Why SOA is the 'perestroika of IT'

Software AG's goal is to become a €1bn company within three years — and, according to chief executive Karl-Heinz Streibich, SOA will be a key engine for growth.

The German software company started life in 1969 selling the Adabas mainframe database-management system but, by the mid-1990s, the world seemed to have passed it by. It started the slow march to modernisation in 1997 when it introduced its EntireX integration middleware. Six years later, Streibich joined the company as chief executive and deepened its focus on middleware, steering it towards the nascent SOA (service-oriented architecture) and BPM (business-process management) markets.

A major move was the $546m (£350m) purchase of integration specialist WebMethods last year. WebMethods now acts as the company's SOA and BPM business unit, while its enterprise transaction systems unit handles mainframe application integration.

Rivals SAP and Oracle see their applications as a Trojan horse in corporate accounts, as they introduce an underlying SOA framework into which all other SOA components will plug. Unlike them, Software AG believes customers prefer to take an independent middleware approach in order to avoid lock-in. But it's not clear which approach will predominate — if either — as enterprises may choose one or the other, depending on their focus. Whichever way the market goes, what happens over the next few years will be crucial.

ZDNet UK caught up with Streibich at Software AG 's Business Innovation Forum 2008 in London this week, to find out how and where the company intends to make a mark as it makes its bid for growth and leadership.

Q: One of your key business focuses at the moment is SOA. But Gartner has indicated SOA implementations have halved over the last year, due to the poor availability of staff with the right technical skills and general difficulties in building a viable business case. What impact is this likely to have on the company, and how do you intend to address it?
A: Regardless of quarterly movements up and down, SOA is a paradigm change, not a trend. It's an inevitable movement, like the one from mainframes to client/server, which also had its ups and downs. So SOA will happen over the next 10 years, not in 10 years, and I'm sure it will be the IT industry's dominant theme.

The time of silos is over. Silos cause more problems than opportunities, and so integration and automation are a key element of SOA. But SOA is a destination. There are no big-bang SOA projects. They're step-by-step and are driven by specific opportunities. So they're incremental in their nature, and that's their value.

But with SOA, you don't just implement a technology solution. SOA is a business value architecture and capability, and that's what has been put into WebMethods. Our focus is on reducing the total cost of ownership of a middleware platform, which is driven through ease of use. When we implement a proof-of-concept at a customer site, we take two days with three or four guys, while our rivals take 10 days with 10 guys, because we have the benefit of simplicity.

So 'getting there faster' isn't just a marketing slogan — it's realising that a step-by-step approach is key. It's Software AG's key differentiator.

But we also consider long-term relationships to be important, because we believe the long-term journey to transforming the business to SOA requires long-term trusted partnerships. The City of Philadelphia, for example, has been a customer for 35 years.

How much of a threat do you expect Oracle and SAP to pose in the SOA space? Some analysts believe the fact they are building their Fusion and NetWeaver middleware into their applications, to act as a default SOA framework, may make life difficult for independent middleware vendors.
Middleware is the new agility hub, and the last thing that customers want is to be subject to monopolies. SOA is the perestroika of IT. The CIO of Volkswagen said, for example, that he doesn't want an IBM or Microsoft SOA, but a Volkswagen SOA that matches his requirements.

So SOA allows people to partially get rid of the monopolistic grip they have experienced in the past, and people are very aware...

...of the opportunity, which only happens once every other decade. They need interoperability between proprietary platforms, between sub-frameworks and departmental frameworks.

This is important because in a more complex world, companies need more flexibility. Businesses have to be run from a more entrepreneurial viewpoint, and managers have to be empowered to run the business.

Your stated aim is to become a €1bn revenue company by 2011. Is this still achievable in a recessionary climate? If so, how are you going to do it?
The aim is for 10 percent top-line growth per annum across all areas of our business, and we see four key pillars for this growth.

The first is innovation: Gartner says we are market leader in almost all our product segments.

The second is geographic expansion: we operate in 70 countries and have a globally diversified presence. Phase one saw us take over the business of our distributors in countries like Brazil, South Africa and Japan, and the next phase is about exploiting those customer bases more effectively. The recession hasn't affected all countries of the world in the same way, and we're seeing exceptional growth there.

The third pillar is partnerships. We're focusing on expanding them, especially in vertical markets. For example, in the UK, a third of our business comes from the public sector, a third from finance and insurance, and the rest from a mixture of areas such as retailers, manufacturing and services companies.

Surprisingly, business in the finance and insurance sector hasn't been hit so far — in fact, the reverse is true. Banks rely on technology, particularly things like trading systems, which handle high volumes of transactions. Market volatility means that the number of transactions is going up and so people need more software to make more money.

The final pillar is mergers and acquisitions. We're interested in purchasing complementary products, customer bases and developing our professional services business, which currently contributes less than 30 percent of our revenues, to support growth in all countries and in all segments.

Does this mean your SOA and BPM business is now more important than your mainframe-integration one?
The mainframe is our core product segment, and it helps that maintenance is an operating expense, not an investment area. Companies in sectors such as utilities and telecoms have proved relatively stable due to their business models, and their share prices haven't declined to the same extent as more investment-dependent ones. So some of our product segments are in more stable areas. But growth areas focus around WebMethods, which we acquired 18 months ago.

How important do you think cloud computing will be, and what will be your response to it?
Cloud will have its place. It's similar in concept to software-as-a-service [Saas] and application service providers before that. The idea is that you share your applications with other users and someone else operates them. So it's an application-oriented idea, but we provide business infrastructure software. It's the SAPs, Microsofts and Googles that will focus on that, but we could sell them some backend infrastructure so it may be a market opportunity for us.