Skip to main content

Common error scenarios

This topic covers well-known error scenarios when querying your data.

Starting assumptions

For each example query below:

  • The sample dataset is used

  • The date is July 31, 2021

  • Default Time Filter = Latest month

  • Default Currency = US Dollar (USD)

 

No measure

Your query must include a measure — that is, a numeric, quantitative field in your data.

Query

Answer

What is the total for Nutrali?

I'm sorry, I can't answer a question without a measure. Please try to ask a question with a measure such as Profit or Sales and try again.

Here are two ways you could validly rephrase the query:

Query

Answer

What is the Sales total for Nutrali?

In July 2021, Sales for Nutrali are $3.07 million.

What is the Profit total for Nutrali?

In July 2021, Profit for Nutrali is $0.57 million.

 

Top

Multiple measures

Your query can't contain more than one measure, unless it's a target-based variance query.

Query

Answer

What are my Sales and Profit for Nutrali?

I’m sorry, I couldn't find a comparison phrase in your question. I can’t handle multiple measures (Profit and Sales) except in a comparison question (Profit vs Sales). Please ask a comparison question or a question with only one measure.

You could break the query above into two separate queries.

Query

Answer

What are my Sales for Nutrali?

In July 2021, Sales for Nutrali are $3.07 million.

What is my Profit for Nutrali?

In July 2021, Profit for Nutrali is $0.57 million.

Or you could rephrase the original query to make it a target-based variance query.

Query

Answer

Tell me about Sales versus Profit for Nutrali.

In July 2021, Sales for Nutrali are higher than Profit by $2.29 million (81%).

Note

A target-based variance query — also known as a comparison question — can handle a maximum of two measures.

 

Top

Multiple dimensions

Your query can't contain more than one dimension.

Query

Answer

Tell me about Sales by Product and Segment.

I'm sorry, I can't handle multiple dimensions (Product and Segment). Please ask a question with only one dimension.

To learn about Sales by Product and Sales by Segment, you must ask separate queries.

Query

Answer

Tell me about Sales by Product.

In July 2021, the top Product is Glower with $3.27 million (26%), and the bottom Product is Ointmo with $0.71 million (6%) of Sales.

Tell me about Sales by Segment.

In July 2021, the top Segment is Departmental with $6.47 million (51%), and the bottom Segment is Channel Partners with $135,000 (1%) of Sales.

 

Top

Unsupported aggregation type

You can't use a measure when it's configured to use an unsupported aggregation type.

Assuming Sales is configured to use Count aggregation:

Query

Answer

Tell me about my Count(Sales).

I'm sorry, I can't answer queries using measures with an unsupported aggregation type. The supported types are none, average, max, min, and sum. Please review each measure's aggregation type and try again.

To make this query work, configure the measure to use a supported aggregation type and try again.

Important

Currently, the supported types are Sum, Average, Minimum, Maximum, and Don't summarize.

Assuming Sales is reconfigured to use Sum aggregation:

Query

Answer

Tell me about my Sales.

In July 2021, Sales are $12.75 million.

 

Top

Wrong aggregation type for query

The measure's aggregation type must be the correct type for the query.

Assuming Sales is configured to use Minimum aggregation:

Query

Answer

What are total Min(Sales) for Nutrali?

I'm sorry, I can't perform a sum query using a measure aggregated by min. Please enter a different query or change the measure’s aggregation type and try again.

The word total makes this a sum-aggregation query, so change the measure's aggregation type to Sum and try again.

Assuming Sales is reconfigured to use Sum aggregation:

Query

Answer

What are total Sales for Nutrali?

In July 2021, Sales for Nutrali are $3.07 million.

Note

See Query types for guidance on the required aggregation type for each type of query.

 

Top

Multiple instances of same dimension

You can filter your query using dimension instances, but you are restricted to one instance per dimension.

Query

Answer

Tell me about Sales in Germany and Mexico?

I'm sorry, I can't handle multiple instances from the same dimension (Germany and Mexico). Please ask a question with only one Country instance.

Both Germany and Mexico are instances of the same dimension (Country), so you must request the information in separate queries.

Query

Answer

Tell me about Sales in Germany?

In July 2021, Sales in Germany are $3.67 million.

Tell me about Sales in Mexico?

In July 2021, Sales in Mexico are $2.02 million.

You can have multiple instances of the same dimension when they form an exclusion filter.

Query

Answer

Tell me about Sales but exclude Germany, Mexico and France?

In July 2021, Sales excluding Germany, Mexico and France are $3.85 million.

 

Top

Dimension plus instance of that dimension

Your query can't include both a dimension and an instance of that same dimension.

Query

Answer

Tell me about Sales by Country for Germany?

I'm sorry, I can only answer a question where the group and filter are different. You could try asking about either Germany or Country.

The query above returns an error because it includes Germany, which is an instance of the dimension used to group the data.

Valid ways of rephrasing the question include:

Query

Answer

What are my Sales in Germany?

In July 2021, Sales in Germany are $3.67 million.

Tell me about Sales by Country?

In July 2021, the top Country was Germany with $3.67 million (29%), and the bottom Country was the United States of America with $1.54 million (12%) of Sales.

Tell me about Sales of Glower by Country?

In July 2021, the top Country is Germany with $0.99 million (30%), and the bottom Country is Canada with $398,000 (12%) of Sales for Glower.

 

Top

Currency measure vs. non-currency measure

A target-based variance query must compare two currency measures or two non-currency measures.

Query

Answer

Tell me about Sales versus Units Sold?

I’m sorry, I can’t compare a currency measure (Sales) with a non-currency measure (Units Sold). Please ask a question comparing either two currency measures or two non-currency measures.

This error occurs because Sales is a currency measure and Units Sold is a non-currency measure. One response is to accept this query is invalid and enter a different one. For example, you could write a query comparing two currency measures:

Query

Answer

Tell me about Sales versus Target Sales?

In July 2021, Sales are lower than Target Sales by $0.53 million (4%).

To compare Sales and Units Sold, you must first reconfigure the data attributes to remove the currency vs. non-currency conflict. For example, the first example query in this section would work if you made Sales a non-currency measure.

 

Top

Two time points but no time interval

A query containing two time points is only valid when the time points combine to make a time-interval filter.

Query

Answer

Tell me about Sales by Product for January and February?

I'm sorry, I can’t currently answer that. You could try asking about either January or February.

You could split this query into two queries (one for January, the other for February) or turn January and February into a time-interval filter.

Query

Answer

Tell me about Sales by Product between January and February?

From January 2021 to February 2021, the top Product was Karateon with $6.07 million (58%), and the bottom Product was Nutrali with $97,000 (1%) of Sales.

 

Top

Time-interval filter with identical time points

A query containing a time-interval filter must contain two distinct time points.

Query

Answer

Tell me about Sales of Glower in France between January and January?

I'm sorry, I can't understand a time interval where start and end periods overlap. You could try a question with two distinct time points.

To fix this, make the time points more specific and therefore distinct.

Query

Answer

Tell me about Sales of Glower in France between January 2020 and January 2021?

From January 2020 to January 2021, Sales for Glower in France were $2.99 million.

 

Top

Drilldown query with no available dimension

A drilldown variance query must leave at least one dimension available for drilldown analysis.

Query

Answer

Why did Departmental Sales of Karateon in France change in Q1 year on year?

I'm sorry, I can't provide drivers or offsets since there is no dimension available to drill down further.

This error occurs because there are three dimensions in the sample dataset (Country, Product, and Segment) and an instance of each (France, Karateon, and Departmental) in the query. To fix this, rewrite the query so at least one dimension is available for drilldown analysis.

Query

Answer

Why did Sales of Karateon in France change in Q1 year on year?

Sales for Karateon in France increased by $0.51 million (43%) in Q1 this year compared to the same period last year, driven by Departmental and Brand Store and offset by Chemists.

There is no instance of the Segment dimension in this revised query; therefore, that dimension is available for drilldown analysis.

 

Top