API Reference › AI Mode › People Are Saying

People Are Saying

Community-discussion cards that Google surfaces on the results page, delivered inline as part of the standard Google Search response.

Overview

The "People Are Saying" block — Google sometimes labels it "Trending posts and discussions" — collects short cards that link out to forum and Q&A threads, most often Reddit or Quora conversations tied to the searched query. cloro parses this block and hands it back to you as structured data.

You do not call a dedicated endpoint for it. Whenever the module appears on the SERP, its cards ride along inside the same payload produced by the Google Search endpoint, exposed under the peopleAreSaying array. Each element of that array represents one card.

When Google does not render the module for a given query, cloro leaves peopleAreSaying out of result entirely rather than returning an empty list. Write your integration to treat the field as optional and check for its presence before iterating.

Card structure

Field	Type	Description
position	number	Rank of the card inside the module, counted from 1.
title	string	The headline text shown on the card.
link	string	The URL the card opens when clicked.
date	string	Google's original relative-time label, passed through verbatim (for example `"5 days ago"` or `"2 weeks ago"`). cloro does not convert this to an absolute timestamp.

Response example

{
  "success": true,
  "result": {
    "peopleAreSaying": [
      {
        "position": 1,
        "title": "Best running shoes 2026 - what runners are saying",
        "link": "https://www.reddit.com/r/running/comments/example/",
        "date": "5 days ago"
      },
      {
        "position": 2,
        "title": "Honest review of the Gel-Nimbus 28 after 200 miles",
        "link": "https://www.quora.com/example",
        "date": "2 weeks ago"
      }
    ]
  }
}

Notes

Because date is a raw relative label rather than a normalized value, parse or interpret it on your side if you need an exact publication time. The set of cards and their ordering reflect what Google served at request time and can shift between otherwise identical queries.