Revolutionizing Search Experience: Enhancing User Discovery

Duration

2023.03-2023.06

2023.03-2023.06

My Role

Product Designer

Product Designer

Responsibility

Product designer, focusing on UI/UX and search logic.

Product designer, focusing on UI/UX and search logic.

Company

FunNow

FunNow

Overview

Team Goal

Improved FunNow APP's search functionality to enhance user experience, focusing on optimizing search efficiency, expanding search scope, and increasing search utilization to better meet user search needs and boost user retention.

My Role

I'm a product designer working closely with PMs and engineers, with a focus on early user research and defining search logic.

Challenges

The existing search functionality has the following issues:


  1. High error rate: Users often encounter misspellings or abbreviations during input, leading to search failures.

  2. Restrictive search logic: The search function only supports exact matches of store or product names, limiting the types of searches users can perform.

  3. Low usage: Only a small number of users discover and use the search feature, necessitating increased visibility and usage.

Result and Impact

After the project went live, we observed data for one quarter and found the following:


  1. Improved search efficiency by 23%: Successfully reduced search error rates, increasing user success rates during searches.

  2. Expanded search scope: Introduced search support for industry names and locations, offering users more exploration options.

  3. Increased search utilization by 9%: Redesigned the search entry, enhancing the visibility and usage of the search feature.

  4. Adjusted search result weighting, improving result accuracy and relevance.

Background

FunNow is an urban entertainment instant booking platform that offers reservations as fast as 15 minutes before the desired time. As part of FunNow's User Growth Team, our primary goal is to increase the number of unique users and enhance their user experience. We concentrate on improving the search and browsing functionalities as these are users' primary actions before discovering and making reservations for their desired activities.


Through data analysis, we found that search results rank third in terms of product card views among all the sources leading to the product detail page. This indicates that when users engage in search behavior, they already have a high degree of intent and willingness to purchase. However, we observed that only 12% of users utilize the search feature.


We aim to optimize the search experience so that even users with clear objectives can complete their bookings faster and more efficiently through the search feature.

User Reasearch

The search functionality can be categorized into Entry, Pre-input, During Input, In Progress, and Search Results phases.

From user interviews, we've identified two main types of search users:

  1. Users with a specific venue in mind, either for a revisit or to check for deals.

  2. Users with category/sub-category and location preferences, seeking to explore available options.

Entry

Pain Point 1:
Hidden Search Entry

The search entry isn't obvious, and some users haven't discovered it, so they stick to their usual ways of finding products.

Before Typing

Pain Point 2:
Unclear Search Option

Uncertain about what to search for, and the popular searches don't appear genuinely popular or regularly updated.

When Typing

Opportunity 1:
Keyword Trends in User Searches

Users primarily enter "store names" as keywords, with occasional use of "industry names (e.g., restaurants)," "sub-categories within industries (e.g., burgers)," and "locations (e.g., Zhongxiao Fuxing)."

Pain Point 3: Uncertainty in Search Inputs

Users often feel uncertain about whether their input will yield relevant search results due to the lack of system feedback during input.

Result

Pain Point 4:
Search Failures Due to Misspellings and Abbreviations

When entering "store names," misspellings or abbreviations can lead to search failures and no results, causing user frustration.

Pain Point 5:
Limited Search Results for Industry and Location Entries

When entering "industry names" and "locations," the system's lack of support for this type of search only retrieves search results with keywords found within store names, leading users to believe there are fewer businesses on the platform.

Pain Point 6:
Limited Search Results Sorting

Search results sorting doesn't meet user expectations, and there's no option for users to customize it as they desire.

Problem Define

Based on the user research above, we can identify 3 key issues and challenges that will become the core focus of our project:

  1. Search Efficiency Issues: Difficulty with typos or abbreviations in search queries leads to failed searches, frustrating users and impacting conversions.

  2. Restrictive Search Logic: FunNow's exact match search limits results when users input industry names or locations, suggesting fewer options and not meeting diverse search needs.

  3. Search Visibility Issues: The search feature is not easily noticeable, leading some users to miss it and rely on familiar methods, possibly overlooking search opportunities.

Establish Hypothesis

Idea Workshop

Goal

  1. Ideate based on search objectives, metrics, and user pain points.

  2. Concentrate on improving elements within the search journey.

  3. Finalize ideas into experiments or product plans through discussion and consensus.

Hypotheses

  1. Assisting users during searches can minimize errors and steps, offering direction and improving efficiency.

  2. More precise search results can expedite finding needed information and hasten the buying process.

  3. Strengthening search prompts and visibility can boost usage and gather data for result optimization.

Deliverables

Expand Search Database

We will add search database fields such as "Program Tags," "Regions and Business Districts," and "Common Keywords" to provide users with richer search options.

Program Tags

Enable finer category searches within industries, e.g., dividing restaurants into burgers, all-you-can-eat, Japanese cuisine, etc. For massages, options like Thai massage, acupressure/oil massage, female masseuse, etc.

Regions and Business Districts

Users can search by specific districts, subway stations, etc., using Google Maps' central coordinate and area mapping.

Common Keywords

Include abbreviations and common misspellings, e.g., "北車" for Taipei Station, "政大" for National Chengchi University, "安森" for Daan Forest Park, and correct searches for commonly misspelled terms.

Auto Complete

Elevate search with instant suggestions as users type, boosting efficiency effortlessly. With added flair, enjoy highlighted matches and captivating icons or images for an intuitive and engaging search experience.

Search Entry Optimization

We're revamping the homepage search entry by replacing the magnifying glass icon with a more inviting input box. This upgrade includes the addition of a keyword ticker

Search Result Optimization

Our search results now feature a new map mode and customizable sorting, complemented by an optimized SERP algorithm. This guarantees higher visibility and priority for results closely linked to user's Auto Complete selections, streamlining their search journey.

Usability Test

Search Entry Accessibility: All participants found the search entry within 2.4 seconds, improving from 4.7 seconds, making it almost instantly accessible. The homepage's vibrant design with the mascot was well-received.


  1. Auto Complete Feature: All noticed the Auto Complete, but only two used it, with others preferring to hit "Enter." The feature's convenience and safety were highlighted, though some desired more intuitive icons or images.

  2. Finding Specific Products: The time to find specific products dropped from 17 to 10.3 seconds, showing the effectiveness of Auto Complete in speeding up the search process.

  3. Discovering Desired Products: All participants were able to find products previously not found on FunNow, highlighting the platform's improved search capability and content diversity.

Outcome

Following the project's launch, we monitored the search feature's data variations for a quarter and obtained the following detailed outcomes:

Improvement in Search Result Rankings on Product Detail Pages

We noticed that among the sources leading to product detail pages, the ranking of search result product cards improved from 3rd to 2nd place, increasing their share of overall traffic from 22% to 27%.


  • Users found what they were looking for more efficiently through the search function.

  • The conversion rate for users viewing search results via Auto Complete doubled compared to those taking a general path to product detail pages, highlighting the effectiveness of the Auto Complete feature.

Increased Usage of Home Page Search Functionality

Among users accessing the homepage, the proportion using the search function rose from 12% to 21%.


  • An increasing number of users became aware of and began using the search function to find their interests.

  • Enhanced visibility of the search entry point made it easier for users to locate the search function.

Higher Proportion of Search Results

The ratio of searches yielding results surged from 54% to 77%, indicating a significant improvement in search functionality. This enhancement allowed more users to find what they were looking for, reducing instances of no results or irrelevant outcomes and elevating the overall search experience.

Next Actions

Following the project's launch, we monitored the search feature's data variations for a quarter and obtained the following detailed outcomes:

Incorporate Discovery into Search

Our research shows that making the search process more exploratory can spark interest and encourage users to actively discover new restaurants, events, and experiences.

Personalized Recommendations

We believe personalization is key to engaging users. By developing a system based on users' past actions and preferences, we can enhance engagement and help them find new, interesting options.

Data Tracking and Analysis

We'll keep monitoring user behavior and feedback to refine our features and optimizations through data analysis, ensuring we meet user needs and provide valuable search experiences.

Thanks for watching.

Next On

Crafting Design Excellence: Building a Robust Design System

Redefining Membership: A Fresh Approach to Loyalty Programs

Contact

Hello! I'm here to connect and chat. Don't hesitate to reach out!

2024 Anniie Lee, who is swinging between ESFJ and ISFJ 🧋

Contact

Hello! I'm here to connect and chat. Don't hesitate to reach out!

2024 Anniie Lee, who is swinging between ESFJ and ISFJ 🧋

Contact

Hello! I'm here to connect and chat. Don't hesitate to reach out!

2024 Anniie Lee, who is swinging between ESFJ and ISFJ 🧋