The most useful Yelp reviews are the ones Yelp doesn't show you. The platform's recommendation algorithm hides what it considers "not recommended" reviews behind a tiny, easy-to-miss link at the bottom of the page — sometimes a third or more of a business's total review count lives there. The 2023 fake-review filter controversy made this very public; what didn't change after the controversy is the fact that the filter still runs and the hidden pool is still where most of the interesting signal sits. If you want to see the real picture for a business, you need both pools, in one spreadsheet, with the Recommended-vs-Filtered flag intact. This guide walks you through using ExportComments' Yelp Reviews exporter to pull every review — visible and filtered — for any Yelp business into Excel, CSV, or JSON.

Why export Yelp reviews

Yelp is the rare review platform where what's missing from the page is more interesting than what's on it. The recommendation algorithm decides which reviews get shown by default and which get pushed into the "not currently recommended" bucket. The criteria are opaque, the algorithm has been publicly contested, and restaurants have spent years trading notes about which positive reviews from real customers got filtered while obvious troll reviews stayed visible. The other major Yelp signal worth pulling is the Elite flag — the platform's invitation-only reviewer program, especially influential in the NYC dining scene where Elite reviewers can shape the listing of a new restaurant in its first month. Once it's all in a spreadsheet:

  • Recommended-vs-Not-Recommended split — pivot rating by recommended status. The hidden pool almost always has a different average than the visible one. The gap is the story.
  • Elite reviewer flag — filter to Elite reviewers and read what they said in the first 30 days after opening. That window disproportionately shapes the listing's trajectory.
  • Multi-location chain audit — bulk-pull every location of a chain and pivot by location. The two outlier locations — best and worst — are usually the most useful to study.
  • Photo-review extraction — buyer-uploaded photo URLs sit in their own column. Useful for menu-photo banks, hospitality marketing, and competitive UX research.
  • Hospitality and restaurant industry use — overlay weekly review volume against your reservation data and you'll see how Yelp impressions actually convert into seated covers.
  • Sentiment-vs-rating mismatch — find 4-star reviews with negative sentiment in the body, and 2-star reviews with positive sentiment. Both are common and both tell you the rating column alone is lying.

How to export — step by step

Step 1: Grab the Yelp business URL

Open the business's page on Yelp — for example https://www.yelp.com/biz/some-restaurant-new-york. The canonical business URL is what the exporter uses; you don't need to click into the reviews tab or scroll to the filtered section first. The exporter handles the Recommended pool and the Not Recommended (filtered) pool in the same job, and tags each row with which pool it came from.

Step 2: Paste the URL into the exporter

Head to the Yelp Reviews exporter and paste the URL. If you're auditing a chain — every location, or every NYC dim sum restaurant in a 10-block radius — switch to bulk mode and paste one URL per location. Bulk runs return one Excel file per URL, packaged together in a single ZIP at the end of the job, so each location stays in its own file for clean per-location pivots.

Step 3: Pick a format

Excel (.xlsx) is the default — the Recommended-vs-Filtered pivot is the first thing you'll do, and it's a 30-second job in a spreadsheet. CSV if it's heading into a BI tool or a structured pipeline. JSON if you're feeding the export into a sentiment model or running topic analysis on the bodies in a notebook.

Step 4: Start the export

Click Export. The job runs server-side and paginates through both the Recommended pool and the Not Recommended pool for the business, including the Elite flag, the owner reply where one exists, photo URLs, and the recommended-status flag on every row. Larger listings with thousands of reviews take a few minutes; close the tab if you want — the file lands in your dashboard and your inbox the moment it's ready.

Step 5: Open the file

Open the .xlsx in Excel, Numbers, or Google Sheets. Each row is one review, each column is one field, and you're ready to filter and pivot.

Inside the export — what fields you get

Each row is a single Yelp review. You'll find columns for:

  • Reviewer name — display name shown on the review.
  • Reviewer location — the city/region Yelp shows on the profile.
  • Elite — true if the reviewer carries Yelp's Elite badge.
  • Rating — the 1–5 star score.
  • Body — full review text.
  • Recommended — true if the review is in Yelp's visible pool, false if it's in the "not currently recommended" (filtered) pool.
  • Photos — URLs of any reviewer-uploaded photos.
  • Owner reply — the business owner's response text, when present.
  • Useful / Funny / Cool counts — the three engagement signals Yelp lets readers tag a review with.
  • Created at — review timestamp in UTC.

Common workflows

  • Recommended-vs-Filtered split — pivot rating by the recommended flag and recalculate each pool's average. The gap between visible and hidden is the most honest reading of a Yelp listing's real reputation.
  • Elite-reviewer first-month influence — filter Elite = true, sort created_at ascending. The Elite reviews in the first 30 days of a new business set the tone for everyone who follows. In NYC especially, this window shapes the listing's first year.
  • Multi-location chain audit — bulk-export every location, pivot rating by location URL. The best and worst locations are where the operational lessons live; the mid-pack rarely teaches you anything.
  • Sentiment-vs-rating mismatch hunt — run sentiment on the body column, then filter for 4-star reviews with negative sentiment and 2-star reviews with positive sentiment. Both are common; both reveal that the star slider lies more than your dashboard assumes.
  • Owner-reply audit — filter ratings 1 and 2, count rows with non-empty owner reply. For hospitality businesses the reply rate on negatives is one of the most reliable predictors of recovery.
  • Photo bank for ads and the brand site — filter photos column non-empty, sort by useful count descending. A ranked, customer-shot UGC pool, free to source from.

Plan limits and API access

The Free tier returns up to 100 reviews per export, which is enough to confirm the columns and the recommended-vs-filtered flag look right for your use case. Personal scales to 5,000 reviews per export, Premium to 50,000, and Business to 250,000 — enough to capture every review for the largest multi-location Yelp listings on the platform. If you'd rather refresh a chain audit on a schedule or trigger an export from your own pipeline, the same job is available through the REST API and via webhooks. See pricing for the full breakdown.

FAQ

  • Does the export include Yelp's filtered (not recommended) reviews?
    Yes — both the Recommended pool and the Not Currently Recommended pool come back in the same file, with a recommended flag on each row so you can split them in a pivot.
  • Can I tell which reviewers are Yelp Elite?
    Yes. The Elite badge is a separate column. Filtering to Elite = true is the fastest way to see who's been shaping the early read of a business.
  • Why does the visible-pool average differ from the filtered-pool average?
    Because Yelp's recommendation algorithm decides which reviews to show. The two pools rarely have the same average — and that gap is the most honest signal you'll pull from a Yelp listing.
  • Can I export every location of a chain in one batch?
    Yes. Use bulk mode and paste one Yelp business URL per location. The run returns one file per URL packaged into a ZIP, so each location stays in its own file for per-location pivots.
  • Are owner replies included?
    Yes. Owner reply text sits in its own column on every row that has a reply, which is what makes the reply-rate audit on 1- and 2-star reviews possible.
  • Can I extract reviewer photos?
    Yes — reviewer-uploaded photo URLs are in their own column. Filter for non-empty and you have a ready-made customer-photo pool for marketing or competitive UX research.