The Engine Behind Your Social Listening
When you activate a campaign in Socialhose, you don't manually go out and find social media posts—the platform handles that automatically through what we call live search jobs. Understanding how these jobs work gives you insight into when and how your data is collected, helps you troubleshoot issues when they arise, and enables you to optimize your monitoring for better results. This guide explains the complete lifecycle of a search job, from activation to the moment a mention appears in your dashboard.
What Happens When You Activate a Campaign
The moment you click "Activate" on a campaign, Socialhose springs into action behind the scenes. For each platform you selected—Twitter/X, Instagram, Reddit, LinkedIn, Facebook, or TikTok—the system creates dedicated search jobs that will query that platform using your configured keywords.
This isn't a one-time operation. Social listening requires continuous monitoring because new conversations happen every minute. The system creates a structured collection mechanism that will run repeatedly for as long as your campaign remains active, ensuring you capture both historical context and ongoing conversations.
Bootstrap Jobs: Capturing History
The first job that runs for any new campaign is the bootstrap job. This is a special one-time operation designed to give you immediate context by collecting historical data—mentions that already exist on social platforms matching your keywords.
Bootstrap jobs reach back up to 30 days into the past, depending on what each platform makes available. This historical collection serves several important purposes. It establishes a baseline for your monitoring, showing you the existing conversation landscape before you started tracking. It provides immediate value—you don't have to wait days or weeks to see meaningful data. And it helps you validate your keyword configuration by revealing whether your terms capture the right conversations.
Bootstrap jobs typically complete within 15-30 minutes, though this varies based on how much historical content exists for your keywords. A very broad keyword set might take longer because there's more content to process. A highly specific set might complete quickly but return fewer results. After the bootstrap completes, you'll have a foundation of historical mentions in your campaign, and the system transitions to periodic collection.
Periodic Jobs: Continuous Monitoring
After the bootstrap, periodic jobs take over. These run on a schedule—hourly by default—to capture new content as it's posted across social platforms. Unlike the bootstrap, which runs once and completes, periodic jobs run indefinitely as long as your campaign is active.
Each periodic job execution queries the platforms for content posted since the last run. The system keeps track of what it's already collected to avoid duplicates. New content matching your keywords gets ingested, processed through AI enrichment, and appears in your mentions list.
The frequency of periodic jobs depends on your subscription tier. Standard plans might collect hourly, while professional and enterprise plans can collect more frequently—every 30, 15, or even 5 minutes for near-real-time monitoring. If you need faster collection than your plan provides, consider upgrading to ensure you catch time-sensitive conversations.
Each platform has slightly different characteristics. Twitter/X and Reddit tend to have the fastest turnaround because of how their platforms expose content. Instagram and Facebook may have longer intervals due to platform-specific limitations. LinkedIn collection depends on public content availability in professional contexts.
Monitoring Job Status
You can view and manage all jobs for a campaign through the "Manage Jobs" tab. This interface shows you exactly what's happening with your data collection.
Each job displays several key pieces of information. The status tells you the current state: Running means the job is actively collecting data right now; Completed means the last run finished successfully; Scheduled means the job is waiting for its next run time; Failed indicates something went wrong on the last attempt; Cancelled means the job was stopped (usually because you paused the campaign); and Idle means the job is between runs.
The last run timestamp shows when the job most recently executed. If you're troubleshooting missing mentions, this tells you the last time collection happened. The next run timestamp shows when the job will execute again, helping you understand the collection rhythm.
Run count tracks how many times this job has executed since the campaign was activated. This cumulative counter helps you understand overall collection activity. Document count shows the total number of mentions this specific job has collected across all its runs—a useful indicator of how productive your keyword configuration is on each platform.
The Data Flow: From Post to Mention
Understanding the complete journey from a social media post to a mention in your dashboard helps demystify how the system works.
It starts when someone posts content on a social platform that matches your keywords. During the next scheduled job run, Socialhose queries the platform and retrieves this content along with any other matching posts since the last collection.
Each piece of retrieved content goes through a normalization process that converts platform-specific formats into a standardized mention structure. Regardless of whether content came from Twitter, Reddit, or LinkedIn, it ends up in a consistent format with the same fields: content text, author information, engagement metrics, timestamp, and source URL.
The campaign association happens automatically based on which job retrieved the content. Since each job is linked to a specific campaign, mentions automatically inherit that association—there's no keyword matching happening at this stage to determine campaign assignment.
Next comes AI enrichment. Each mention passes through natural language processing that analyzes sentiment (positive, neutral, or negative), extracts named entities (people, organizations, products, locations), detects intent (is this a question, complaint, praise, or informational statement?), and identifies key phrases and topics. This enrichment runs automatically and completes within seconds of ingestion.
Finally, the fully processed mention appears in your campaign's mentions list, ready for you to review, filter, export, or include in alerts.
Understanding Quota Consumption
Your subscription includes limits on various resources, and it's important to understand how search jobs consume these quotas.
Each job execution counts against your monthly search job run limit. If you have 5 campaigns, each monitoring 3 platforms with hourly jobs, you're running 15 jobs per hour—360 per day, roughly 11,000 per month. Higher-tier plans include more generous job limits to support more campaigns and faster collection frequencies.
Each mention collected counts against your monthly mention volume limit. This is separate from job runs—a job might execute and find zero new mentions (if nothing was posted matching your keywords), or it might find dozens in a single run.
You can monitor your current usage in Settings → Subscription. The usage dashboard shows how much of each quota you've consumed in the current billing period. If you're approaching limits, consider optimizing by pausing campaigns you're not actively using, reducing the number of platforms in lower-priority campaigns, or upgrading to a higher tier for more capacity.
Troubleshooting Common Issues
Several issues can affect job performance, and knowing how to diagnose them saves time and frustration.
If jobs aren't running at all, first verify your campaign is in Active status. Paused or draft campaigns don't trigger job runs. Check the job status in Manage Jobs—a Failed status indicates something went wrong, often with temporary platform issues that resolve on subsequent runs. If you've exceeded your quota limits, jobs may skip runs until your next billing period or until you upgrade.
If jobs are running but collecting no mentions, the issue is usually keyword configuration. Keywords that are too specific might not match any content. Try broadening your terms or adding more variations. Also consider that some platforms simply may not have much content for your particular keywords—a B2B software product might have extensive LinkedIn discussion but minimal TikTok presence.
If you're collecting too much irrelevant content, your keywords are too broad. Add exclusion terms to filter out noise patterns you've identified. Review your first batch of mentions to understand what irrelevant content is getting through, then add appropriate exclusions like job postings, unrelated products that share a name, or spam patterns.
If mention volume suddenly dropped, check whether your keywords still match how people discuss your brand. Language evolves, products get renamed, and platform conventions change. A keyword configuration that worked six months ago might need updates to stay effective.
Job Lifecycle and Campaign States
Jobs are tightly coupled to campaign states. When you pause a campaign, all its jobs stop running—no new data is collected during the pause. Existing mentions remain accessible; you just won't get new ones. When you resume the campaign, jobs restart from the current moment. There's no backfill for the paused period—mentions that happened while paused are missed.
If you archive or delete a campaign, its jobs are permanently cancelled. Archived campaign data remains accessible but static. Deleted campaigns and all their data are removed.
When you edit a campaign's keywords or platform selections, the changes take effect on the next scheduled job run. You don't need to deactivate and reactivate the campaign—just save your changes and wait for the next collection cycle.
Understanding live search jobs transforms social listening from a mysterious black box into a transparent system you can monitor, optimize, and troubleshoot. The jobs are your tireless workers, continuously scanning social platforms and delivering relevant conversations to your dashboard. Treating them as a visible, manageable component of your monitoring strategy helps you get more value from your social listening investment.