Percent of total questions are a valuable way to better understand your data. For example, a call center manager might want to see what percent of calls are resolved in under 5 minutes for different cities. This can help managers to see where they might need to hire more employees to decrease customer wait time or to provide additional training on common issues to decrease the resolution time.
Available now, Q&A supports a broader range of percent of total question intents including:
- Filtering on a specific value
- Asking questions with a distinct count field
- Asking questions without needing to explicitly group by a field
Let’s cover some examples following a financial analyst at a card payments company.
Filter on a Specific Value
Percent of total questions now support filtering on a specific value.
The marketing team reached out asking if running an ad campaign on eating out and traveling would have a meaningful impact. I ask Q % of revenue last year for restaurants or travel and I can see that restaurants and travel make up a significant 40% of total revenue last year. From the bar chart on the right, I can see that restaurants is the top contributor at 30% and travel as the 4th highest at 10%. I share this answer with the marketing team and let them know that they might want to consider strategies for targeting the grocery store and fast food merchants as well.
Distinct Count Field Support
Percent of total questions now support distinct count fields like Transaction ID in this case.
Next, the product management team wants data to help prioritize their purchase method roadmap. I ask Q % of transactions by payment method and I can see that Recurring, Online, and Contactless together contribute to about 60% of total transactions. I share this answer with the purchase methods product team.
Ask Questions Without Explicitly Grouping by a Field
Percent of total questions now support asking questions without needing to explicitly group by a field.
For example, previously questions needed to always include the group-by field, like “% of revenue for US-based Visa holders by network”. It is not intuitive to add the field, so now Q identifies the correct field based on the filter values. Now, I can ask Q % of revenue for US-based Visa holders and see US-based Visa card holders make up 17% of total card holders.
Visual Chart Selection
As you’ve likely noticed in the above answers, Q also automatically picks the most relevant chart type based on the number of values in the answer. For example if there are fewer than 6 values, Q displays as a donut chart with a KPI in the center as donuts are easy to visually interpret parts of whole. You can always manually change the chart type from the visual menu or specify with natural language! Then save to your pinboard for future access.
