Introduction: Full Warehouses. Empty Shelves. One Very Bad ERP Decision.
Halloween, 1999.
Walmart. Kmart. Target. Shelves that should have been packed with Hershey's Kisses, Jolly Ranchers, and Reese's cups — empty.
Not because Hershey's ran out of chocolate. The factories were running. The warehouses were full. The candy existed. But because of one catastrophically mismanaged ERP implementation, Hershey's couldn't process orders, pack shipments, or get products to stores in time.
The result? A company that had been making chocolate for nearly a century failed to deliver on the single most important sales day of its year — and it made front-page news in The Wall Street Journal.
This is the story of one of the most famous ERP disasters in business history. And it has lessons every student, business owner, and consultant needs to understand.
⚠️ This case study is studied in MBA programs and ERP certifications worldwide. At ERP Beans, we train students to be the consultants who prevent these disasters — not the ones managing the fallout.
The Setup: A $112 Million Modernisation Dream
In late 1996, Hershey's made a necessary decision: replace its fragmented, outdated IT systems with a fully integrated ERP environment. The investment? $112 million across three enterprise platforms deployed simultaneously:
| System | Vendor | Purpose |
|---|---|---|
| SAP R/3 | SAP | ERP — financials, order management, logistics |
| Manugistics | Manugistics Group | Supply Chain Management |
| Siebel | Siebel Systems | Customer Relationship Management |
This ambitious project was called "Enterprise 21."
The vendor-recommended implementation timeline: 48 months.
Hershey's executives said: "We'll do it in 30."
That single decision set everything else in motion.
The Fatal Mistakes
1. Y2K Panic Drove a Rushed Timeline
The late 1990s were gripped by Y2K hysteria — the fear that legacy systems would crash when the calendar rolled to 2000. To avoid this, Hershey's compressed a 4-year project into 2.5 years, pushing for a July 1999 go-live date.
July is not a random month for a confectionery company. It is when Hershey's processes the bulk of its Halloween and Christmas orders from major retailers — the single most operationally critical window of the year. There was zero room for error.
2. Big Bang Instead of Phased Rollout
Most ERP experts recommend going live one module at a time — stabilise, then expand. Hershey's chose the opposite: launching all three systems, company-wide, on a single date.
The result was immediate chaos. Departments were in conflict. Data migration was flawed. Training was inadequate. And most critically — testing was cut short to meet the deadline.
Integration points between SAP, Manugistics, and Siebel had never been properly stress-tested. Nobody knew what would happen when real orders started flowing in at peak volume.
They found out in July.
3. Going Live at the Worst Possible Moment
When the systems went live, critical issues surfaced immediately. Orders weren't flowing correctly. The ERP couldn't match inventory to outgoing shipments. Shipping instructions weren't generating.
"Following their go-live in July 1999, Hershey was still struggling in mid-September to fix errors in its order processing and shipping functions." — Decision Resources, October 2025
Halloween orders — already placed by retailers weeks earlier — simply couldn't be fulfilled.
The Damage: By the Numbers
In September 1999, Hershey's CEO Kenneth L. Wolfe disclosed the crisis to Wall Street analysts, confirming the company could not deliver $100 million worth of Kisses and Jolly Ranchers for Halloween.
| Impact | What Happened |
|---|---|
| Lost Orders | $100 million in undelivered Halloween candy |
| Revenue Drop | Q3 1999 sales down 12.4% year-over-year |
| Profit Collapse | Quarterly profits dropped 19% |
| Stock Price | Fell more than 8% in a single day |
| Investor Trust | Analysts didn't trust Hershey's delivery promises for nearly 9 months |
📊 Sources: CIO Magazine | Panorama Consulting | Pemeco Consulting
The most painful detail? The chocolate was sitting in the warehouse the entire time. Hershey's didn't run out of product. Their $112 million ERP system simply couldn't connect inventory to orders to shipments.
A system meant to make the business faster made it unable to function at all.
Root Cause: What Actually Went Wrong?
This was not a failure of software. SAP R/3, Manugistics, and Siebel are all enterprise-grade platforms used by hundreds of large companies successfully. The failure was entirely in how the implementation was managed.
"Hershey's downfall wasn't about choosing the wrong software. Failure was rooted in shortcuts related to systems testing, data migration, and training." — Jonathan Gross, Geneco Consulting, as cited by Accent Software
| What Went Wrong | What Should Have Happened |
|---|---|
| Compressed 48-month plan to 30 months | Stick to the vendor-recommended timeline |
| Big Bang deployment of 3 systems at once | Phased rollout — one system at a time |
| Testing phases were cut short | Complete UAT and integration testing before go-live |
| Go-live during peak season | Choose the slowest business month to go live |
| Insufficient staff training | Dedicated training program weeks before launch |
📊 Source: The Case Centre — ERP Implementation Failure at Hershey Foods
Why This Matters for Every Business
Hershey's was not a small business that didn't know better. This was a Fortune 500 company with billions in revenue, experienced leadership, and $112 million to spend. If this could happen to them, it can happen to any organisation that treats ERP as just another IT project.
ERP is not software. It is your business infrastructure.
Every order, invoice, delivery, payroll, and financial report — once you go live on ERP, everything runs through it. If the system breaks, the business breaks. There is no fallback.
This is what makes ERP one of the most high-stakes decisions a business ever makes — and why the people who understand ERP correctly are so valuable.
What Modern ERP Gets Right (That 1999 SAP Didn't)
The ERP landscape has changed significantly since 1999. Modern platforms like Odoo are built with these exact lessons in mind:
| 1999-Era ERP | Modern Odoo (2025) |
|---|---|
| Monolithic — all or nothing | Modular — activate only what you need |
| Separate systems, brittle integration | All modules natively integrated |
| Big Bang by design | Phased rollout built into the platform |
| Requires large implementation teams | Accessible to trained functional consultants |
| Expensive and enterprise-only | Designed for SMEs — India's 63M+ MSMEs |
A business today can go live with just Odoo Accounting, stabilise, then add Inventory, then Sales — the exact phased approach Hershey's skipped.
Conclusion: The Chocolate Was There. The System Failed.
Hershey's 1999 disaster is taught in MBA programs, ERP certifications, and business schools worldwide — not because it's dramatic, but because it reveals a fundamental truth:
A brilliant ERP system, implemented badly, is worse than no ERP at all.
The consultants who understand this — who know what correct ERP implementation looks like from the inside — are the people every company needs before, during, and after a system rollout. They are the ones who ask the right questions, flag the wrong timelines, and insist on proper testing even when executives are pushing for speed.
That is the professional ERP Beans trains you to be.
Learn ERP the Right Way with ERP Beans
ERP Beans is India's dedicated Odoo ERP career program for students — built to make you genuinely job-ready in 6 months.
✅ Hands-on training in a live Odoo environment
✅ Learn O2C, P2P, R2R cycles — not just theory
✅ Functional track (Commerce/MBA) + Technical track (IT/Engineering)
✅ Official Odoo certification preparation
✅ Resume, LinkedIn & interview coaching
✅ Placement support with Odoo partner companies across India
🚀 Don't just learn ERP. Learn it right. Join ERP Beans and Train on Live Odoo →
FAQs
No. SAP R/3 is a proven, enterprise-grade platform. The failure was entirely in how Hershey's managed the implementation — rushing the timeline, skipping testing, and choosing the worst possible go-live window.
Yes, but it took years. The company stabilised its systems and completed a successful SAP upgrade in the early 2000s. However, the financial damage and lost investor trust were real and lasting.
Going live with all ERP modules across the entire company on a single date — the highest-risk implementation method. The safer alternative is a phased rollout, one module at a time.
Because the same mistakes still happen today. Students who understand why ERP fails become the consultants who prevent failures — and those are the professionals companies hire first.