Lessons From Making A Beach Holiday Website

For most of the last half of 2016 I was working with a startup called BeachFix on the first non-beta version of their product. The idea behind BeachFix is simple: give people the tools to find their dream beach holiday.

The product is built atop a mountain of rich beach data, a lot of it gathered by BeachFix staff themselves while on their travels.

I was brought in to work with the team on applying the lessons they had tirelessly learned over the course of the beta period. I was tasked with coming up with design ideas and then to put into action as a device independent web application.

The team deserve maximum props for all the work they put into measuring how people interacted with the existing website before I even arrived on the scene. They could answer almost every question I had for them straight away, and if they didn't have the answer immediately, they would usually just bring out the data. This made life easier for me.

Thanks to the team's tenacity when it came to research, they were able to identify patterns in user behaviour that I think we were all surprised to see. Coming up are three things that shook up my idea of how people behave when looking for a holiday online.

Lesson Three: Users Window Shop All Year Round

It came as a surprise to me—as someone who spends as little time on holiday websites as necessary, and assumed everyone did—that some people actually quite enjoy browsing for holiday destinations as far as a year in advance.

I thought this was particularly remarkable because of how unremarkable most of the big holiday websites are (more on that below). Why would anybody want to spend longer than necessary on one of these noisy websites? Its difficult to tell. But we observed the same users returning often, running different searches, and leisurely exploring different combinations of search options.

Lesson Two: Be Disciplined With Data

Over the course of the project I developed a theory about holiday websites... There is such a massive amount of data available for use with holiday websites that if you don't draw a line somewhere, before too long your website or app will start looking more like this:


Which might explain why most holiday websites do, in fact, end up looking something like that. There seems to be an epidemic running through most holiday websites of 'more is better'.

This lesson isn't exclusive to holiday websites though. It should be applied to everything we make. The reason we design interfaces in the first place is to arrange and filter information so that it can be used constructively by humans. When information is shoehorned in to a UI, it ends up detracting from the overall goal of the product.

Lesson One: Measure Everything

This is another one that isn't necessarily unique to holiday websites, but given that users of holiday websites are likely to spend big chunks of time perusing, you'd be missing an opportunity if you're not using that to measure what works and what doesn't.

The most valuable (and slightly frightening) tool that the BeachFix crew introduced me to was something named 'Full Story'. Full Story is like a window into the user's device while they use your website, kind of like you're peering over their shoulder. We used Full Story to catalogue any tripping points and places where user's would repeatedly leave, and used the information to improve those parts of the product. This is also known as the build measure learn loop.

If you're interested in a deeper dive into the workings, successes, and failures of BeachFix, you should read this article by Nishul Saperia (the CEO of BeachFix) over on Medium: My first startup is worth $15Bn now. My latest one failed though. Here is why..