That one time I made a jobs website and failed. Hard.

Joboti.com – “it makes job hunting easy”

This was an exploration into building a complex website during the evenings and weekends. I was approached by a friend who wanted to turn his idea into something real and I agreed to join in.

Joboti.com was all about finding work locally in the form of small one-off jobs. We wanted to address the unemployment rate in the UK by linking people together with small jobs (it was 2010 and we called it ‘social recruitment’ and felt pretty darn good about it).

petervanlanschot.com_joboti_screenshot

The scope for the project almost immediately ballooned into a full service recruitment website that allowed the user to search for a selection of different job types: from one-off jobs through to contract work and permanent vacancies in local businesses.

Employers could post all types of jobs and choose to make the payment type negotiable which allowed candidates to offer specific skills in return for completing the job.

For example,

“Drive me to London and I’ll teach you Spanish during the ride.”

or

“Help me move house and I’ll buy you a crate of beer.”

or even

“We need a new logo and menu designed for our cafe.”

Without testing it for a market fit, we expanded the scope of the functionality to fit into an incredibly competitive industry with big dreams of addressing every single type of customer need (both job seeker and employer).

Blog-Image-2

What’s more, the business model was wildly optimistic; Joboti was totally free to use. Its main aim was to appeal to people in their 20’s who were looking for a range of different type of jobs in their local area. The idea was to only charge once we got to a level where we had enough candidates to entice recruiters to pay to look at our CV database.

So many mistakes, so many lessons learned.

Bearing in mind it was 2010 and I was eager but still relatively inexperienced. I worked on this website during my evenings and weekends, and it very soon became the only thing I did after working all day on large site for clients such as Sony, Waitrose, and Nokia. I dived in head first and decided to really nail this like I would with a big client account at the day job.

In hindsight, this was my biggest mistake. With no thought to an MVP or basic user requirements, I launched straight into creating a site map, a complete set of user flows, and a wireframe document detailing full functional specifications.

sitemap

This was an expansive site with user profiles, on-site emailing, and search functionality for jobs and candidates. I used our limited cash savings to vet and hire an overseas development team, and a visual designer.

The build was lengthy and complicated. I still wonder why didn’t I make an MVP first? It could have easily been an app instead! Too much focus on the dozens of wireframes and stressing over the finer details of the build with no room to think about the growth plan and distribution model.

 

It was like taking a sledgehammer to a walnut. There was no need for such detailed wireframes and functional specifications.

Instead, it was endless evenings of bug reports and checklists sent back and forth between myself and the development team.

wireframes8

 

The vision was to keep the site free for both job seekers and job posters. In that way we could challenge the competition (Gumtree.co.uk is essentially the Craigslist of the UK) with an easier to use and better looking service. We would target certain areas (“hyper-local communities” we called them) and build up a user base by offering an email newsletter for when the type of job the user wanted became available.

wireframes7

Blog-Image-1

In hindsight, it was a big learning curve about how to focus on the wrong things for the wrong amount of time.

I got lost in the detail and totally missed the concept of starting small and building features when the research proved that they would solve a user problem or take advantage of an opportunity.

In 2014 I revisited this idea when a conversation one evening started revolving around how simple would an app have to be to barter for things that you either have or want.

“What about a Tinder for jobs?”

I made some quick mockups of it, but didn’t take the concept further. The experience of Joboti had left me wanting to explore areas other than recruitment and the jobs industry.