Creating an awesome StumbleUpon alternative, one year on

Archive note: This post was originally published in May 2021. It has been lightly rewritten for the new Bored Hoard while keeping the story intact.

With the release of our 52nd Box, an obvious piece of maths finally caught up with us: 52 boxes meant we had been doing this for a full year.

Bored Hoard had started as a side project for finding interesting websites. Somewhere along the way it became a proper product, picked up a community, survived a redesign and briefly got flattened by Reddit traffic.

So this is the story of that first year.

The people behind the squirrel

Bored Hoard started with two product people who had already spent years working together and constantly throwing project ideas at each other.

A suspicious amount of that happened while cooking in the office kitchen. One of us was Irish, the other Italian, so food was usually involved. When we were not discussing products, we were comparing zombie survival plans, Football Manager tactics and whatever other extremely important problem had appeared that week.

We liked building things, trying things and following ideas further than was probably necessary. Bored Hoard came out of exactly that habit.

Why build another way to discover websites?

We missed StumbleUpon.

Not because the internet lacked websites. Quite the opposite. The problem was finding the interesting ones without already knowing what to search for.

StumbleUpon had made wandering around the web feel effortless. You could start with no plan and end up somewhere completely unexpected. After it disappeared, we tried alternatives, but nothing quite scratched the same itch.

Eventually the obvious solution presented itself: stop complaining and build something ourselves.

Bored Hoard V1 — April 2020

The first version was basic.

The homepage revolved around weekly “boxes” containing hand-picked websites. Open a box and you got a set of finds organised by category. It was simple, slightly odd and good enough to prove the idea.

It also had a PWA, because apparently if you are building a small side project you must find at least one unnecessary technical hill to climb.

The problem appeared as the archive grew. Releasing a new box every week made the product harder to navigate. What worked with a handful of boxes started becoming friction once there were dozens of them.

The format had helped us start. It was not going to help us scale.

V2 — January 2021

At the beginning of 2021 we rebuilt Bored Hoard around discovery rather than boxes.

The new version made individual websites easier to find and browse, and it brought the experience much closer to the original reason we had started the project: giving people somewhere to go when they wanted to find something interesting but had no idea what that something was.

Feedback from people on Twitter helped us find bugs, rough edges and features worth pursuing. We also started sharing Bored Hoard in communities where old StumbleUpon users and people building side projects tended to gather.

That worked rather better than expected.

The Reddit hug of death

At one point a Reddit post sent thousands of people to Bored Hoard in a very short space of time.

The site did what small sites traditionally do in this situation: suffered.

We increased capacity, got things running properly again and watched a side project suddenly deal with a level of traffic we had not planned for. Painful, useful and considerably more interesting than a quiet analytics dashboard.

More importantly, a lot of those visitors stuck around and gave us useful feedback.

What we wanted Bored Hoard to be

From the beginning, the point was never to dump a giant database of URLs onto a page.

We wanted curation.

Every site was picked by a person. We tried it first, decided whether there was something worth keeping, and added a bit of context around it. A website did not have to be productive or important. Useless but entertaining was perfectly acceptable. Sometimes preferable.

The product was meant to serve two groups.

For people exploring, it should make the enormous internet feel smaller and more interesting.

For creators, it should provide another place for a side project, experiment or strange little website to be discovered by people who might actually care about it.

That balance — discovery for the visitor, attention for the creator — became one of the main ideas behind Bored Hoard.

One year later

After a year we had accumulated a sizeable stash of websites, rebuilt the product, survived unexpected traffic and met a lot of people who seemed to miss wandering around the web as much as we did.

The product would keep changing. The weekly boxes would not last forever. Features would appear, disappear and occasionally turn out to have been terrible ideas.

But the underlying reason for building Bored Hoard was much simpler and has aged better than most of the implementation around it:

the internet is enormous, and the interesting bits are surprisingly easy to miss.

That seemed worth building around then. It still does.