The 24-hour Fake Startup

A few weeks ago, Sophie (a fellow Project Getaway participant) and I both launched a business within just 24-hours (actually, we had both hit out goal after just 18-hours).

We both had come up with ideas for products that we thought could be interesting to sell online, but we didn’t want to stake a large investment in a complex website or sourcing inventory. So we formulated a plan to instead ‘fake it’. We created an online store, showcased some mock products, promoted the site through AdWords, and waited to see the results.

This concept takes ideas from Tim Ferris and the Lean Startup; namely the concept of minimum viable product (TechCrunch has a nice piece on how Dropbox used MVP to create demand without a working product), whereby you launch a business with the very bare bones of a product or service to test the waters before making a larger commitment.

My venture centered around fascinators; those fancy hair broche type of things that have seen a boom in the UK since the Royal Wedding. I figured that they could be sourced cheaply from China, shipped over to the UK in bulk and sold online at a price that reflected good value.

After a quick bit of brainstorming, Sophie & Fox was born. We used the superb Shopify platform to launch our webshop, taking advantage of Paypal for simple payment processing.

Here’s a screenshot of the Sophie & Fox homepage (the site has now been put offline). I made use of a standard template, with a bit of customisation and a custom-made logo. I’m pretty happy with the look; it feels very professional and slick.

I spent most of the day hacking the template to make it more bespoke, working with a graphic designer on some elements (which turned out to be quite painful) and adding some products to the store.

We whacked together a super quick AdWords campaign (featuring the most obvious broad keywords, coupled with a few ad texts and some negative keywords) and set it live immediately with a budget of $30/day. Traffic acquisition through Google might not be commercially viable in the long-term, but it served us perfectly for this experiment – if we can’t convert highly targeted visitors, then it’s unlikely that there’s demand for our product.

This Google results page shows the Sophie & Fox ad on top for the “fascinators” keyword.

After $101 and 262 clicks of ad spend, I had received just 1 order. (Of course, I refunded the payment immediately and apologised for not being able to supply the product at the time).

My experience suggests that fascinators aren’t a great product to sell online. Of course, perhaps a more conversion-optimised page could have performed better, and building the brand would help enormously, but I was hoping to see a much higher conversion rate which would show a much greater opportunity to work with.

But I wouldn’t consider the test a failure. I had only spent a hundred bucks and a day of my time to turn a seed of an idea into fully fledged proof of concept. I feel confident that I could now go on to test out other products in the same way and I’m sure after a while I’d find something that sticks.

Sophie worked on building Sticky Inspiration; cool little Post-it like notes with inspirational quotes that you can stick anywhere. Her test continues.

 Tags: Business   Published: 21st November '11


2 Archived Wordpress Comments

Dilanka Wettewa

Hey Keith:

Pretty neat dude. I am running about 2 experiments myself at the moment. I see a potential issue with your testing phase though (In addition to running maybe 2-3 versions of the page to test the page itself): ADWords can be a bit expensive to run tests – especially for competitive keywords. It might be a good idea to incorporate Facebook ADS / BING (Can purchase the same/similar keywords for cheaper) – and my two favorite: StumbleUpon and SurveyMonkey :)

I use StumbleUpon first – not necessarily to make sales, but to see the overall response rate (I don’t spend more than $20-30 here), then move onto SurveyMonkey and Bing/Adwords. Also I have bought (although un-targeted) cheap traffic from sites like fiverr.com just to test response rate(s). As for your specialized idea, I might even try advertising in similar/competitor(s) sites if possible!

Good experiment dude. Can’t wait to hear about your next one!
Dilanka :)

Keith Mander

Thanks for dropping by Dilanka. Super feedback; I’ve never thought about trying StumbleUpon – I must give it a go for the next experiment. I agree that $100 is still quite some money for an experiment and I should explore ways to reduce it. That said, it’s still obviously much cheaper and risk-free than the alternative of taking the plunge and producing the product first.