The secret of my dot.com success

When Boo.com went bust - blowing millions of pounds of investors' money in the process - Steve Homer could only smile to himself. For he's managed to set up a Web business on a fraction of what Boo burned up. Here he tells how it can be done

Boo.com is dead and the dot.com stocks are in the toilet. So lets all enjoy a little schadenfreude. But just because an online clothes retailer managed to burn its way through nigh on £100m, it doesn't mean that the party is over.

Boo.com is dead and the dot.com stocks are in the toilet. So lets all enjoy a little schadenfreude. But just because an online clothes retailer managed to burn its way through nigh on £100m, it doesn't mean that the party is over.

The internet revolution still has a long way to run. If the distance from the old economy to the new is a marathon, I doubt we have come the first two miles. But for all you would-be Amazons, might I suggest a slightly novel approach?. Don't look for £10m in venture capital; build your business yourself, on the cheap. You can set up a pretty complicated Web business on a fraction of what Boo.com burnt up.

The editor of Network asked me to write about how I set up a successful Web company without going bust. We will come to the successful bit later, but the reason QAZ.com has not gone bust is simple - it's been done the old-fashioned way. Hard graft, scrimping and saving. Not the fancy investment-banker route. No IPOs at dawn. We haven't spent much money and what we've spent has been our own, so it has been managed frugally.

At the beginning of November 1998, I interviewed Jeff Bezos, the founder of Amazon, for these pages. I liked the guy so much I wanted to go and work for the company. In trying to convince the company to employ me I bought my first internet address. By the end of the month I came up with my first internet idea - an international directory of gift websites that is now called TheGlobalGiftGuide.com. It was born of my frustration in trying to send presents to a girlfriend who lived in California.

I knew nothing about putting websites together, but I was put in touch with Steve Browne, an ex-journalist who runs dvdupdate.com. For next to nothing he helped me put up the site in about three days. Over the following two weeks I fleshed it out and Gift-Net.com, as it was then known, launched.

Today the TheGlobalGiftGuide.com has about 300 sites in around 30 countries, but it is still really in a pre-commercial phase and needs a sponsor to help get its name known around the world. But I learnt an awful lot setting it up.

Mainly I learnt the basics of HTML, the clumsy and occasionally tricky language websites are written in. The big discovery here is that HTML is not all that complicated. Stick to the basics and you really can do it yourself - and a simple site can be faster and easier to use than some of the horribly overdesigned sites that are making the rounds.

In building my gift site, I realised that there were a lot of useful websites that I was not easily able to locate with traditional search engines like AltaVista and directories like Yahoo. What I was looking for was a nice, clean list of sites that, for example, sold CDs when I wanted to buy a CD. Or I wanted to find a nice listing of wine-selling websites when I wanted to order some wine, jobs sites when I wanted to find a new job, house-selling sites when I wanted to buy a house. It did not seem much to ask, but no one seemed to supply such a service. So I decided to build it myself.

I wanted a short, easy-to-remember name (Gift-Net.com had taught me to avoid the dreaded hyphen). I looked around but could find nothing I liked in available domain names. I liked QAZ.com (Quick Access Zone) because it was made up of the three left-most characters on the keyboard and it had a nice look to it.

As it turned out, buying QAZ.com was an absolute godsend. It cost me $1,500, but Jim van Johns, the owner of Infa.net who owned the name, got interested in what I was trying to do and, for the fun of it, created the addressing system at the heart of QAZ.com for almost nothing.

The addressing system is unique. You say what you are looking for before you even get to the site. How? Simple. You make the object of your search part of the internet address. To go to the CD page, and you type in cd.qaz.com. For books, book.qaz.com etc. The extra little wrinkle is that there is no dog.qaz.com, but if you typed that in, the addressing system delivers you to pet.qaz.com.

It took a year of hard work to build up QAZ.com. The site officially launched at the end of April with over 100 sections, some 600 aliases and with in excess of 1,500 comprehensively reviewed websites.

Nearly all the reviewing was done by a wonderful young English graduate who, for three months, sat looking at websites eight hours a day, five days a week. It nearly drove him mad, but it got the site built at a very reasonable cost. Inevitably, he has a small percentage of QAZ.com.

Now to the technology behind building the site. A small firm called DDPLUS created the content management system for a very reasonable fee after our one big expensive mistake. I believed the Microsoft hype that its Office products would work together. I foolishly assumed its Access database software would integrate with its Web authoring package, FrontPage. It cost me about £3,000 to discover that this approach was doomed. Anyone trying to build a business is going to make mistakes. This was my big one.

QAZ.com has cost me about £15,000 to build. Five grand of that has come from friends and family. The rest I have put in from my earnings as a journalist and an internet consultant. I have also put in an awful lot of time and effort. I suppose part of the cost that has to be weighed up is virtually no social life and no partner for over a year.

So, is QAZ.com a success? In the first month, our launch quiz attracted 3,400 entries - not bad for a site with no real marketing budget. The feedback has been fantastic. Comments left in our simple guestbook include "I loved Google, I loved Ask Jeeves; but you got me exactly what I was looking for, first shot" and "I like the idea very much. The interface is great". A regular stream of complimentary e-mails, saying that we have found a novel and useful approach, make me believe we have a future. In the first month, nearly 300 people registered on the site so they can be kept informed as it develops. Not a lot, but at this rate we could easily have 10,000 people registered at the end of year one.

QAZ.com is already roughly breaking even. We have started charging £29 to review mainstream sites and gain a small amount of revenue from referenced sites. How can we break even then? Because where Boo.com was burning millions a week, QAZ.com eats up less than £200 a week.

The job now is to build up our traffic smartly and cheaply . Yahoo.co.uk delivers 200 million Web pages to visitors every month. We would settle for 1 per cent of that - not, I think, an unassailable target. More importantly, we need an investor to help move QAZ.com, a WAP site (provisionally named wapmama.com) and our other ideas forward.

Ideally, I want to take QAZ.com's unique approach international and that might well mean moving into the big league. No, I don't mean £100m but I might mean £5m-£10m. It will cost about £15,000 to build the site in each country, which would reach break even within a year. The rest would go on advertising. That, though, for the moment is a pipe dream.

I have built QAZ.com in a very traditional way. Napoleon called us a nation of shopkeepers and I have built QAZ.com as if it was a corner shop. I have worked hard and long behind the counter. If you have a dot.com idea or you want to move your company on to the internet, I cannot recommend this more rational approach too highly. Work hard, don't rush too much, read a lot and don't spend too much money. I can't guarantee that you will end up with a dot.com success, but I can guarantee you won't end up wasting millions of other people's money.