Docs / Sources

Sources

Sightlabs monitors 14 platforms across three categories: social media, search engines, and AI chatbots. Here is what is covered and what to expect from each.

Social media

6 sources

Social sources return the most raw mention volume. Reddit and Twitter / X are usually the highest-signal channels for brand and keyword tracking.

Twitter / X@mentions, keywords, threads

LinkedInPosts, company pages, comments

RedditPosts and comments across all subreddits

YouTubeVideo titles, descriptions, comments

InstagramPosts and captions (public accounts)

TikTokVideo descriptions (public accounts)

Note

Instagram and TikTok are limited to public accounts. Private posts and Stories are not accessible.

Search engines

3 sources

Search sources show how your brand appears in organic results and news. Google AI Mode captures the new AI Overview answers that appear above organic results, increasingly important for GEO (generative engine optimization) tracking.

Google SearchTop organic results

Google NewsNews tab results

Google AI ModeAI overview answers and citations

AI chatbots

5 sources

AI sources query each chatbot with your brand or keyword and capture whether and how it appears in the response. This is what Sightlabs calls AI share of voice: your brand's presence inside LLM answers. Available on Pro and Team plans.

ChatGPTAnswer text and cited sources

PerplexityAnswer text and linked sources

GeminiAnswer text and citations

CopilotAnswer text and linked results

GrokAnswer text and sources

Warning

AI chatbot responses vary by session. Two identical queries to ChatGPT may return different answers. Sightlabs runs multiple probe queries per platform to reduce variance, but results should be read as indicative, not exact counts.

Coverage gaps and rate limits

Sightlabs does not have access to:

Private or friends-only posts on any platform
Direct messages or group chats
Paywalled content (e.g. Substack paid posts)
Discord servers or Slack workspaces
Facebook (access restrictions prevent reliable scraping)

Some sources apply rate limits that cap results per scan. Reddit and Google may return fewer results during high-traffic periods. If a source returns zero results, try re-running the scan. Transient timeouts are the most common cause.

Adding source filters

The source picker in the scan widget lets you toggle platforms on and off before each run. Your selection is not saved between scans by default. If you save a scan and re-run it, it uses the same source set you originally selected. You can change the sources before re-running from the scan history view.