Guitar Tab Project

Can UX help a market leader to retain their position in a highly competitive landscape?

Ultimate Guitar has established itself as the go-to database for learning songs. Although their interface is packed with several tools to aid the user in playing, users are still left wanting when it comes to an experience. Prototyping involved playing guitar while designing the interface. One particular feature seemed to make all the difference for usability and customer retention....

UX, Business Modeling, Redesign, Music

Ultimate Guitar - Redesign

Test Subject: Me

The world of music made a shift years ago, when graphic notation was used less and less. It was a broken method with a high learning curve. Even those who are familiar with staff/sheet music sometimes prefer the simplicity of tablatures (a.k.a., tabs) and chords sheets.

One of the most prominent databases derived from Russia - Ultimate Tabs. As of late, they have several mobile apps that do incredibly well - tab database, sheet music database, tutorials database, guitar/ukulele tuner, etc. While the mobile experience has advanced, a lot of their existing website has not really met the mark. I reached out with them with my own ideas on how to build a valuable experience. I was positively biased - I was a long-time user of their assets and a pretty okay musician myself.

Here’s me in 2007 performing an original piece. Man, I was super skinny. Ew, and look at that fedora.


While Ultimate Guitar remains king in the landscape of other tablature sites, they really a weak grasp on that title, because they don’t own any of the tabs. They can easily be snatched and placed in a better interface (which it indeed happens, pretty often). So they are consistently trying to create new value in their UI - for example, an auto scrolling module in their tab interface. The right sentiment (observing the music reading behavior of their users), it still falls short in being truly useful.

Build Around the Task of Tab Search

  • Show only what’s important (the existing homepage was a hodgepodge of too many things) - in this case, searching for the right tab should be very clearly the main element.
  • People are searching based on their mood, and by the difficulty level. Build search to accommodate these two parameters.
  • A person’s taste in genre can help self segment - give priority to high propensity songs based on genre.
Eliminated the fluff - less words, more clear indication for the guitarist to get to what they want.

Moderate/Curate for Tab Quality

In the world of tab search, it is highly likely to find tabs that are incorrect - because it is all user generated content. Therefore their rating system helps immensely when looking for the right tabs. While some people may look at the lower ranking tabs, it would be a better experience to highlight the high ranking-tabs and hide the low-ranking tabs - so users don’t have to spend time sifting through BS.

Syndication is Paramount

  • Users don’t know there are other features/apps - have global links to all of them.
  • Each module/feature on the site should be built with an experience in mind. For example, the chord lookup module can be designed with UX principles, be aesthetically desirable to use.
  • Integrate to other credible music databases, instead of building own - like Soundcloud.
Only showing what's necessary. Artist Imagery to soften the interface.

Not all great value comes from hi-tech solutions

Part of my research process was to identify key moments of frustration in music reading. So I chose songs I wasn’t very familiar in playing.

  • Scrolling sucks when I have to use both my hands on the guitar.
  • I lose track of where I am in the song when I play - I close my eyes sometimes.
  • Chords can be confusing - there are diminished and augmented version of minors and major chords…and it doesn’t help that the letters are super tiny on the screen.
This is how some of them look, it can get hard to look at.

It came down to prototyping several methods - approaching the size of the text, clarity of fonts, even with symbols. In the end I put the chords in colored boxes. It made it easy to identify chords, but also pick up where I left off if my eyes wandered away from the screen when I played.

Color-coded chords - from prototyping the interface myself. Turns out I was conditioning myself to read the colors as they were easier to read. Conditioning presents an interesting case for buiness modeling for market leadership.

I noticed something interesting - I was training myself to follow the colors instead of merely the chord letters. I find this interesting, because it was substantially easier as an addition to the letters…but I was building a dependence to the colors. A dependence could equate to loyalty from a business perspective. Being one of the largest databases for tabs/chords in a sea of competitors, I figured this was more than a UX insight find, but a strong method of solidifying a brand’s market share in a commoditized landscape.