Oracle sheds light on its acquisition strategy

In the applications software business, acquisitions are common, but Oracle has taken the practice to an unprecedented level.

Over the past year or so, Oracle has bought some 40 software companies, from small players to very large ones, such as JD Edwards and PeopleSoft (which had already merged before Oracle bought them), as well as Siebel. Along the way, there have been other large purchases, like that of Hyperion, the business-intelligence software supplier bought for £1.7bn in March 2007. Most recently, the company has made efforts to get hold of systems software company BEA Systems.

After its initial bid of $6.66bn (£3bn) was rejected, Oracle came back in January with a renewed bid of £4.3bn, which was accepted by the BEA board. However, Oracle still has other hurdles to overcome before the deal is complete, such as any possible objections from the European Commission.

ZDNet.co.uk sat down with Oracle's senior vice president for applications in the UK, Stuart Turner, to discuss some of the issues faced at a company that is filled with people who are used to working at other businesses, with different policies and strategies.

Q: So, Oracle has been pretty hungry recently?
A: We have been going on with acquisitions, and we are pretty close to 40 acquisitions in 14 months. That is a lot of acquisitions, and it all hangs together. It is a very exciting place to be and there is a lot of growth around. There is a lot of growth from an organic point of view, as well as an inorganic point of view, with all the acquisitions. The important thing is that both are growing.

The size of our user base now gives us a very fertile ground for selling our acquired products, and there is also a lot of cross-sell between the acquired products and the existing products. It does appear to work. It does appear that we are able to increase the gap between ourselves and our main competitor, SAP.

Can you put an estimate on the gap between yourselves and SAP?
We became bigger than them two years ago. The gap has gradually been growing bigger since then, but we haven't run a specific proof point of the size. We are doing that work and we will run some specific news when we have it.

It must be an interesting time for you because, at one level, SAP is a direct competitor and, at another, a very large customer?
Well, if you picture a big SAP customer, they are at the centre of our attention. Take a typical customer: they are a big customer of ours on the one hand, [but] then, on the side, you find they have PeopleSoft. Then you find that, on the other side, they have Siebel. And they all run on an Oracle technology footprint. Then you find that, even in banking, if we said it was a banking customer, then over the top they have Hyperion.

That is an example of a user where we used to say: "They are 100 percent SAP" and now we say: "Actually they are 20 percent SAP and 80 percent other Oracle stuff".

So now we have to ensure that our products work well with SAP and add value to the customer. You have to be cognizant of that, and some of the products we are releasing, especially around the applications integration architecture, are helping us do that. So now we are building out integration naturally into SAP.

It is obvious that, at some stage, at the right time, when it adds value to the customer, we would want to make the play for replacing SAP. But it must be when the time is right for the customer.

So, meanwhile, you have got this strategy of ring-fencing SAP, wherever they are, wherever they are going to look?
Yes, and it just seems to work.

It is a bit like the database wars, the war between Oracle, Ingres, Informix and Sybase in the middle of the 1990s. People would choose Oracle or they would choose Ingres and it was almost a cultural thing. Once you had chosen one of them, you would never need to cross to another. Now, when you get a bit ahead of the competition in terms of what the software does and the reach, then partners build on top of it, and experience tells me that you get a long way ahead very quickly. It becomes a little exponential.

You must have collected a lot of good people from all the acquisitions. Do you keep them?
What we have found is that a lot of acquisitions involve losing a lot of people. What we have tended to do is keep as many as possible of the technical people, the support people, the consultancy people and the sales people. Normally, when companies go into sales mode, they tend to rationalise sales teams. What we have found is that there are not enough sales and accounts people.

We want to keep as many as possible inside the company, and, in all those categories, we have got a very high retention rate. People have an expectation of Oracle but, once they get inside, they see that we put a lot of support behind...

...teamwork and development, and we put a lot of store behind making sure that we are coaching people from a career point of view; it is not just performance-related.

So, explain a little about how these acquisitions are being developed and moved forward.
The first one we did was JD Edwards and PeopleSoft. JD Edwards had been taken over by PeopleSoft the year before and their people felt pretty aggrieved. That was the first one and we didn't do as much as we needed to do at that point in time. We were a bit new.

Where we are now is that we have listened to the users. The user-group survey has gone from saying the PeopleSoft and JD Edwards customers are the least happy to [them being] the most happy. That is for three basic reasons.

Firstly, we have released versions of every product. We released new versions of PeopleSoft; new versions of JD Edwards EnterpriseOne; but also the first release of JD Edwards World since 1998. So they have seen that the company is serious about investing in those products.

Then we picked up on the fact that they were less happy than the Oracle customers and we focused people on resources and programmes — so on changing that [perception] and listening. They have obviously reacted really well to that; hence the survey results.

Finally and most critically, from a customer point of view, is that they see us signing new-name business using those products — so, new PeopleSoft customers that have never been PeopleSoft customers before. The same [goes for] JD Edwards customers. They realise that these are real products with a future that we are going to market with in a serious way.

Some products are PeopleSoft products that fit the legal market really well and [some are] software from JD Edwards that fits the manufacturing market really well, so it is horses for courses. Nevertheless, we are looking at the JD Edwards user base, for example, and asking: "Where are the other customers who are similar, so we can go and sell this to them?"

Some people are not very happy about the way Oracle licenses its software, and some draw comparisons with the on-demand model and its inexpensive licensing. How do you feel about that?
We have to make sure that customers understand what the licensing is and why it is like it is. At the end of the day, you have to have some way of doing the licensing. What works for some people is more difficult for others. What we want to do is make it applicable as far as it goes and accountable as far as is possible.

We have recently come up with some "unlimited deployment" licensing, which some of the bigger customers have bought into. That takes away all the worry of counting anything.

How does that differ from site licensing?
The one thing about site licensing is that it is based on "what it is now". If you are in a fast-growing company, then site licensing has typically been restricted to the size of the company as it is now. The key here is unlimited deployment, so, if you double in size over a period, say three or four years, then it is based on that expectation. You can deploy as much as you like in that time. Then you have put a peg in the ground and said that is what the answer is and, if you deploy it, it's yours, and, if you don't deploy it, then we stop doing this. The good thing about that is it makes it very predictable. Licensing is predictable and it especially makes the support predictable.

So that licensing is over a period of time. At the end of that period, we can review it or whatever?
Yes, and that sort of thing seems to work really, really well for technology. From an applications point of view, then necessarily you have to be a little more variable. So, for instance, for transportation management, that tends to be done under a metric based on the freight and the management — something very specific and a metric that the industry uses all the time.

How about the on-demand model?
Well there are two parts to that model. The first part is running the systems at the customer's premises. We do some of that. We provide our products as a service on top of the standard products. We provide it running at our partner's [premises] and off-site. Finally there is software as a service [SaaS], and we already do that with our CRM product, which is a multi-tenanted offering. That is going really, really well. You will see some of our partners in a year or so offering SaaS as part of our applications — some of the bog standard applications.

The older applications?
Well, in terms of the Oracle E-Business Suite. "Additional" is a better word.

So that means you get all the upgrades and so on?
Yes.

It is a few years now since Salesforce.com popped up in the area, but how do you see the CRM field panning out?
CRM for us would be on-premises, where we have been pretty successful in the UK. The advent of Siebel just changed the game entirely. We are easily the leading product on-premises. When it comes to the on-demand model, then Siebel coming into the fold really gives us an outstanding competitive product [with which] to go and compete with Salesforce.com.