Social Listening Pipeline

Automated social media research → client-ready reports

Sign Out
Pipeline
Saved Briefs
Run History
Help

Quick Load

Client Brief

Platforms

Influencers

Report Context / Vision

Budget

Split evenly across platforms. 70% discovery, 30% enrichment (transcripts + comments).
Loading...
Loading...

How It Works

The pipeline runs 8 stages automatically. You fill in a brief, hit Run, and get a client-ready report with trends, audience insights, content opportunities, and creator spotlights.

1-2
Brief & Strategy
3-5
Scrape & Enrich
6-7
Review & Research
8
Final Report

Brief Fields Guide

Client Name

The brand or company you're researching. Used in the report header and to give the AI agents context about the brand.

Example: H&M, Nespresso, The Ordinary
Category

The market category or niche. This shapes what the AI looks for in the data — trends are reported relative to this space.

Example: fast fashion, specialty coffee, skincare, home fitness
Hashtags

Comma-separated hashtags the pipeline will search for on each platform. Include the brand hashtag, campaign hashtags, and 2-3 category hashtags. More hashtags = more data scraped = higher Apify cost.

Example: #hm, #hmfashion, #hmhaul, #fastfashion
Tip: 5-10 hashtags is the sweet spot. Over 15 can exhaust your budget on discovery alone.
Keywords

Optional search terms (without #) used alongside hashtags. Good for catching content that uses natural language instead of hashtags.

Example: hm haul, hm try on, h and m outfit
Platforms

Select which platforms to scrape. Budget is split evenly across selected platforms. Each platform uses different Apify actors.

Tip: If budget is tight ($5-10), pick 1-2 platforms. TikTok is usually the richest data source for trend reports.
Influencers

Optional. Add specific creator handles per platform to scrape their recent content. Useful when you know key voices in the space.

Example: @theordinary, @hyaboron (TikTok handles)
Tip: Include handles with the @ for TikTok, without @ for Instagram.
Report Context / Vision

Free-text guidance that steers the AI agents. Tell it what you need from the report, what to focus on, who the audience is, or what business question you're trying to answer. This is injected into every AI stage so the entire pipeline is shaped by your input.

Example: "We're launching a new coffee pod range and need to understand the competitive landscape. Focus on Gen Z engagement, sustainability messaging, and home barista culture."
Tip: Be specific. "Focus on sustainability" is OK. "Focus on how Gen Z talks about sustainability in skincare, especially The Ordinary vs. CeraVe" is much better.
Apify Budget ($)

How much to spend on data scraping. 70% goes to discovery (finding videos), 30% to enrichment (pulling comments and transcripts). Split evenly across platforms.

$5 — Light scan. ~100-200 videos. Good for narrow categories or single-platform runs.
$10 — Standard. ~300-500 videos. Recommended for most briefs.
$15-25 — Deep dive. ~500-1000+ videos. Use for multi-platform, broad categories.

Tips for Better Reports

1. Be specific with hashtags
Generic hashtags (#fashion, #food) return noisy data. Use brand-specific and niche hashtags that target the conversation you care about.
2. Use the context field
This is the single most impactful field for report quality. Tell the AI what business question you're answering, who the report is for, and what kind of insights matter most. Without it, the AI generates a generic category overview. With it, you get a focused, strategic document.
3. Match budget to scope
Running 3 platforms with 20 hashtags on a $5 budget means each search gets pennies. Either increase the budget or narrow the scope. Fewer platforms + fewer hashtags + more budget = richer data per search.
4. Add influencer handles
If you know the key creators in the space, add them. Their content gets scraped directly (not via hashtag search), so it's more reliable and adds depth to creator spotlights.
5. Set a recent date range
The pipeline filters for content within your date range. A 30-day window gives you timely trends. Going beyond 60 days dilutes the "what's happening now" signal.
6. Save and iterate
Save your brief before running. If the first report isn't focused enough, tweak the context field or hashtags and run again. Each run costs a few dollars, so iteration is cheap.

What Each Stage Does

Stage 1 — Brief Validation
Validates your form inputs. Checks required fields, valid platforms, date range logic.
Stage 2 — Strategy Review
Two AI agents (Community Manager + Brand Strategist) review your brief and generate initial hypotheses about what trends and insights to look for.
Stage 3 — Discovery Scrape
Scrapes TikTok, Instagram, and YouTube via Apify using your hashtags, keywords, and influencer handles. This is where most of the Apify budget goes (70%).
Stage 4 — Data Review
AI agents review the scraped data, select the most relevant videos, and refine their hypotheses based on what was actually found.
Stage 5 — Enrichment Scrape
Pulls comments, transcripts, and thumbnails for the top videos. Uses the remaining 30% of Apify budget.
Stage 6 — Pre-Report Review
AI agents do a final review of the enriched data and generate desk research queries to validate findings.
Stage 7 — Desk Research
Runs web searches to corroborate claims and add industry context to the report.
Stage 8 — Report Generation
Claude Opus generates the final report: executive summary, trends, audience insights, content opportunities, creator spotlights, and visual language analysis. Outputs HTML, JSON, and Markdown.

FAQ

How long does a run take?
Typically 5-15 minutes depending on the number of platforms and data volume. Stage 3 (scraping) and Stage 8 (report generation) take the longest.
What does it cost?
Apify cost is set by your budget field. Claude API cost varies but is usually $1-4 per run on top of the Apify spend. Total cost is shown in the live tracker during the run.
Can I run it again with tweaks?
Yes. Save your brief, adjust whatever you want, and run again. Previous reports are preserved in Run History.
What if a stage fails?
The pipeline will show the error in the log. Common causes: Apify budget exhausted (increase budget or reduce hashtags), API rate limits (wait a few minutes and retry), or invalid brief fields.