Microsoft Office Reaches for the Back Office

Creating user interfaces for ERP systems just got a little more complicated for developers.

Making the desktop safe for enterprise resource planning (ERP) users has been a goal of enterprise software vendors ever since client/server interfaces began replacing green screens in the early 1990s. One of the largest, single R&D items in every leading ERP vendor's budget has been user-interface design.

So it's with no small amount of irony that a big part of the solution to this eternal quest turns out to be something that 500 million or so users are already using: Microsoft Office. And the growing acceptance of Office as a primary user interface for all major ERP systems -- from Dynamics to SAP AG's MySAP to Oracle Corp.'s E-Business Suite -- is giving the impression that the sky's the limit.

But it turns out that real limits abound. Some are purely functional; others are due to the rather untimely emergence of Windows Vista and Office 2007 just as this revolution is taking hold. The result is that deploying this most obvious interface places some less-than-obvious demands on partners, developers and users as they try to understand the nuances of what Office-to-back-office connectivity means.

The first problem lies in deciding what target platform to expect on the user's desktop. Office 2007 running on Vista can create some absolutely amazing mashups of graphics, back-office and third-party data, and Office functionality. But the number of enterprises, large and small, that have adopted Office 2007 and Vista present a limited market today -- and one that still needs some time to develop.

What to do in the meantime is less obvious. Targeting the least-common denominator -- Office 2003 and Windows XP -- leaves a fair amount of functionality on the table. Not enough to be a bad choice by any definition, but still vastly less compelling.

Picking one set of target platforms over the other can become a zero-sum game. If a partner goes for the least-common denominator, it risks being usurped by a competitor with similar functionality but a far better Vista/Office 2007 experience. Similarly, if the partner takes a chance on the latest-and-greatest option, its customers could potentially be forced to justify a triple-threat upgrade -- Vista and Office 2007 may require a hardware upgrade as well -- that could add significantly to those customers' costs.

These are classic choices that have bedeviled developers since software development emerged from the primordial ooze more than 50 years ago. What's unfortunate is that this classic choice is being played out at a time of unprecedented recognition for the value of Office as a back-office interface. The result is a Hobson's choice that most would be better off not having to make.

Which brings us to the other problem facing the growing armies of developers and partners targeting this opportunity: How far does the market go in developing for Office, at the risk of producing something that is actually less useful than could be produced otherwise?

Again, there are real limits involved. All the Office-to-back-office stakeholders -- including Dynamics, the Office Business Applications (OBA) gang, and partner/competitors like SAP and its Duet Office interface -- caution there's only so much that Office can do as a front-end to the back-office.

It's an ill-defined limitation -- "you'll know it when you see it" is the best description -- but an important one: Setting limits on users' expectations for Office is key to its success as a back-office interface. That's partly because no one wants to stall demand for new, high-priced "classic" ERP seats, thereby harming license revenue. But it's also because many of those zillions of dollars spent in interface design weren't wasted at all: Many sophisticated business processes will need a much more sophisticated interface than Office can provide.

So, popular demand notwithstanding, the Office-to-back-office revolution won't eliminate traditional interfaces as much as it will expand ERP usage to a larger class of users, which is exactly the goal that everyone has in mind. Just don't think that the user experience begins and ends in Office. There's so much more to ERP functionality than just a pretty interface.

About the Author

Joshua Greenbaum ( is founder and principal of Berkeley, Calif.-based Enterprise Applications Consulting.