• silicon.com
• Management
• CIO Insights

By Julian Goldsmith, 2 December 2008 14:23

COMMENT

Can a company exist with no IT of its own? Majedie Asset Management, a fund management boutique, is proving it can be done.

Based in the City of London, the company manages more than £4bn in funds using hosted applications for business critical systems. It employs a relatively new business model using cloud computing to create a highly streamlined company not hamstrung by established thinking on IT provision.

silicon.com recently spoke to the company's business manager Simon Hazlett about how the business came about, why it opted for the cloud model and how he adapted the military approach to information management to suit the needs of the City.

silicon.com: What was the background of Majedie starting up?
Hazlett: We all came from a company called Mercury Asset Management, a fund management business that grew very strongly in the late 1990s, so in 2002 five of us left, four fund managers and myself.

We set ourselves up to run a small part of very large, very sophisticated pension schemes just in UK equities, so in the jargon, a specialist boutique doing one thing but doing it for very highly sophisticated clients who are used to consuming very high degrees of information from fund managers.

It was going to cost enormous amounts of money to run IT in the traditional networked way but we needed to be able to compete with the likes of Fidelity and Schroders on the service side. The only way to square that circle was going down the ASP route and just being an investment firm and a consumer of information products rather than owning its own information factory.

Our documents have ended up in an old salt mine in Utah. We use Microsoft SharePoint but on an outsourced basis. We use an outsourced email service but we use Bloomberg as our core operational application that presents to us our portfolios and allows the functionality of trading. We have very sophisticated compliance engines on the back and straight-through processing trading engines.

Cheat Sheets

♦ Faster Payments
♦ Basel II
♦ MiFID
♦ Sarbanes-Oxley

We do have outlook on the desktop, which was the only thing that could churn through the immense amount of emails we had every day from the [Microsoft] Exchange server, but every single bit of the shared processing is happening on other people's servers.

What impact has this model had on your ability to carve a niche in the market?
First it allowed us to grow like topsy but without any loss of quality in terms of the ability to outperform for our clients.

We won our first client in 2003 and by December 2006, we got £4.5bn or so under management of some 85 or so institutional clients, all with a staff of about 15 over that period.

One reason for that is the idea that the system has no tangible deployment. Its footprint is essentially virtual so its ability to scale with that exponential growth was massively important - and it could only be done with this virtual ASP strategy.

Secondly, because of our virtual nature, our systems recognise what we call the real network between the fund manager and their client. In an ASP strategy everybody can go into the cloud and our clients can have the same access to the systems, where their information is concerned, as I do.

Because you have critical systems looked after by other people, how do you maintain control internally with no IT staff?
If you are a business front-office person, IT people are outsiders whether they happen to work in your building or elsewhere. To an extent, we don't need to control them. We are clearly very keen that they remain available and that's something we are checking every day.

We get the benefit of these suppliers' specialisations and we have a contract that states minimum uptime. We spent a lot of time enunciating exactly what it is we want these people to do but we don't second guess the way that they do it.

So who is making sure that you get the quality of service you've been promised by your suppliers?
We have an operations manager here and to some extent, myself as well, but for us the monitoring is the uptime.

It works because it's inherently more stable in my opinion than technology that's locally deployed, because we're not allowed anywhere near the pieces of those applications that make them actually work.

We're simply configuring the top five per cent of the APIs for our purposes.

That means there are none of the confliction issues of an old-fashioned network where heavy applications are fighting each other for space.

We have had, I think, a 15-minute Exchange server outage in about four years. About the same for our CRM systems and - touch wood - SharePoint has never failed.

Click here for page 2 of this interview...