Adjust your search results with a simple drag and drop.

AI Search
Edit Mode

OVERVIEW
Problem

With Yext’s entrance into the eCommerce market came a new persona, the merchandiser needs to tweak search results to meet a business objective while not necessarily knowing about the inner workings of search like our usual users. Our current users had to edit search results through Query Rules, which required a time consuming flow configuring pairs of criteria and actions.

Solution

Create a simple drag-and-drop WYSIWYG interface where a user can directly reorder and manipulate search results. Query Rules will be auto-generated based on where the user drags search results. This solution was two-fold:

  • It enabled Yext to be a stronger option for current and potential eCommerce clients, such as Samsung, by having a tool essential to their business needs.

  • It greatly reduced the time needed to complete the task of modifying search results. This was a welcome improvement by our current users who often interacted with Query Rules.

Team

Michael Cho

Senior Product Designer

Max Davish

Product Manager

Max Shaw

SVP of Product

Jessie Smith

Upward Analyst

Daniel Bagel

Upward Analyst

Jim Henris

Software Engineer

Emily Zhang

Software Engineer

Scope

New Feature

Timeline

4 months (Jan '22 - Apr'22)

Tools

Figma, Snowflake

My Contribution

As the product designer on the team, I conducted a competitive analysis, took part in brainstorming sessions, created mockups, gathered feedback, built Figma prototypes, performed user testing, and participated in a bug bash.

Reorder Entities
Boost or bury to get the right results

Ensure certain products appear at the top of the search results when users search a specific query.

Reorder Verticals
Organize the correct order of categories

Sort each type of result to add an extra layer of customization at a higher level.

Train the Algorithm
Edit the answer given by the algorithm

Approve, reject, or modify the answer to the query so that the search algorithm gets better over time.

Our Process
Research
Brainstorm
Mockups
Prototype
User Tests
Iterate
Development
DESIGN
Research

Max, our PM, created a Snowflake dashboard for us to see our users’ current usage for key-related features involved. Jessie, our analyst, provided great insights through her experience working on Walmart’s eCommerce platform. I performed a competitive analysis on other brands to compare different approaches.

Brainstorm

The team brainstormed multiple approaches to how we wanted this feature to look and operate, and I designed rough mockups of some ideas that we had.

Mockups

After receiving feedback from stakeholders, designers, and engineers, I increased the fidelity of the mockups of the approaches and develop the key flows. Given that the feature required many net-new design components, I took the opportunity to update the aesthetics to better reflect today’s UI trends.

Prototype

Our PM and SVP quickly built a basic prototype that mirrored the functionality of my designs that used real data from our platform.

User Tests

I conducted and recorded a few tests with active users within Yext that would more closely match the merchandiser persona. They performed basic tasks and gave valuable feedback.

Test Recording

User Testing Slack Channel

Iterate

I updated my designs based on the feedback to help address issues that our participants came across, and I prepared my Figma file for engineering handoff through utilizing Figma's interactive components to provide clear organization and annotation.

Old Empty State

New

Old Warning

New

Old Drag-and-Drop Interaction

New

Organized Figma Handoff

Interactive Components
(Go ahead and click around!)

QA + Bug Bash

Throughout development, I regularly QA’d the feature to ensure everything matched my designs and to catch any edge cases. We also hosted a bug bash that presented both bugs and UX feedback that we could address within the last weeks before launch.

QA Sheet

Bug Bash Slack

LAUNCH
Spring Release 2022

After making final fixes and some adjustments to the design, we successfully shipped Edit Mode as a major part of Yext’s Spring 2022 release.

Tracking Success

Our team collaborated with the Analytics team to come up with a database schema that will help us answer important questions about adoption, retention, and ease of use.

We received many slack messages about our users praising the new feature! However, I had left Yext right after this project, so I did not get to take a look at the numbers to see measured impact. I would have been interested to see how we could iterate on the feature to improve it further.

Database Schema

Questions

Tradeoffs and Possible Improvements
Limited Scope

We simplified this feature to have a limited set of capabilities to not overwhelm users. There are more advanced editing options that are very powerful but we decided not to include those to simplify the flows and edge cases, reduce scope, and launch an MVP version.

List View vs Grid View

Most eCommerce sites provide a grid view of search results, and ideally, we wanted our feature to do the same. However, this would have increased the scope by creating/developing two separate designs for list and grid view results. Due to time and resources, we kept the list view as our MVP approach since the vast majority of our customers use it.

CONCLUSION
Final Thoughts

I was proud to have collaborated with my team to create this feature as part of Yext’s emergence into the eCommerce market. In addition, adding an updated design aesthetic that increased the platform’s perceived value was a great contribution to the company’s design.

Figma Mocks + Live Site

Take a look through the actual file! Things can get pretty messy as always.

other projects