Use Volume as a comparison tool
Volume can be approached like a pivot-table creator for qualitative data.
Before configuring the view, decide:
Which concepts do I want to measure?
Which records should count toward the calculation?
How should the results be compared or displayed?
For example, a product team monitoring crashes, login and password could compare those concepts by app version or operating system. This may reveal whether a growing issue is widespread or mainly associated with one technical environment.
Understand how top concepts are selected
When displaying top concepts, Journey Insights Hub considers how distinctive they are within the text analysis project. The number of concepts selected affects the level of detail shown.
A smaller selection, such as Top 10, provides a focused overview of the most distinctive concepts. A larger selection, such as Top 500, allows more specific and nuanced concepts to appear.
For this reason, Top 10 may not always contain the same concepts as the first 10 results displayed when Top 500 is selected. Changing the number of concepts can affect the selection itself, rather than simply adding more concepts to the existing list.
Discover emerging concepts
Use Top concepts by prevalence under Configure to explore prominent language across the current selection. When filters are active, use Unique to Filter to identify concepts that are particularly distinctive to the selected group.
For example, you could use Unique to Filter to find concepts that stand out among:
Customers using a particular app version
Responses with low satisfaction scores
Feedback submitted after a product release
Customers from a particular region or segment
Concepts that have not been saved display a blue + icon. Select the icon to add a useful concept to a concept list for continued monitoring.
Monitor an established set of topics
Create or select a concept list containing the topics your team regularly tracks, then display Active concepts in Volume.
This is useful for recurring reporting because the same definitions and colors can be reused across analysis periods.
You might organize concepts by:
Product experience
Customer-service issue
Journey stage
Business priority
Strength, weakness, opportunity, or threat
If you add concepts discovered in the visualization, save the concept list before leaving the text analysis project.
Compare Volume across metadata
In Configure, Volume provides several feature-specific display options:
Overall concept volume compares total reach across the current selection
Over time tracks changes using a date field
By category compares groups such as products, regions, channels, departments, or customer segments
By score value compares concept volume across individual ratings
Across a numeric range groups a numeric field into more readable ranges
A category comparison can place products or customer groups side by side, making it easier to identify which concepts are shared and which are concentrated within one group.
For example, if mentions of crashes increase over time, compare the concept by:
App version to determine whether the increase is associated with a particular release
Operating system to determine whether the issue is concentrated among Android or iOS users
Compare explicit and broader discussion
In Configure, include conceptual matches when you want to understand the broader conversation around a topic.
Exclude them when the question specifically requires explicit mentions. For example, a brand-monitoring question may require the exact company name, while an experience analysis may benefit from related language that customers used instead. The default bar colors distinguish exact and conceptual matches. Concept or sentiment colors can be used when they better support the analytical story.
Check the percentage denominator
When filters are active, Volume can calculate percentages using:
The records included in the current filters - Use the filtered records when asking, “What percentage of this group discusses the concept?”
All records in the text analysis project - Use all text analysis project records when asking, “How much does this filtered group contribute to the text analysis project’s total conversation?”
The same match count can produce a different percentage depending on the denominator, so verify this setting before comparing or reporting results.
Read more: Use Configure Visualization, Concepts, Filters and Records.
Interpret Volume carefully
High volume alone cannot establish that a topic is positive, negative, important, or responsible for an outcome. A common concept may be expected background language, while a lower-volume concept may represent an emerging risk or specialized need.
Use Galaxy to explore the surrounding language
Use Drivers to see whether the concept is associated with a numeric outcome
Use Sentiment to understand how it is discussed
Use Records to confirm what people actually meant
The strongest Volume findings combine reach, comparison and source context.