Topic Performance

Aggregated metrics by topic — called “keyword” in the underlying data source fields below — for topic-level analysis.

Dimensions
Metrics

Field	Type	Description
`date`	Date	Analysis date (YYYY-MM-DD)
`provider`	Text	AI provider name
`keywordId`	Text	Keyword unique identifier
`keyword`	Text	Keyword text

Field	Type	Aggregation	Description
`totalPrompts`	Number	SUM	Number of prompts with this keyword for the provider row
`totalAnswers`	Number	SUM	Total responses analyzed
`answersWithBrand`	Number	SUM	Responses mentioning your brand
`answersWithAnyBrand`	Number	SUM	Responses mentioning your brand or a direct competitor
`totalSelfMentions`	Number	SUM	Total mentions of your brand
`totalDirectMentions`	Number	SUM	Total mentions of direct competitors
`sentimentSum`	Number	SUM	Sum of sentiment scores for your brand mentions
`sentimentCount`	Number	SUM	Count of scored brand mentions
`promptsWithSelfBrand`	Number	SUM	Prompts where your brand appears at least once, across all providers
`totalPromptsAllProviders`	Number	SUM	Total prompts for this keyword/date across all providers
`coveragePromptsWithSelfBrand`	Number	SUM	Normalized prompt numerator used by `coverage`
`coverageTotalPrompts`	Number	SUM	Normalized prompt denominator used by `coverage`
`brandMentionRate`	Percent	Calculated	`SUM(answersWithBrand) / SUM(answersWithAnyBrand)`
`shareOfVoice`	Percent	Calculated	`SUM(totalSelfMentions) / (SUM(totalSelfMentions) + SUM(totalDirectMentions))`
`avgSentiment`	Percent	Calculated	`SUM(sentimentSum) / SUM(sentimentCount)`
`coverage`	Percent	Calculated	`SUM(coveragePromptsWithSelfBrand) / SUM(coverageTotalPrompts)`

The percentage metrics are calculated from raw counters in Looker Studio. This avoids averaging provider rows and keeps topic-level totals aligned with the Qwairy interface.

coverage is calculated at keyword/date level across all providers, then exposed through normalized counters so provider rows do not overweight dates when reports aggregate a date range.

Best For

Keyword performance tracking
SEO-aligned analysis
Topic prioritization
Identifying high-performing topics

Getting Started

Data Sources

Best For

​Best For

Best For