Showing posts with label Swete. Show all posts
Showing posts with label Swete. Show all posts

Tuesday, April 07, 2009

HCM SaaS Provider Workday's Advanced Architecture Brings New Business Agility Benefits to Enterprises

Transcript of BriefingsDirect podcast on how Workday's SaaS delivery model for human capital management applications provides better business intelligence and architectural advantages to end users.

Listen to the podcast. Download the podcast. Find it on iTunes/iPod and Podcast.com. Learn more. Sponsor: Workday.

Special offer: Download a new white paper on Workday's latest update to System 7.

Dana Gardner: Hi, this is Dana Gardner, principal analyst at Interarbor Solutions, and you're listening to BriefingsDirect. Today, we present a sponsored podcast discussion on the virtues and paybacks from designing a strong IT architecture.

We'll examine how such architectural benefits promote business agility as a service while lowering total cost for both the deliverer -- and the receiver -- of pure application services. This perspective looks at IT architecture with a new twist, not just in terms of developing architectures on-premises, but ... for architectures that support the providers of services.

We'll take a look at how IT architectural best practices at a software-as-a-service (SaaS) provider help not only that provider's operations. We'll show how the users, the receivers, of those services can benefit in new ways as well.

By examining the experiences and approaches of Workday, a human capital management (HCM) SaaS provider, we can better understand the benefits of modern IT architecture and gaining new levels of business intelligence (BI), innovative search capabilities, and the ability to extend business processes out to mobile devices.

Here to help provide an in-depth look at how proper architecture allows the SaaS delivery model and business agility to come together, we're joined by Stan Swete, CTO of Workday. Welcome back to the show, Stan.

Stan Swete: Hi, Dana. How are you doing?

Gardner: I'm doing well. Workday has had an advantage in that you knew what your goals and objectives were when you started your architectural journey. You knew that you were going to go out as a service and you knew that there were some new modern technologies, approaches, and best practices to take advantage of.

That's a little different from a lot of enterprises that have, in many cases, had decades of IT to adjust and, in a sense, drag along with them. Let's first hear about the level set. When you started out from green field, what things went through your mind, and how was that refreshing, given that you were starting from scratch?

Swete: First, it definitely was refreshing to get the opportunity to start from scratch. I'm sure that if you talk to a lot of IT professionals, they'd all want that chance. At Workday, we had that chance and we started our company with a lot of background in what had gone before in terms of architectures to support enterprise resource planning (ERP).

We had that backlog of information, a list of what worked and what didn't work so well with previous client-server architectures. As you said, just like everyone else, we definitely had an appreciation for all the new developments in technology that make different approaches possible today.

Gardner: Today, enterprises are faced with a number of challenges. They're trying to adjust quickly for very dynamic business environments. They have to watch their costs. They'd like to modernize, but there's a significant time lag between how and when they can take advantage of these modern concepts and when they can't.

Do SaaS providers like yourself allow them to leapfrog by taking advantage of what you've been able to do and then bring those benefits into their business practice?

Complex environments

Swete: We absolutely think it does. You're right. In IT today, people are in a difficult spot. They have complex environments. The complexity has grown for a variety of reasons. Everyone sees the opportunity to modernize and to improve efficiencies, but how do you do that in the midst of a complex environment that is constraining just how aggressive you can be?

That's where SaaS can come in. If you've got a provider like Workday, or someone who's able to take a clean approach, there might be the opportunity to take the right portion of a certain set of your applications and, instead of having to deal with the complexity of managing all the multiple instances and different architectures you might have, use the unified SaaS service as a way to achieve some integration and a way to drive against cost. Today, it's all about cost.

A lot of the discussion around cost has driven IT to look at a variety of dimensions. Just the consolidation of some of the complexities and different instances of architectures that large companies all have is one area where they see opportunity, but they are bound by having to support where they have in place.

Gardner: What intrigues me, Stan, about what you're doing at Workday is that you have gone beyond the concept of just delivering an application or set of applications. You're positioning yourself as a partner for these businesses and how they can then relate to the outside world. You're extending the enterprise boundaries for them, and this notion of business agility as a service kicks in.

Explain for our audience what you mean by extending your value, not just as a provider of applications, but as an extension to enterprise architecture.

Swete: The space we're in is HCM, trying to be the core system that captures all the information about the workforce -- how it's organized, how it's compensated, and what work is actually done.

If you're trying to play that role at an enterprise, you can't be standalone and independent of the enterprise. You're going to have many, many ties into other things that the enterprise has going. You're going to have to be agile, in terms of how your solution can fit into different instances of different enterprises, because everyone has a slightly different picture and puzzle.

What that means to us is that there's a demand on the solutions to be agile, to be able to change shape and to be able to integrate to the variety of different scenarios you have. Otherwise, SaaS doesn't become a productive option for the IT professional to look at, if it's just one-size-fits-all and take-it-or-leave-it. They're going to have to make some trade offs that they can make.

So for us, if we can be viewed as just a solution that can meld itself to fit into whatever else it has got going on and to be an effective core that we can handle in the cloud for them. That's the best way for us to engage.

Gardner: You're also becoming sort of an adjunct and maybe even an accelerant to service-oriented architecture (SOA). If organizations have begun their journey toward SOA, you can provide a catalyst to that as a go-between, a services exchange, if you will, with a number of other providers.

Maybe it's payroll, maybe it's healthcare and insurance benefits, perhaps it's reaching out to partner with other organizations and the labor resources that they have. Do you see it that way -- that this is, in a sense, SOA but as a service?

Embracing Web services

Swete: Absolutely. You mentioned before about us having the ability to look at new and emerging technologies and approaches to architecture. Certainly, we've got the religion of SOA and firmly believe that the right way for us to tie into other systems in the cloud and other systems on-premise of our customers is via SOA and an embrace of Web services.

We embrace that and we think to some extent that it can accelerate SOA adoption within enterprises. Enterprises we talk to all see the technology the same way we do. They all see the appeal of newer SOA architectures, but I go back to the fact that they have the same issue. They have the whole other set of architectures that they've got to be concerned about maintaining.

So, if they can see a segment of their application stack -- in our case, the governance and control of the global workforce -- that absolutely can be an SOA project that we can jointly embrace and can get them down the road toward this new architectural style.

Gardner: For those of our listeners who might not be familiar, Workday is relatively young. You are only few years old. You're based in Pleasanton, CA. You’re focused on HCM, but why don't you fill that out a little bit. Then, let's also talk about the philosophy that you embraced when you started building out your services. Finally, we'll try to take a look under the hood. So, first, a little bit about Workday and then what you have got supporting your architectures?

Swete: I'm happy to introduce the company a little bit. Workday is about a four-year-old startup, as you said, based in Pleasanton. The company was founded by Dave Duffield, former founder of PeopleSoft, and co-founded by Aneel Bhusri. Basically, the company got together four years ago and began development of its products. We launched publicly in November of 2006 and then we had our first version of our HCM business services and two production customers.

Since then, we've grown the customer base up to north of 75 customers, with more than half of them in production. The idea behind Workday, in addition to being a SaaS company and in addition to focusing on HCM, was to focus on enterprises in a space that we define as the upper mid-market and we started with a focus on companies with between 1,000 and 5,000 employees.

In the past two years, we've had great success in selling not only to that target market, but we've been able to move up market and attract larger customers. Today, the average customer we're engaging is probably closer to 10,000 employees than 1,000. So, we'd say that we service companies in a range between 1,000 employees and 20,000-30,000 employees.

We've even attracted larger customers. Flextronics is the example of our largest customer. There are over 200,000 employees, and have selected Workday as a way to consolidate a number of HR instances that they have around the world.

Gardner: Given the fact that you needed to be modern, you needed to be flexible, but you also needed to scale, what were some of your requirements, and what did you end up with in general terms to make this possible?

Success at a high cost

Swete: Let me back up a little bit and just say the other bit of information to toss in about what was motivating us to start the company was just taking a look at the enterprise solution space and starting to identify some of the complexities of owning and implementing these applications.

It's our belief that enterprise applications have driven a lot of success and a lot of value in enterprises, but that success and value has come at a very, very high cost. Essentially the systems come down to being very hard to use, hard to change, and hard to integrate. Those were three thoughts in our heads as we started Workday. We wanted to go after the same space, which is a complex space. There are complex processing requirements and hence, the need to have a solution that’s going to scale.

So, it was taken as a given that we were going to have feature-rich applications that needed to scale to support large work forces, but we wanted to achieve that, while also attacking the issues of being hard to use, hard to change, and hard to integrate.

That's what led us to evaluate the new technologies in terms of how could we take an approach that would allow us to progress against those issues, while still being able to satisfy the enterprise-class functionality that you have to have to play in this game.

Gardner: And what did you decide on?

Swete: The approach we ended up taking, as we took it to the next level down, was that we started to investigate where some of these complexities and difficulties of use came from. As we evaluated the prevalent architecture enterprise systems, that of client-server, there were a couple of dimensions that we looked at.

From the point of view of being hard to use, the user experience for those applications typically was, first of all, designed for a highly trained back-office user and was deployed as a menu-driven application with tons of fields on each particular screen. Then, that user experience was migrated to the Web. So, these were not native Web applications. They're applications that found their way into the browser, but retained their old complexity and difficulty of use.

From the beginning at Workday, we knew we wanted another approach. Rich Internet application (RIA) experiences were emerging and supported by new technologies. We committed ourselves to being an application that was built first and foremost for the browser and one that was also built to consider the needs of not just the back-office, but the rest of the workforce. So, we started to look at technologies like Adobe Flex to give us an ability to be in the browser but to still deliver some of the rich experience that customers expect.

On the side of being hard to integrate, we took a look at the client-server architectures. The way these applications were written was to think about regulatory information that was required. Design a data model to meet that regulatory information and then build the transactions to feed the data model. That gave you a monolithic application that could generate the reporting you needed to satisfy your regulatory reporting, HCM, and financial management.

The integration of other systems to feed data into those systems or to get data out of those systems was left as an afterthought. Integration was not thought about upfront. Integration was not thought about in terms of the new Internet standards we have in the form of Web services, or the new architectural forms and approaches we have in SOA.

From the beginning, we thought about a system that would be able to deal natively with producing Web services to get data out of and back into the application and would treat the conversation with other systems as a first-class conversation, just like the conversation with individual users.

Turning to the ESB

We began an investigation of tools that could help us do that and really looked into the SOA space. We thought that what future state enterprise business applications needed was to embed some of the technologies you find in enterprise service bus (ESB) technologies. So, we've gone ahead and done that.

We do have some of the transformational and delivery options in multiple formats available to us in our data-center, so that the Workday applications can generate Web services. Beyond that, we can transform those Web services into other data formats that might be more meaningful to legacy applications or the other applications we need to tie to. We did a lot of work in that area and came up with the need to embrace Web services and embed in an ESB in our case.

Gardner: You also created what you call an "object management server." Why don't you explain what that is and why that makes sense?

Swete: That comes from the third area of these applications as being hard to change or just the general theme of this discussion today, which is this whole issue of agility. How do you make complex applications configurable, not only when you're initially implementing them but postproduction.

As you move forward, it's not like your business stops changing after you initially implement the enterprise application. Your business is constantly changing, and client-server based applications have shown a real inability to keep up with changing at the pace of anyone's business changes. They've shown a real high cost to be able to change to incorporate new functionality postproduction.

In working on that problem, we took a look at client-server architectures. It's our view that there's a lot of the rigidity in these architectures, and rigidity is what leads to a lack of agility. We think the rigidity in these architectures comes from the fact that you've got a complex logic layer.

Typically, it's multiple languages, but you have an executable that's built out of a lot of lines of program code. Millions of lines of code, in most cases, are backing the logic layer of enterprise systems. That layer has a complex conversation with the relational database, which also has its own complex structure -- typically thousands of relational tables to model all of the data.

Bringing change to that environment is difficult, because changes to the logic layer have to be synched up with corresponding changes to the data layer, and both layers are constantly changing. We decided to take an entirely new approach in this area and embrace an approach that leveraged the concept of encapsulating data with some of the logic into an object.

At Workday, the primary logic server is what we call our Object Management Server. It's a transaction processing system, but it's entirely based on an object graph, and that is just a class structure that represents not only the application and its data, but also the methods that process on that data.

The important difference is that we have that layer and we don't have a correspondingly complex and changing data layer. We have a persistent data store that is a simplified version of a relational database that can persist changes that happen from the object layer. But, as we're developing our applications and changing our applications, we don't need to constantly change the shape and form of the persistence layer. It's an unchanging relational schema that can persist, even as we make changes up in the object layer.

This frees us up to, one, develop our applications more rapidly, but two, to change those applications, even for our customers in production, without having to take the system down and make structural changes to the underlying database.

So, our two new approaches really reduce the coding you do up in the object layer. We try to define the application more as metadata and reduce the complexity of the relational model that you have.

Leveraging modern architectures

Gardner: So, we’ve recognized some of the handicaps of some of the older approaches, recognized the new set of requirements for the modern day, understood that SOA principles can be applied here quite advantageously, and recognized that rich Internet application interfaces are the way to go when the browser is the ultimate client target. Without getting into too much more detail, we've certainly established some improvement in modernization around the architecture. Let's get into a little bit of what that does for you.

We talked in general terms about agility, but there are some interesting add-ons here -- things that you couldn’t have gotten otherwise in terms of benefits. We're not just re-paving cow paths in terms of delivering applications and services. We're actually now able to do interesting things around BI, around scale and customization, and around different services federated to different users, but with more commonality under the hood. That brings more total cost reduction for the end user.

Let's get into what these modern architectures do not only for you in terms of cutting your costs, but advantageously creating new business benefits for your customers.

Swete: When you combine the architecture we talked about with the SaaS delivery model, you get some of these benefit categories. We get benefits in all the categories that you just mentioned, and you're absolutely right. Our view is that it's equal opportunity. There are definitely benefits for the customers that we're serving and, frankly, we think that in the approach there are tons of benefits for us, as a vendor, to take cost out of what we're doing and pass those savings on to our customers.

You named a lot of categories, so let me let me start with one area, which is the benefits we get out of doing more about integration. With an architecture that really facilitates integration and especially, if you combine that architecture with a cloud-based approach or delivery of SaaS, you get what we at Workday call "hosted integration" or "integration on demand."

We use our embedded ESB to do exactly what you just said. We take the ESB and package up integration so that it can be reused across a wide set of customers. The best example of packaged integration within what we call the "Workday Integration Network" is our benefits carrier network.

Here's a solution where Workday has used Web services to tie our HCM solution to a variety of benefits carriers and we offer customers the ability to sign up for this network. They would pay us for the use of the network just as they would pay us for any of other business service that we deliver. What we're able to do is offload the need for them to convey their benefits data out to the carrier and to get information back from the carrier into their human resource system.

This has been a very popular option for most of our customers, because most of our customers are large enough to have multiple carriers. All of the carriers inconveniently have multiple data formats, and the formats are always changing, and the mapping and testing of data access of those formats is always a cost. So, we're able to lift that off of them and just give them a service, which is a tie of all of the carriers they select into the Workday benefit system.

Gardner: Stan, one of the interesting things about cloud and this whole notion of centralizing allows for different things to be done with data. Now, the data is often in little nooks and crannies, in different formats and inside of different architectures. But, as we centralize the architecture, we're also getting more access to different types of data. In doing so, we can do joins, overlays, and comparisons in ways that hadn't been done before at a scale that hadn't been possible, at least at an acceptable price.

Let's get into this notion of BI. What can your architecture bring to the table, and what can your clients start to do in terms of gaining insights into what's going on inside their companies, but just as importantly, in conjunction with their business processes and extended business processes.

Built-in business intelligence

Swete: That's a big question. Built-in BI, as we call it, is absolutely an advantage of our offering. That is what we're offering today and the future that's possible. I can go through maybe a couple of levels on this, because as the customers that we've attracted look at the Workday solutions, they see unusually rich access to data as the first basic offering.

Having an object model that allows us to link more data attributes together than a classical relational database to establish relationship is a lot lighter weight than having to build the foreign key into another table. We're able to cross-link a lot of information that we're tracking inside the object model that we have, and so we're able to offer unusually rich reporting to the customers.

One example is that, just like many HR systems, we can give a straightforward headcount report, but with the Workday system, the headcount report isn't just a flat report in the system. We offer you the ability to give the headcount by organization to link to other information about each organizational entity, get its details if you want to, or see other reports related to that organization, probably more of the point for BI for the actual summary headcount that's in that particular organization, without the need of a third-party tool.

Workday is going to offer you the ability to drill down on that aggregate number, and take a look at the number in terms of all the dimensions that go into it. An HR professional could look at the headcount for an organization and analyze it in terms of gender by business site, for example, or job code by business site. Our transactional application is facilitating multi-dimensional analysis without the need to have to take the data, off load it into an OLAP cube, and then, by a third-party tool, query that cube.

The barrier that we have broken is -- again, going back to client-server systems -- you had these data models that were defined to provide one kind of reporting. Typically, it was regulatory financial reporting or the regulatory HR reports that you have to provide.

Workday provides all of those, but it crosses over into information that could be more interesting to the people who are not just back-office HR professionals, but maybe managers who wanted to get information about their workforce. That is all built into the application, and that's the level of increased BI we're delivering today.

As we look forward into other things we could do, you mentioned reaching across different elements of information. We do a bit of this today in terms of the automated business processing that we offer. We have built-in workflow in Workday and we have a system that is able to track all of the performance of the automated business processes we deliver.

Our BI today can reach across and look into their performance information and give the manager information, such as the average time it took you to do a hire, which department had the longest time to do it, and which department had the shortest time to do it. We'll continue to improve this kind of analytical information over your business performance. We see this as a very valuable area going forward and we'll get richer analytics supported in that space. That's an area that we're verging into.

The third opportunity, which you’ve also alluded to, is the benefits carrier network. There's an ability to drive intelligence across the population of users who are engaging in this network. For example, statistical information about the most widely used carriers might be interesting for any company in the network, whether they have those carriers or not. Usage information about the relative number of people using a carrier inside the company would obviously be interesting.

There is just a large world of opportunity to expand into, but the important thing is that we think as a transaction processing systems, which is what these systems are classically looked at as, we're growing out of that to also provide BI without the need to buy third-party tools to do it.

Social networking

Gardner: We're talking about people here -- workers, productivity, habits. People don't just live in the workplace, they have lives as well, and we have this phenomenon now that's going on around social networking and the ability for people to connect in new and different ways. It seems to me that this offers the potential for yet another large data source to perhaps be compared, contrasted, brought-in, and in some ways leveraged, vis-à-vis what's going on in the HCM apparatus inside the organization.

Swete: You're absolutely right. That actually even widens beyond social networking to the proliferation of productivity tools that get called broadly Web 2.0. For me, what that means is that to be an enterprise application that's relevant in a world where those applications are gaining increased usage, you not only have to have the great system-to-system integration that I talked about before, but you have to be an application that, with good security, is mashup-able, if I can use that non-word.

That's the second dimension of integration that you really can deliver on, if you’re an application that’s built for the Internet, just like the Web 2.0 applications, and certainly Workday is that. The application appears to our users as a Website, and the data in the application is accessible via the Web services I talked before, but you don't always have to construct or communicate in a heavy-duty Web service.

Workday has the ability to have RSS feeds of our data, the ability to instantly tie to emailing systems, the ability to link out to make calls to a phone numbers in Workday. The most popular mashup is always mapping locations of, let’s say, business sites. Workday is wide-open to that type of integration, and we expect that to explode, as the use of our applications reaches out into the workforce.

Gardner: I certainly expect that the ability to integrate to the social tier is becoming all the more important, and can be extremely valuable. I don't think people have plumbed the depths of what productivity benefits are inherent in that.

Swete: People are thinking about that a lot, and there is huge opportunity for a certain class of social network apps to bond with enterprise apps and deliver real value. A great example of that is LinkedIn as an application that really can facilitate a new way to do recruiting, for example. They certainly see that as an opportunity, and it's incumbent upon modern enterprise applications to be able to tie into services such as that, so you can get that kind of benefit.

Gardner: Another important tier to integrate to, or to reach, is the mobile tier. You guys have apparently put some thought into that. Tell us what a SaaS provider like yourself can bring to the table for an enterprise that would love to be able to get more data, more applications, and more business processes extended out to mobile devices across these mobile network.

Widening access

Swete: That's extremely important for us. I think I said this earlier. With our applications, in order to overcome the application being hard to use, you want to work on your native user experience, but you also want to consider that the wider user population is just not always going to use your native user interface (UI). You're going to have people who want to use your application without getting into the pages that your application actually renders. Mobile is a great example of that. We absolutely see widening out access to Workday on the mobile devices.

We're just taking our first step in that direction right now, and this is improving the benefits of the modern architecture. We've been very quickly able to extend the business-process framework that we have. This is the framework that delivers the automated workflow that I spoke about before. We've been able to extend that out so that approvals that are done within that framework can now be completely processed on a mobile device.

We’ve picked the iPhone as the first starting point and we'll be expanding out to other devices. The benefit here is that you get access to the Workday solution without having to be at the mercy of a browser on a tiny device running the rich UI that Workday generates.

We're able to mark-up a subset of our data and have that appear in a native client on the iPhone that you can get on the App Store, just like you get any other iPhone application. Then, with security, you're just utilizing a native app, which is acting on Workday data. We use that for manager approvals, the management of to-do lists, and for enterprise search of the workforce. That's been a successful example of leveraging this modern architecture. We didn't have to go in and rewrite our applications.

Our applications developers merely extended the existing application to say, "for small profile devices here's the data that's relevant to show" and they stopped there. We're able to use the separate UI layer and extend that UI layer to generate a completely different view of what the approval would look like for the mobile device. Then, process that just as if it were a transaction coming in from our own native user experience.

Gardner: I can see where a line-of-business manager could really benefit from this. They've got some ideas about what they want to do with their business. They, of course, have to bring the employees on line. There are going to be approvals, there's going to be a necessity for dealing with the HR department on that.

If they could find a way of entering into that process or workflow around approvals through the mobile device, through these interfaces, do it quickly, and get it automated, I think that might be a very interesting opportunity.

Critical opportunity

Swete: More than interesting. It's critical. If you look at a lot of the managers I know, it's just binary. It's the difference between the using the system and not using the system. Even though we have a very accessible and very user-friendly user experience, we're talking about busy people who use what they use.

They use their portals of information to get access to information on the Web and they use their intelligent cellphones now to get access to information and even to update information.

If we can provide this functionality on the devices that they use, then they will use this. If we can't, then they will get someone else to enter their information into the system, and that's not the way we want it. So, absolutely, it's tying in another theme of Workday as wider access to our applications. We think it's absolutely essential to increase the value that you get from your enterprise applications.

Gardner: One of my favorite sayings is that convenience is the killer application and I think that's what you bring in here, right?

Swete: Absolutely. Mobile is a good place to start with that, but we won't stop there. There is a lot of information that is currently presented well within Workday, but it could be presented just as well within a gadget and someone else's portal. We'll be looking for that opportunity as well. It's a way to put information in front of the people who need to see it without having to draw them into your user experience.

Gardner: Let's move into some examples, perhaps some anecdotes, about how this is being used in the real world and then we'll also look into the crystal ball and see what you might have in mind for the future. Do you have any case studies or anecdotes of how any of your customers have actually put into use, or are getting returns and paybacks from, some of the benefits that we have talked about that have their underpinnings in the architecture?

Swete: I have a couple of examples I think might help. I can go from general to specific. The one general example that I always like to quote is a real payoff and testament to the new architectural approach and combination of the new architectural approach and the use of SaaS in terms of delivering the business services. All Workday customers are always on the current version, and this is very, very different for the world of business enterprise application.

Workday delivers new updates three times a year, and our growing customer population whether they are in implementation or in production, comes up on each of those updates within the six weeks following the initial release of the update.

We did four updates last year. We will do three this year. Just this past month, we have been bringing our user population up to Update 7. If we have had this call two months ago, the only update that anyone would have known about in the customer population was Update 6. Sitting here today, Update 6 is a complete memory, and the only update that anyone knows about, whether they are just implementing on Workday or whether they've been in production for three years, is Update 7. That just brings a whole new opportunity to the customer conversation.

Special offer: Download a new white paper on Workday's latest update to System 7.

When we're talking to customers, we're all looking at the same code line and the same set of functionalities. We don't have to think about what version, what tech stack are you on, what version of your database are you on, or what version of Workday are you on? When they are asking a question about a new feature they want, we're all looking at the same feature set. It really helps to facilitate the conversation about what new features might be appropriate for the coming update, to say nothing of what it does for the cost profile of supporting these guys.

We have our stack, and that's it. They all run on it and we're able to keep them current on it. That's the benefit from the vendor side. From the customer side, the benefit they get is new functionality delivered to them, and all the manual work and data conversion work that is still necessary is done by Workday and not by them. They can focus on how they want to implement this new functionality on what timetable.

That's data point number 1 for some of the payback to the new architecture. A lot of us who came from the enterprise space are really impressed and pleasantly surprised at how rapidly we can move production customers and implementing customers forward.

The psychological effect

Gardner: I suppose there's a psychological effect there as well on the receiving end of these services. All of a sudden, you're getting more and better, but you didn't have to pay more and you didn't have to go through the pain of implementation and debugging. In adopting new things in the past there was an actual penalty for adoption, whereas the SaaS model gives you all rewards. It's like "hit me again," right?

Swete: It's absolutely that. As a customer, you avoid the high cost of having to set up QA environments and duplicate environments on-premise, having to deal with data conversions, or having to deal with installation instructions for the software that you've got on your premise. That's all handled by the vendor.

The other thing that's not widely talked about that's also valuable, though, is this notion of chunking up how new features come. We're on a regular clock here. So, three times a year, you're going to get a set of new functionality from your vendor and you are going to get converted to it.

In the on-premise world, where the delivery is deferred, what's building up is just a larger set of functionality that's going to have to get consumed some time at very high cost. Part of the way we look at this is that the incremental approach is just much more cost effective for our customers and for us. We know that, because even in our brief history here, we've had an on-going conversation with our customers about how long we want the update periods to be.

Our first update period was seven months and we delivered a ton of functionality. It was harder for them to consume and harder for us to support the upgrade to it. Now, we’re on a really good balance, where we can have meaningful functionality, but come out in a way that can be readily consumed. I don't think -- well, I don't think, I know the on-premise world just does not work that way.

Gardner: Okay, a quick look to the future before we wrap up, Stan. What are some of the implications from what we've been discussing, perhaps in terms of BI, perhaps in terms of extended business processes, or more of this integration agility?

Swete: Let's focus on the last two. We talked about a lot of the value of leveraging new technology to deliver enterprise applications in a new way and then combining that with doing it from the cloud. That combination is going to profoundly change things going forward.

Today, it's a new option for enterprises to look at in terms of offloading some of the applications that they're trying to support in their existing environment. It's a vehicle for consolidating some of the complexity that you have into a single instance that can be managed globally if you have architected globally, as Workday has done.

As I see that playing out going forward, you'll see more vendors taking this approach and you'll see those vendors partnering. If you think about the combination of modern architectures and cloud-based modern architectures, what will happen when two vendors that have taken that similar approach start to partner in terms of integrated business processing is that the bar will get raised significantly for how tight that integration can become, how well supported it can be, and how it can functionally grow itself forward, without causing high cost and complexity to the consuming enterprise that's using both sides.

As I look in the future, I think enterprises will see an ecosystem of their major application providers be cloud-based and be more cohesive than a like group of on-premise vendors. Instead of having a collection of different architectures and different vendors all in their data center, what they will see is an integrated service from the set of providers that are integrating with Web services in the cloud.

Gardner: So, when the cloud model becomes the common denominator, it allows for a lot more, I don't know, co-existence collaboration but I suppose really just integrated processes.

Swete: It allows for a lot more integrated processes. The key thing with enterprise functionality is you're never done. The requirements are always changing, because business is always changing. What it allows for us is not only cloud-based integration, but the ability to change that integration without placing additional cost on your customer.

That's what's the key is, that you will be able to deliver enhancements to the integration between vendors without getting caught up in how that integration might have been deployed at customer A, versus customer B, versus C. That will allow the same kind of agility we've been talking about. That will allow integrated solutions, the cross-vendor integrated solutions, to keep pace with the change of business, which is absolutely not the case today.

Gardner: I want to thank the sponsor of this discussion, Workday, for underwriting its production and a special thanks to our guest, Stan Swete, CTO of Workday. I really appreciate your inputs, Stan.

Swete: Dana, thanks a lot.

Gardner: We've been learning about how IT architecture is destiny and how a SaaS provider's operations can mean more to its customers and simply lower costs and baseline delivery of services as a Web application. We have seen a multiplier effect, if you will, in terms of how new and additional productivity and agility benefits are gained from a modern architecture regardless of its location and ownership.

This is Dana Gardner, principal analyst at Interarbor Solutions. Thanks for listening and come back next time.

Listen to the podcast. Download the podcast. Find it on iTunes/iPod and Podcast.com. Learn more. Sponsor: Workday.

Special offer: Download a new white paper on Workday's latest update to System 7.

Transcript of BriefingsDirect podcast on how Workday's SaaS delivery model for human capital management applications provides better business intelligence and architectural advantages to end users. Copyright Interarbor Solutions, LLC, 2005-2009. All rights reserved.