Through Chattermill's table chart view, you can now apply up to three levels of breakdown to your data. This enables you to uncover more precise insights about your customers, by displaying how different factors are performing within one another.

After selecting the table view from the top right-hand corner, click on the left-hand sidebar to select a breakdown, using any of the metadata available to you.

In this instance, let's say we're a regional manager looking to get insights into how our organization's stores are performing regionally. The first breakdown, 'Country', gives us a broad look into how our stores are performing at the national level.

Whilst this gives us an outline of the story, it doesn't show us the full picture. By selecting a second breakdown, 'store name', we can segment our data more precisely.

This generates a table which allows us to compare how individuals stores are performing within each region. This is useful as it allows us to spot whether the scores from the regional view were the result of a general trend in that area, or down to a few underperforming or overperforming stores.

What's more, is you can segment your data further by adding a third breakdown - let's say by the theme 'Product: Quality' - to see how your product quality fares across each store. Multiple breakdowns can be endlessly customised to suit your business - just select the breakdowns that are most relevant to you.

Here's some more examples of how you can use multiple breakdowns to uncover nuanced insights from your data.

Contact Centre Agent Performance

Contact centre managers often need to monitor performance at both a team and individual level. By first viewing overall sentiment by team, they can then breakdown by individual agents to see who is driving positive or negative feedback. If one agent is regularly receiving low sentiment scores, the manager might step in to provide additional training and improve the centre’s overall performance.

Product Teams

To assess how different versions of an app or product are performing across various devices, Product Managers can break down customer sentiment by app version, comparing themes between iOS and Android users, or even across different device types. This helps them pinpoint whether a particular version is causing issues, or if one platform consistently delivers a better experience.

Supplier Sentiment Analysis

For an organization that sells products supplied by other companies, it’s crucial to understand how each supplier is impacting overall customer sentiment. For instance, a travel agent might break down customer feedback by airline, revealing whether certain airlines are consistently generating negative reviews or compliments. This helps their focus on working with or improving specific suppliers, such as addressing issues with a poorly performing airline or promoting a high-performing one to boost their customer satisfaction.

While using multiple breakdowns can provide deep insights, it’s important to be aware of some limitations to ensure you get the most out of this feature.

Maximum Number of Elements

To maintain optimal performance, there’s a cap on the total number of data elements that can be displayed when using multiple breakdowns. The system tries to return at most 10,000 elements in total. This means that the more breakdowns you apply, the fewer elements will be shown for each breakdown.

Breakdown Limits Based on Number of Breakdowns

Here’s how the limits generally work:

• One Breakdown: Up to 100 elements can be displayed.

• Two Breakdowns: Each breakdown can display up to 21 elements.

• Three Breakdowns: Each breakdown can display up to 10 elements.

These are worst-case scenarios. If you’re using breakdowns that naturally have fewer unique values (like ‘NPS Answer’, which has 3 categories), the limits for other breakdowns might be higher.

Influence of Filters and Historical View

The number of filter groups you apply and whether the ‘Historical’ view is enabled can also affect the maximum number of elements displayed. Applying multiple filters or viewing historical data adds complexity, which may reduce the number of elements the system can return.

Individual Breakdown Caps

Some breakdowns have their own specific limits:

• Segment: Capped at 500 elements independently.

• Score Breakdown: Typically produces 5–10 values.

• NPS Answer: Produces 3 values.

• Data Source & Data Type: Generally produce only a few values.

Tips to Optimise Your Analysis

• Prioritise Key Breakdowns: Use the most important breakdowns first to ensure they’re fully represented in your analysis.

• Apply Specific Filters: Narrow down your data with filters to focus on the most relevant information and stay within element limits.

• Monitor Data Volume: If not all expected data is displayed, consider reducing the number of breakdowns or adjusting your filters.

• Leverage Breakdowns with Fewer Values: Combining breakdowns that have fewer unique values can help avoid hitting the element cap.

By keeping these limitations in mind, you can effectively utilise multiple breakdowns to gain deeper insights without overwhelming the system.