Just under two years ago, NetSuite announced a set of partnerships with Microsoft, including plans to move some of its computing workloads from Amazon Web Services to the Microsoft Azure public cloud.
This was interesting in that it seemed to indicate that a large software company was agreeing to use data center infrastructure from Microsoft (msft), which is a distant second to AWS in public cloud—basically a huge set of computing, storage, and networking resources that one company owns, manages, and rents to multiple customers.
The thing is: This promised move never happened. This week, a spokeswoman for NetSuite—now part of Oracle (orcl)—said the "migration as stated in the release didn't take place as it wasn't a big priority and we focused on more pressing customer needs at the time."
She also noted that no NetSuite production workloads run in AWS.
Given that NetSuite was acquired last year by Oracle (orcl) for $9.3 billion and that Oracle wants to compete with Microsoft Azure and AWS in public cloud, it's not surprising that Oracle would prefer to keep NetSuite's workloads in-house. Amazon (amzn) and Microsoft are doing their best to get big software providers to use AWS and Azure respectively instead of their own data centers. Oracle, presumably, will want to do the same.
It's not all that unusual for a big multi-vendor partnership to die on the vine. The tech landscape is littered with the remains of alliances, often announced with great hoopla that never amounted to more than a stack of press releases and (maybe) a cocktail party. What would be more newsworthy is a partnership that paid off in terms of real products and services that customers want and need.
Many tech observers cite the great VMforce fizzle of 2010 as a primary example. That year, Salesforce and VMware announced plans to jointly develop "the world's first Enterprise Java cloud." Java is a popular computer language used to build business software.
Less than a year later, a VMware (vmw) exec said VMforce wasn't going to happen, prompting one Gartner analyst to quip that "the only real partnerships are acquisitions." VMforce only lives on in press release archives and some slideshows on the web.
So why does this happen? The issue seems to be companies feel pressured to come up with news at their various events. And, in this cloud era in particular, they are threatened by AWS' dominance in so many areas that they want to team up to combat the threat. But it's hard for multi-party alliances to really take off: Too many cooks in the kitchen united only by a common threat.
To compound the issue, reporters do not often follow up to see what happened with a given pact. Credit VMware here: At least it publicly announced that VMforce would never happen. Many companies just bury the dead alliance and hope no one notices.
Get Data Sheet, Fortune’s technology newsletter
Another example: In 2014, Cisco (csco) announced Intercloud. its grand plan to get telcos and other service providers around the world to run Cisco's cloud software so that they, in aggregate, could compete better with AWS and Azure in public cloud. A year later, Cisco announced 35 new partners in that effort. A year after that? Intercloud public cloud was scuttled.
Carl Brooks, analyst with 451 Research summed it up: "Boy, that was a big swing and a miss."