Optimizing The Review Flow

8 Weeks

UX/UI Designer / Researcher / Data Analyst

Android, iOS, Web

This is a project I did while I worked at Wongnai (Thailand's Yelp).


Wongnai gives out restaurant recommendations based on the users’ reviews. The reviews are rated from 1-5 stars (one being the worst, five being best) but sometimes judging a restaurant can be a hassle. Sometimes users don’t know whether to give a restaurant 4 or 5 stars. So we have a couple of ideas and requirement that we want to try out.

  • Allow user to post without giving the rating
  • Users can convert from post to review
  • Asking whether the users recommend a restaurant instead of asking for rated score


We use big rating buttons in the middle of the screen to persuade users’ to rate the restaurant.

With the new requirement, I tried out the design that allow users to post stuff to the restaurant without having to give ratings.

Without the rating buttons, the team doesn’t feel confident that the new design would benefit the review platform. (Since we rely on these ratings to give out recommendation)

So in another iteration, we added both the ability to post and rate. But this falls into the paradox of choices. There are a lot of things going on. It seems complicated and confusing. I felt like it’s better to just have one main action for the users to do. So we scrapped this design.

Focusing on What Matters

With each iteration, the design above was becoming more complicated. In the end, we decided on what actually helps people find restaurants and decided we want to focus on :

  • Rating: To rank different restaurants
  • Recommended Menu: Quickly get a grasp of what interesting
  • Photo (tagged menu): Obviously help people identify the menu
  • Short text highlight: Easier to read and write!

Mini Review Revisit

About a year ago when we redesign the business screen we created “Mini Review” to make writing review simpler. We’ve always kept “Classic Review” for elite users who contribute a lot of reviews to the platform here’s the overview of “Mini Review” and “Classic Review”

Since we’ve decided what we’d like to persuade users to do from the section above we rethink the Mini Review and came up with “Review Popup”.

When users click on rating instead of getting a text field to fill they are met with a popup that asks them for menu recommendation, short highlight of the place and photos.


Every time we want to utilize popup we get the stigma that it would annoy the users. Here’s we’re going to look at the data and find out if the Review Popup actually annoys users.

We look to compare the “post button” conversion of Mini Review and Review Popup to find out which one is best.

Turns out we see a 120% increase in conversion rate.

We also see the increase in photos posted (9.57%) and recommended menu posted (29.7%).

The Review Popup makes the users’ tasks clearer. It’s far more in the users’ face. That’s the reason for the increase.

Repeat Rate

We compare Mini Review and Popup on how often the user rate a restaurant and found no change in the frequency. This shows us that the users are not annoyed by the popup because if they are they would use the feature less frequently. It is safe to assume that this popup isn’t out of users’ expectations when they click to rate a restaurant.

No Brainer Adjustment

To top it off, we did something we should have done a long time ago. Writing a review requires users to be logged in. We let the users fill in the details before asking them to log-in or signup. Once people put it the effort to fill in the detail they are much less likely to abandon when asked to log-in.

This improves the conversion dramatically. We gain 40% more new reviewers per week.