ERP, 20 years ago: the move to go global

ZDNet's 20th anniversary: ERP in 1991 was a markedly different world than it is today.
Written by Andrew Nusca, Contributor on

ERP in 1991 was a markedly different world than it is today. The sector was undergoing a change that would usher in a new group of software market leaders and send another group to the scrap heap. The shift at that time was to move from green-screen to Unix/Client Server solutions and some newcomers, like PeopleSoft (now owned by Oracle), were introducing a new user interface with their applications: Windows.

In 1991, I ran Andersen Consulting’s (now Accenture) global Software Intelligence unit. We advised companies on what to buy and what to run away from when it came to big ticket software purchases. In 1991, we were discussing ERP options from firms like: Walker Interactive (which changed to Elevon which was acquired by Infor), The Dodge Group, Tesseract, Software 2000 (which changed to Infinium which was acquired by Infor), Global Software, Quality Software Products, Computron, Ross Systems, IMRS (which became Hyperion and now owned by Oracle), Lawson (currently being acquired by Infor) and JD Edwards (now part of Oracle). Other ERP vendors like Cullinet (later acquired by Computer Associates), McCormack & Dodge (M&D) and Management Science America (MSA) (both later merged into Dun & Bradstreet Software and later sold to Geac) were on most prospect short lists at that time. But, those shortlists were changing in 1991.

SAP was selling R/2 at the time. SAP was only then beginning to make major inroads into the North American markets having already established itself in Western Europe. The product had undergone some additional function enhancements to make it more acceptable to U.S. and North American buyers. Around 1991, SAP really started to pick up steam in North America.

Prospective software buyers were seeing a lot of changes emerging in the ERP space. American Software, ASK, Mapics, JD Edwards, MSA and Marcam had manufacturing solutions. Even McCormack & Dodge had one briefly. But, SAP was the one that gaining the most momentum. Walker, Data Design Associates and McCormack & Dodge all had financial applications that could support large, flexible user-definable accounting code blocks. This functionality had given them an advantage in regulated industries (e.g., utilities and insurance) and helped their sales move up nicely in the late 1980s and early 1990s.

But, large code blocks alone weren’t enough to help older green-screen products survive. The most disruptive influence on ERP in 1991 had to be PeopleSoft. PeopleSoft was selling a Human Resources application that looked (thanks to the Windows user interface) unlike anything else on the market. I used to say to software buyers that PeopleSoft’s competitors looked like “motherhood, apple pie and the American way” while PeopleSoft was “sex, drugs and rock and roll” in comparison. The new user interface was coupled with a client-server architecture.

Client-server architectures from PeopleSoft and many other vendors were game changers in ERP. Vendors that predominately sold platform-specific solutions (e.g., to the IBM AS/400, IBM iSeries, HP3000, DEC Vax and, of course, IBM mainframes) were in trouble. Customers were looking to buy commodity Unix-based servers and non-proprietary databases. Suddenly, no one wanted to buy dumb terminals. No one wanted ‘monolithic’ or proprietary CPUs. Open was in. Closed was out.

The definition of ERP was really changing then, too. ERP, a term attributed to a Gartner Group analyst in 1990, was basically MRP (materials resource planning) software with integrated financial and human resources applications connected to it. ERP then did not address CRM (customer relationship management), SFA (sales force automation, the Internet (amazing as that sounds), telephony or mobile devices. No, ERP was an internal application for the most part. The best solutions might have supported EDI (electronic data interchange) as a way to enhance communication between customers and suppliers. Very little of ERP in 1991 could occur outside of one’s firewall.

ERP customers in 1991 were also deciding how many ERP component providers they wanted. Previously, software buyers purchased many component applications from niche, best of breed providers and integrated them together. The convenience of buying pre-integrated suites was definitely driving up the market success for large suite providers like SAP and JD Edwards. Other vendors, through internal development or acquisitions, re-doubled their efforts to expand their product lines to remain competitive. Sadly, some did not.

Global and ERP were two terms that rarely existed in the same sentence prior to 1991. However, that was starting to change then as some ERP providers had crossed company size and geography barriers that were permitting them to open offices in multiple countries and to develop language specific versions of their products. Global ERP was tough to deliver in the green screen days as it often meant country specific versions of the software. All of the screen display logic would be changed to support a new language. That meant changing a lot of COBOL, Fortran or other code. Additionally, most packages weren’t originally designed to support different currencies, let alone languages. In 1991, the global ERP winners were the vendors who could externalize screen, currency and regulatory changes by user and legal entity. Almost no ERP vendor in 1991 could consolidate financial data easily across different legal entities that had different accounting calendars, different currencies and different languages. European vendors like CODA, QSP and SAP had the market advantage then as these companies had more multi-country functionality built into their products from the onset. United States based ERP companies usually were way behind in the global functionality game. The global economy was still a new concept to ERP vendors in 1991.

Oracle was undergoing some interesting times in 1991. The company was still new to application software and only had a line of financial software applications. Unfortunately, the company hit some choppy operational and earnings issues around that time. Today, Oracle is clearly one the largest software companies in the world. 1991 was a growing pains time for Oracle but it clearly grew up a lot since then.

Since 1991, the leader board for ERP continues to change. Following the massive sales that occurred in the space prior to Y2K, ERP sales went into the doldrums for a couple of years. 2 and 3 level client server architecture debates ended when true web-based applications arrived early last decade. Sales began to fall off again when the recession of 2008 occurred. But, at the same time, SaaS (software as a service) providers have emerged in the last decade. Newcomers like Workday (established in 2005 with many former PeopleSoft innovators), NetSuite (nee NetLedger and materially owned by Larry Ellison of Oracle), Intacct, Salesforce.com, FinancialForce (built off the Salesforce.com platform) and many others are re-defining the space once again. Their view of ERP is decidedly focused on:

· Cheap to maintain multi-tenant offerings · Support for more than transactions (e.g., Workday’s support for REA) · Platform as a Service capabilities for customers and implementers · Faster implementations · Easier integration with other cloud and on-premise applications

Even the market leaders, like SAP and Oracle, have gotten the SaaS religion. SAP has its Business ByDesign product line and Oracle is shipping new Fusion modules.

What will the future hold? Change! Yes, the leader board will change some more. We can expect vendors who don’t make the big shifts will get acquired or minimized out of market relevance. New upstarts will siphon off market share from the majors and, thanks to the power of competition and innovation, customers should benefit. But, that’s just one guess. We’ll see how right this prediction is in another 20 years when someone else writes this story in 2031.

» Return to ZDNet's 20th Anniversary Special

Editorial standards


Programming languages: It's time to stop using C and C++ for new projects, says Microsoft Azure CTO

Programming languages: It's time to stop using C and C++ for new projects, says Microsoft Azure CTO

Slow internet at home? This adapter is the key to faster wired connectivity

Slow internet at home? This adapter is the key to faster wired connectivity

Meta's AI guru LeCun: Most of today's AI approaches will never lead to true intelligence

Meta's AI guru LeCun: Most of today's AI approaches will never lead to true intelligence