Menu
Menu
Dan Lee is The Business

Dan Lee is The Business

The responsibility of the CIO of a global trading platform vendor kaleidoscopes much of the current conventional wisdom about the value of the role

If there is such a place as CIO heaven, Dan Lee must wake up every morning believing he is there. Lee is at an Olympian distance from the often irritating debate on the whole hazy topic of whether the CIO is a blessing or burden, an asset or a charge upon the business.The CIO of The Marketplace Company is the business, and that is electronic trading platforms. And its most recognisable branded product is Unlisted, the independent second-tier share trading market.

But M-co, as the business is usually called, does more. It operates the cyber market too. “We are end-to-end,” notes Lee, chuckling that the term was once widely used by consultants and vendors to describe a commitment to just a component of a project.

M-co, in contrast, is there from soup to nuts, from go to whoa. “At no stage do we tail off,” says Lee. Indeed, M-co’s involvement starts even before the beginning. It starts at the conceptual stage. “We identify an opportunity. Then from this nucleus we get the buy-in from the market. We take the idea and turn it into a reality.”

M-co began its career trading Australian carbon credits. Then its big break came when the New Zealand government, having embarked upon a policy of power de-regulation, needed an organisation to trade the blocks of electrical energy. It was now that Dan Lee came into the picture.

He was with Logica, then a jewel in the UK’s IT crown. The outsourcer put the energy trading deal together into M-co’s Comit system, the IT node that collects information and shuffles it to and fro between generators, retailers and all the other buyers and sellers in the electricity sphere.

Comit, with its two access mechanisms -- the web client and machine-to-machine dealings – fortunately worked from the outset. And just as well. As Lee points out, all the contracting between M-co and its users was pinioned on 99.9% uptime, a benchmark still enforced by a draconian set of penalties. “Dollar penalties,’’ says Lee.

Logica, then the powerhouse of UK innovation, did its bit for New Zealand by deploying a full slate of highly qualified and experienced specialists. Naming rights were taken on one of Wellington most avant-garde high-rises. Anybody dealing with Logica had the feeling they were dealing with an institution that was not going to go away. But Logica, needed more business than just M-co, and so it went looking for opportunities in the finance and treasury realm.

Alas, growth was not forthcoming -- not immediately, anyway. Oddly, Logica was strangely reluctant -- or unable -- to focus on an even greater triumph in New Zealand -- its much earlier Wanganui Law Enforcement Centre, in which it did all the systems work with Progeni.

Two problems of success ensued from this project. First was to agree on how to divvy up the early completion incentives promised by the government. Then there was the fact that, so advanced was the technology of the Wanganui Law Enforcement Centre, it put New Zealand out of kilter with the rest of the world and was to lead indirectly to the colossal overruns 20 years later when the police sought to replace it.

Some believe now that Logica’s problem was one that it shared with another European technology icon of the era, Philips. Namely, it made the mistake of believing that the quality of the engineering would sell itself.

Anyhow, soon after Logica had got M-co up and running there came a global re-shuffle in which Logica New Zealand came entirely under the aegis of the Australian operation. The Australians promptly closed down the New Zealand organisation.

At the time the Australian management was adamant that the numbers from Wellington were not adding up. No amount of longer-term vision could sway the Australian management when Logica Australia itself had wandered down fanciful and money-consuming alleyways such as multimedia. Down came the Logica sign in New Zealand, leaving M-co with the 99.9% uptime covenant baby -- one that was, incidentally, uninsurable. It could not be sentimental. It simply took on the key Logica people responsible for M-co.

One of these people was Dan Lee, by now back at home in the UK. “You could say that I have seen both sides of the coin,” he reflects now. “I’ve been involved as an outsourcer, and now of course from the in-house point of view – and I can see just how effective that can be. We get more done, and at half the price.” The experience has left Lee with a highly selective attitude towards outsourcers, especially consultants.

In formulating the Unlisted structure off the back of the energy trading coding, Lee installed a “finance guy” to ensure that what was being done was being performed in a way that would make sense to the share trading sector. He brings in an “ad hoc” practitioner/consultant to perform the role of a “sounding board”. Essentially he sees consultants in a “testing” role. At the same time he remains wary of consultants taking becoming too embedded in a business.

When it comes to non-core business, Lee becomes a vigorous outsourcer -- especially with telecommunications and hardware.

Eagle does the hardware and Telstra does the telecoms – or, to be specific, did the telecomms. M-co is “migrating” to Telecom, according to Lee, who will not be drawn on the reason.

He neither confirms nor denies that it has something to do with the 99.9% uptime specification. It is hard to perform any consideration of M-co without returning to this easily remembered percentage. The staff of 12 are somewhat like specialist medical practitioners: they must be reachable and bleep-able at all times. Back-up services are available in Auckland, and these in turn are under replication in Wellington.

One thing is certain -- no CIO, probably not even Telecom’s Mark Ratcliffe -- works quite so hand in glove with the marketing side as does Dan Lee. As the marketing people prowl the world looking for new markets, so Lee must be ready to deliver them their new platforms. And quickly too. For Lee and his 12 there can be no lingering, pizzas in hand, Pepsis on standby, nor any infinite conjecturing sessions.

Bearing in mind M-co’s philosophy of flexibility through independence, it comes as a bit of surprise to hear Lee describe the company as a committed Oracle shop. M-co went with Oracle at a time when the software titan was hardly the darling of the user community. The business stayed with Oracle when the software manufacturer with the ferocious sales arm had a reputation still very much open for debate.

The fact that Lee seems to have enjoyed the experience indicates that 99.9% clarity of specification from the M-co side has probably sidestepped 99.9% of the problems that might have affected other users in their dealings with the ardently market-orientated software supplier.

“Of course we know Oracle’s funny games, and they know that we know their games,” says Lee. “They’ve had to pull their socks up as they’ve realised the competition from Microsoft SQL. “They’ve had to see the danger of losing market share. Over the past two years we have found them to be really good. One problem was their perpetual licence system but now they have term licences for one year.”

Lee also points out a number of other concessions such as licences geared for fewer than two servers, and so on. He gives Oracle high marks for Standard Edition and observes that Unlisted was established on Oracle for just $4000.

Lee likes to eliminate unnecessary features, formerly a black hole in the securities trading field. For example, the London Stock Exchange suffered one of the world’s landmark IT cockups when its Taurus upgrade had to be abandoned at a cost of over $1 billion to the exchange and its members. A key to the Lee approach is to have as few features as possible and arrange these so that they can be armed or disarmed simply and quickly.

As M-co diversifies into providing trading platforms for an endless array of financial derivative categories, not to mention familiar products such as milk, so Lee and his people must be ready in days to fit out the required system. Lee describes this modifying or customisation process as one of “productising”.

Just as Logica failed to have capitalised on its all-performance and no-hype cornerstone installations in New Zealand, so the government, desperate for examples of successful state technology seeding, seems uncharacteristically quiet on M-co and its global role since it was formed in 1993 to administer the wholesale electricity market.

First, though, contemplate, the following … In Australia, M-co administers the gas market in New South Wales. In Singapore it runs the wholesale electricity market. In Taiwan it set up the electricity market operator. In the Philippines, M-co last year was contracted to implement a wholesale electricity market and then run it. It is active in Africa.

In other words, M-co has achieved the holy grail of Kiwi computing: it has grown successfully on foreign soil. Databank tried it in the 80s and Telecom tried it in the 90s. But M-co has done it and not even the industry talks about it. The reason why is probably clear enough – the company has been a fully owned subsidiary of South Africa’s First Rand Bank since 1999.

Nobody at M-co talks about it. But in financial circles there are soundings to the effect that it changed hands for $7 million, a considerable sum at the time, and perfectly timed for the peak of the dot-com bull run.

Today, in the light of the world franchise that M-co is appropriating to itself, First Rand Bank’s purchase looks like a bargain. M-co’s marketplace looks infinite. “You look at what organisations do, and if it is anything fungible then there will probably be a platform for it,” says Lee. Fungible? In this context it means goods and services that can be contracted for. Fungibles under M-co’s purview could include such tradable as racing or golf club memberships, taxi company quotas or fishing quotas.

In very broad terms, Lee’s approach in rapid deployment “productising” is to take the existing product -- the trading platform software, for example -- and then convert it to its new application for its new marketplace. In the case of Unlisted, the entire conversion of existing software into a new application was completed in 12 weeks.

Lee characterises the process of building a new product from the old one as one “pulling it out and tidying it up”. One of the secrets of getting Unlisted on the board so quickly was the understanding that it did not require sub-second response times.

Dan Lee and his team are not so much mission-critical as business-critical. You know they cannot be complacent about their vaunted position. Not while that 99.9% spec continues to loom over them.

Join the CIO New Zealand group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.

Join the newsletter!

Error: Please check your email address.
Show Comments