action output explained:
- Cloud - shows a number of most frequently appearing lemmata. In this view, it is possible to add lemmas to the Lemmata to include or Lemmata to exclude filters by clicking on the needed lemma.
- Text - shows every text case that appears in the query separately. In this view, it is possible to see text case's lemmata, associated metadata, and other details, such as classification results.
- Chart - shows the list of most frequently used lemmata along with a frequency of appearing in the resulting query. It is also possible to add any of the listed lemmata to the Lemmata to include or Lemmata to exclude filters by clicking on Exclude or Include buttons against the needed lemma.
- Co-occurence - shows the number of times words occurred together in one text.
-
Correlation - shows how much more likely a pair of words will occur together in a sentence than separately
Positive correlation - there are more cases when words either appear together in one sentence or do not appear at all.
Negative correlation - there are more cases that they will appear separately than together or not at all.
In MINE, word pairs with positive correlation are displayed only. -
Data labelling - a technology that uses unsupervised models to provide automation in the creation of training-data, without the need for manual training or topic identification. More via manual here.
Comments
0 comments
Please sign in to leave a comment.