1. Find restaurants
Search by name or suburb, filter by cuisine and budget, or use the map. If you choose “Use my location”, your browser will ask for permission before we show nearby places.
2. Read the complaints
We use AI to group written 1–3-star reviews by topic, such as slow service or small portions. Restaurant pages show excerpts and links to the original reviews so you can read more about what happened.
We also show Google ratings and restaurant details where we have them. Google’s rating covers a different set of reviews from our collected sample, and our critical-review score has its own calculation. Our sample can miss reviews or recent changes, and we can put a review under the wrong topic. We haven’t independently checked whether each reviewer’s account is accurate.
3. Filter out complaints you want to avoid
Under “Criticisms to avoid”, select an issue and a percentage. A restaurant is hidden if any issue you’ve selected appears in at least that percentage of its analysed 1–3-star reviews. For example, selecting slow service at 25% hides a place when at least one in four analysed critical reviews mentions it.
That percentage counts analysed critical reviews, so it doesn’t tell you how many of all diners had the problem. A review can mention several issues. Charts that count topic mentions use a different total; the label beside each chart explains what it counts.
A restaurant may still appear because we haven’t collected or analysed enough reviews. Check how much information is available before drawing conclusions from a missing complaint.
4. Compare and save places
You can compare up to three restaurants side by side without an account. Sign in with Google to save places and see them on another device using the same account. Your priorities and comparisons stay in the browser you’re using.
Preferences change the critical-review score. To hide restaurants from your results, use “Criticisms to avoid”. Comments posted on RipperFeeds are separate from collected reviews and don’t affect the score.
How we calculate the critical-review score
The score summarises complaints in the reviews we’ve collected. A higher score means a smaller penalty in the calculation below. Read the complaints alongside the number, since two restaurants can have the same score for different reasons.
We start with the collected 1–3-star reviews. Each has a penalty of (4 − stars) ÷ 2: 1.5 for one star, 1 for two stars and 0.5 for three stars. We subtract the average penalty from 5 to get the starting score.
Your five “most important to avoid” topics get weights from 1.5 down to 1.1. Your five “easier to overlook” topics get weights from 0.5 up to 0.9. All other topics have a weight of 1. We divide each review’s penalty across its topics, so adding more labels doesn’t increase its starting penalty.
The difference between the weighted and unweighted penalties changes the score by up to half a point in either direction. Reviews we haven’t classified still count towards the starting score, but your preferences don’t change their penalties. Positive reviews and review age aren’t part of this calculation.
We need at least five critical reviews and enough usable analysis to show a score. We may also withhold it if the data is incomplete or out of date. Five reviews is our minimum for displaying a score; it doesn’t guarantee a reliable picture of the restaurant.
What the reviews can tell you
Reviews describe someone’s experience at a particular time. Even a complaint that comes up often may have been addressed since. Read recent and positive reviews too, and check current details with the restaurant. Our scores and filters can help you compare the feedback we have, but they can’t guarantee what your visit will be like.