Skip to main content

Overview

Cached Jobs are the execution units for your brand monitoring. When you create queries, MentionLab generates cached jobs that run against multiple AI models to track how your brand is mentioned in AI-generated responses.

Cached Job Properties

PropertyTypeDescription
idstring (UUID)Unique identifier
projectstring (UUID)Parent project ID
query_idstring (UUID)Associated query ID
querystringThe search term being monitored
aiModelsstring[]AI models used for analysis
executionCountnumberNumber of executions per run
languagestringISO 639-1 language code
countrystringISO 3166-1 alpha-2 country code
statusstringCurrent job status
createdAtstringCreation timestamp
updatedAtstringLast update timestamp

Job Status

Cached jobs progress through different states:
StatusDescription
pendingJob is queued and waiting to execute
activeJob is currently running
endedJob (job-executions) has completed - may contains failure, check job-executions

AI Models

Cached jobs can run against multiple AI providers:
Model KeyDescription
openaiOpenAI GPT-4 models
openai5OpenAI GPT-5 models
anthropicAnthropic Claude models
google-generative-aiGoogle Gemini models
perplexityPerplexity AI
xaixAI Grok models
ai-overviewGoogle AI Overviews
ai-modeAI search modes

How Cached Jobs Work

1

Query Creation

When you create a query, MentionLab automatically generates a cached job for it.
2

Scheduling

Based on your project’s recurrence settings, jobs are scheduled to run at specified intervals.
3

Execution

The job runs against selected AI models, capturing responses that mention your brand.
4

Analysis

Results are analyzed and stored as Result Analysis records for you to review.

Execution Count

The executionCount property determines how many times each job runs per scheduled execution. Higher counts provide more comprehensive data but consume more credits.