Exploring the vastness of a website

I read a post a while back about how SEO and search engine ranking algorithms have created a landfill of useless search results across search engines. It’s interesting how dramatically different searching the web was before search engines began ranking results based on popularity and profiling user engagement. In the early days of search, there was a real feeling of exploring the vastness of the web.

In the last few years, websites and social media platforms have introduced explore pages/tabs into their interfaces. Usually these are just feeds masquerading as explore pages. Other times they are simply static curated “staff picks” lists, or personalized recommendation pages. To me, these pages don’t live up to the name “explore”. A feed is based around time, usually sorted reverse chronologically, whereas an explore page reveals the expansiveness of a website by pulling from disparate sources, indifferent to time, allowing one to jump into the depths of something entirely new. Even personalized recommendation pages like Instagram’s explore tab aren’t really explore pages either because they are limited to a user’s interests/similarities to other users on the platform. Being fed similar content based on your likes isn’t the same as exploring the vastness of a platform. Explore pages are unique in that they unearth things that are not necessarily the most popular or the most recent. Interestingly, when explore pages are built this way it allows users to approach a platform with an open mind and with less expectations. Explore pages should make you feel a sense of excitement similar to the feeling of exploring a new place.

When I worked at The Creative Independent, we talked about creating an explore page, but we never got far enough along on the idea to define what the page would actually look like. One thing we did implement was a random button that served up a random interview from over 600 articles across the site. I ended up moving this button into the main navigation so that readers could continue to click the button until they found an interview that interested them. It’s fairly easy to implement a “randomized items/articles” section on a website. In the case of The Creative Independent, this simple addition revealed how expansive the site really was.

In 2008, I came across the site Muxtape, which was the original “mixtape on the web” website. It allowed you to upload MP3s from your computer and curate the tracks into a “muxtape”. Though the site was eventually shut down, I often think about one key design decision. Muxtape’s homepage had a “Random active muxtapes” section that changed on refresh. This simple randomized section gave users a way to instantly explore muxtapes across the website.

Interestingly, this feature alone replaced the need for a search. The site became centered around exploring mixes instead of searching for a particular mix or a song. I remember spending hours refreshing the homepage, clicking through to a mix, listening, and heading back to the homepage to find another. I’d never experienced anything like this on a website. It truly felt like I was exploring a vast archive similar to the experience of wandering through a library and randomly choosing a title from a shelf. It was a completely new feeling and it all came from a simple random function.

Though I doubt anything like Muxtape will ever exist again, I don’t think the idea of exploring a website through randomization has to be completely lost. This simple design decision showed how useful randomization can be without any sort of curation or personalization.

In the last few years, platforms have stripped away any hint of how vast they actually are. As a result, users only get to see a tiny sliver of an entire platform. There’s been an overwhelming push to build tools specifically designed for engagement (like buttons, emoji responses, comment threading) instead of building tools that help users actually explore. This has replaced any sense of play with a bleak struggle for users attention. The marketing line for these new tools could easily be, “engage more, explore less.” I’m proposing that we begin designing with vastness in mind again. The data is already there, all we have to do as designers and engineers is to build tools that reveal how expansive these platforms really are.

Published on August 18, 2019