BACKGROUND

Discogs, a comprehensive audio recordings database, doubles as a marketplace for physical media, accessible through its standalone mobile app and mobile website version.

PROBLEM

The current search functionality on both the website and app is limited to a text input field and a barcode scanner. To enhance the efficiency of music searches and optimize the seller's inventory updating process, it would be beneficial to introduce additional search methods.

GOALS

  • Add “musical search” (search by playing or performing audio excerpts)
  • Add search by release pictures
  • Add text scanner
  • Add voice prompts search
  • Lay all these options out in a way optimal for a mobile device, making them not bulky, not overwhelming, and easily accessible
  • Optimize search results sorting

OBJECTIVES

  • Competitive analysis of digital marketplaces’ and databases’ search engines
  • Implement additional search methods
  • Create a prototype with the added search options

TOOLS

  • Survey users to identify useful and unnecessary search options for optimization.
  • If results are thorough, craft a persona for the chosen features.
  • Competitive analysis
  • Mid- and hi-fidelity wireframes in Figma
  • Prototype in Figma

OUTCOME

COMPETITIVE ANALYSIS

I've analyzed 7 crucial elements that improve our project's search capabilities. I've also examined how 6 competitors handle this aspect.

ASKING QUESTIONS

  • 30 responses were received from different sources
  • The survey was deliberately given to people who may be, but are not necessarily Discogs users
  • Beyond the survey, I engaged a variety of music enthusiasts to assess music search widgets. Their diverse tastes provided valuable insights:
  • Google and Shazam emerged as the top picks, mentioned most frequently.
  • My testing spanned diverse environments: shopping centers, movie theaters, live concerts, music shops, and more.

THE FINDINGS

WAYS TO FIND WAYS

While most prefer text input, our research uncovers surprising alternatives, especially in music searches.

WHAT DID I JUST HEAR?

93% of the respondents find themselves in a situation they want to find the music they hear, to 40% it happens often.

SAFE IN SOUND

90% of the respondents either already use the search-by-sound feature, or would want to use it.

PICTURE THIS

While over half of music seekers don't use image search or text scanning, a solid 33% are keen on exploring these options.

SHAZAM vs GOOGLE

Shazam is the favored choice for many, thanks to its one-click simplicity. However, it came up short when tested against the Google widget in the field.

WHEN OLD IS NEW

The results indicate that new Discogs users highly value and expect familiar search methods, akin to those found in other programs and widgets.

AN UNEXPECTED PROBLEM

I was aware that these search features are employed by music enthusiasts. However, given that Discogs specializes in physical media, understanding the specific needs of our buyers became crucial. To gain deeper insights, I opted for a strategic approach: conducting three interviews and crafting personas that would synthesize both interview and survey findings.

ASKING MORE QUESTIONS

These were the main questions I was seeking answers to:

  • Why continue to purchase physical media when nearly all music is accessible through streaming services?
  • Do you ever buy music you've heard by chance?
  • Do you typically buy music on impulse, or do you prefer to think it over, gather more information, and so on?

PERSONAS

RECORD COLLECTOR

RECORD SELLER

FOOD FOR THOUGHT

In accordance with my findings, the following decisions are made:

  • The most useful search feature would be the “search-by-sound”
  • The next step would be creating the option of “search-by-picture”
  • Combine the Shazam UX simplicity with Google’s effectiveness
  • “Search-by-text” scanning feature leave for the future as the least popular
  • The Discogs database usually serves as a price reference and/or a wish list
  • It is rare that the user buys the music spontaneously, but he/she uses either the wishlist or shopping cart to add the results of the findings as soon as possible

NOTHING IS FLOW-LESS

To help map out the integration into the existing app structure, I created a user flow showing how the functions will interrelate and compliment each other

USER FLOWS FOR MUSIC SEARCHES METHODS

For the sake of simplicity and clearness, this user flow omits the text input and browsing/sorting methods, and reflects only “untraditional” methods of search, two of which are new:

  • Search by Sound
  • Search by Barcode
  • Search by Picture
UI  SYSTEM
THE ICONS
New Icons were created from scratch to coincide with the new features
This icon mocks one of the most iconic artworks ever for one of the best sellers of all time, Pink Floyd’s Dark Side of the Moon
I chose this microphone design to stand out from system icons and give off a karaoke vibe. It should scream 'singing and music,' not 'voice prompts.
The existing icons were changed slightly to create consistency for the renewed UI system
The magnifying glass became thinner but bigger and brighter
The barcode slightly changed proportions to look clearer, and the lines thickness and brightness are adjusted to the rest of the icons
The shopping cart became outlined (instead of filled), with less rounded corners to match with the style
The envelope became outlined as well, and the corners were also became rounded to correspond with the style
Same refers to the close button, hamburger menu, etc
NEW HEADER
  • The new header version has more clickable space around the icons
  • The hamburger menu disappears as soon as all the options for searches appear
  • The searches are grouped together yet are separated enough for improved accessibility

BEFORE

AFTER

WIREFRAMES

I've crafted a system of components and wireframes for our upcoming prototype, encompassing all three innovative search methods

USER TESTING
Overall, the user testing was a success. Even those new to Discogs found the new features easy to use and genuinely exciting.

PROBLEM

Testers showed some uncertainty about the search process stages and were unsure if they should press anything to proceed.

SOLUTION

I've added a progress timer to all three search options.

PROBLEM

If a tester accidentally selects the wrong search method, they have no option to correct the mistake except by going back to the search home page.

SOLUTION

The navigation menu is now accessible at any stage of the search process.

PROBLEM

The scan options had distinct UI differences, making the camera usage choices less clear.

SOLUTION

To enhance consistency, I unified the visual elements and introduced a timer that matches the selected method's icon.

PROBLEM

Due to visual inconsistencies, the testers were confused if some of the options were available only in certain searches

SOLUTION

I fine-tuned and adjusted the visual elements to maintain a sleek and clear visual coherence.

FINAL DESIGN

SEARCH BY SOUND

One of many available ways to look for a venue is looking at the map

SEARCH BY PICTURE

The app now scans pictures and searches for matches in the database

HITS (AND MISSES)

BUYERS’ SUPERHITS

  • wider and more diverse range of search options
  • immediate results in practically any possible situation and/or environment
  • even if search is connected with a third-party service, it’s available directly in the app
  • the search options became clearer thanks to the new UI elements

SELLERS’ SUPERHITS

  • the search allows extremely quick inventory cataloguing and listing items on the marketplace
  • the search allows to quickly detect whether the title is in the inventory — even if neither the client nor the seller know the author or name of the record (by simple humming, singing or playing)
  • the searches speed up the process of pricing, as sales history is accessed faster

MISSES, DARLING

  • the survey could have been wider, with more open questions
  • due to time constraints, not enough interviews were held
  • the respondents were diverse enough, but the groups were rather uneven
  • there was no opportunity to test the app on its devoted fans

LESSONS LEARNED

  • Ah, problems! They're like surprise plot twists in the story of development – unexpected, but they keep the narrative exciting.
  • Prioritization is like a two-act play: first before testing, then the encore after testing – setting the stage for success!
  • Unlocking certain answers can be like finding rare gems outside the confines of your focus groups – hidden treasures waiting to be discovered.
  • In the realm of experience, the path of improvement never ends.

Previous Project

Next Project