Becky’s Affiliated: Ian Sherrington’s journey to creating the world’s first online sports betting software

ian-sherringtons-online-sports-betting-softwareThe world’s first online sports betting software was launched in 1996, built by Ian Sherrington who has been a part of the industry from 1983 until today. That’s thirty years of programming (and partying) with online gambling industry professionals, building businesses, moving across the globe and landing as the head of EveryMatrix’s new Manila office.

Someone who is responsible for such an important “first” in our industry must have a great story, so I sat down with Ian to learn more about his journey to launching the world’s first online sportsbook and accepting the first online bet.

Becky Liggero: Ian, thank you so much for joining me, it’s a pleasure to be in the company of an industry icon.You spent 20 years working for Intertops, a sportsbook that started off as a mail and phone betting business, what were the triggers that made you decide to make the leap to online and create the first ever online sportsbook?

Ian Sherrington: Yes, I had the honor of working for Intertops for 20 years from 1983 until 2003. I’m so glad I had that opportune meeting with Detlef Train in a pub in London at the beginning of it all in 1983. I was a Computer Studies student in London at that time and I bumped into Detlef one evening at my local pub in Pimlico. Detlef was on a mission to become a Bookmaker in London and needed some local help to do that. Detlef is an immensely likeable fellow and so I gladly helped him start his business, find premises and started to write the computer systems that would make the process of accepting and processing bets as efficient as possible.

Intertops’ original business was to accept bets by mail and telephone from German customers. A customer would be sent a coupon in the post each week, which they duly returned to Intertops and their bet would be accepted. The customers could also call Intertops to place a bet. Most of the bets were football parlays or ‘Kombi’ bets as they were known. As the odds were printed onto the betting coupons, they were fixed and could not be adjusted according to the action that was taken. Some days were good, some bad.

However, for me it was all about progress and efficiency. Intertops’ business grew and the computer systems had to become more capable and even faster to handle it. It all started with a single PC running VisiCalc, which was never going to be good enough. The first true Intertops programmed system was written using dBase III+ in 1985 which provided the company with a vaguely scalable solution.

Eventually, Intertops installed networking and Novell servers. The biggest impact on the business was the introduction of computer readable betting coupons using Optical Mark Reading (OMR) machines and a specialist printers / copying process. Business grew amazingly and at one stage in Intertops, I recall printing and mailing over 250,000 coupons to customers in one day. Intertops grew so big it had it’s own postcode.

“So where are we going from here?” Detlef said. It was 1994, Detlef and I were on a fishing trip in The Netherlands,Intertops was ticking along nicely, we had achieved all the goals that we had set ourselves. So why not cause some trouble.“Well”, I said, “there is this thing called ‘The Internet’. Wouldn’t it be nice if the customers could place the bets themselves without any coupons or telephone operators? Transaction costs would be negligible and the customers would have much more control over their accounts.” “So how do we do that then?” he said.

BL: That’s a great story, Ian.  How long did it actually take to create the software, what kind of resources did you have?

IS: It took a while. Firstly I realized that the system currently in place was never going to handle the massive amount of traffic that the Internet could bring. At the core of everything is the database of course and I had to go for Client-Server architecture. Also, I had to decide on the software that I would use to create the user interface for the internal systems such as Bet Management and Customer Accounts as well as for the future web servers; and not forgetting the printers and the OMR machines. Once I started on the conversion, I had to follow through. Finding examples of hooking up a web server to a database proved impossible at that time. No one had really done this before.

It was a herculean challenge. But in the end I chose to base the system on Microsoft SQL Server and used Visual Basic (VB) for the programming. The web server I chose was called ‘Website’ originally enough and was created by O’Reilly (yes the publishers.). Nice thing was that it connected up VB to the web processes using a simple CGI.

So there it was. All of the required systems could be written in a single language and was simple enough for me to understand.

The programming for the core Intertops systems took nearly a year to complete and then the conversion from the old systems to the new took 4 weeks over a summer period. Once this had all bedded down ok, I started on the web site itself. That took about 4 months and by January 1996 we were ready to launch. Really, this was all done on a minimal budget and I only had myself plus a graphics guy to help me with the web site. I had no doubt that this would be a technical success, but would anyone actually bet over the Internet? No one knew.

The website went online on 17th January 1996 and a group of us eagerly watched the hits on our web site monitor. Traffic steadily grew. We knew that when a bet is placed and accepted, it would also be printed on a dot-matrix printer as irrefutable proof. No one trusted much at that time.

The very first bet was placed on 17th January 1996 at around 6pm by a Finnish fellow called Jukka Honkarvaara. A dot-matrix printer never sounded so good…

BL: Ha!  Yes, I’ve heard of this infamous Jukka.  What were some of the biggest challenges you faced before actually taking this first online bet from him?

IS: Intertops had a very reliable and efficient business system. It was making money and everyone was used to how it all worked. And now here comes the madman that wants to change everything and cause mayhem.

Yes, I had a little resistance of course. I had to replace the whole system so that the transactions were handled by a database server and not on the rather dated dBase III+ flat file system operated. The system had to be infinitely scalable with no one device being the bottleneck – at least it should be capable of being so. This would enable the system to process many more transactions per second.

So the conversion from the old system to the new was the biggest issue both technically and politically. Lots to learn, inevitably many mistakes were made and plenty of nerves trampled upon. Once everything was ready to go, after one year of work and four days of no sleep, a radio suddenly sprang to life. It played the song ‘Que Sera Sera’ by Doris Day. It has been my theme tune ever since.

BL: Whatever will be, will be… And it was a success.  If you think back to this software that you created back then and compare it to what we have today, what reaction do you have?

IS: In those days, HTML consisted of 20 tags. Not many. And they easily fitted on to one page of A4 with descriptions.That was the easy part – the issue really was the ability to accept huge numbers of bets at peak times, especially just before NFL kicks off on Sundays. As the hardware in those days were not so zippy – servers had 386 processors – I had to create (for example) a queuing mechanism so that bets were accepted one at a time and not all at once as originally designed. This prevented the database from getting locked up but also delayed the acceptance of bets, which annoyed the customers a bit. The system design was correct and the architecture was scalable (I maintain) but I was slightly ahead of my time. These days, systems are so much faster and more capable of handling high volumes of transactions. The techniques I used early on would most probably not be applicable these days, but I don’t see how I could have done it in any other way at that time.

As for the web pages themselves, the introduction of Javascript and especially AJAX techniques have made such a difference. Features such as floating Betting Slips and dynamic odds pages are possible now, which simply could not be done in the early days without refreshing the page.

Of course nowadays, Sportsbook web pages can be delivered from specialist suppliers such as EveryMatrix and inserted into any web page using iframes. As an operator you don’t ever need to worry about databases again really. So much better now.

BL: If that much has changed over 20 years since you launched Intertops’ online book, what do you predict we’re going to see in 15- 20 years time with online sports betting software?

IS: I did some work recently on a P2P betting exchange based on mobile devices. I think there are many opportunities in that area still.

However, the biggest opportunity I see is in the way that a punter interacts with betting. With every new development in the user interface area comes new ideas of how bets can be placed. For example, the new wearable devices, certain movements could be interpreted to be a place bet. This would make possible faster placement of bets and a seamless involvement in live games. A totally immersive experience could be possible whereby the punter is actually part of a football game. To place a bet, a virtual betting panel is available in a virtual 3D environment. So many options. Ah…Ok leave this one with me and let’s see what I come up with.

BL: Ok, we will.  In the meantime, you’ve made a move to the Philippines as the head of the EveryMatrix Manila office – why Manila?

IS: Simple. The Philippines offers online gaming licenses and is the only place is Asia that does so. Most of our Asian clients are based there and it’s the right place to be for EveryMatrix.

BL: Great Ian, thank you so much- wishing you the best of luck in your new venture.