The Realities of Upgrading SAP and Oracle Cloud ERP

Executive Summary

  • Upgrades for cloud ERP applications are not very straightforward.
  • Not all vendors are equal in the cloud. Vendors like Workday or ERPNext, which started in the cloud and have a long history of providing updates while minimizing disruption for their customers.

*Disclaimer. This article makes statements that apply to SAP S/4HANA Cloud and Oracle ERP Cloud. It does not generalize to NetSuite (an acquired cloud ERP by Oracle) or to SAP’s acquired cloud applications. Brightwork Research & Analysis categorizes Oracle and SAP’s internally developed cloud applications separately from their acquisitions. 

Introduction: The Issues with Upgrading SAP and Oracle Cloud

This article will describe what is often an overlooked aspect of cloud ERP. Moreover, this is the reality of upgrades for cloud ERP. You will learn about the reality of upgrading SAP and Oracle ERP.

A Brief History of Cloud/SaaS Vendors

Cloud/SaaS vendors started up around 15 years ago as a way to leverage the Internet to compete with on-premises vendors that had more abundant resources, marketing budgets, sales teams and other big vendor luxuries. The first SaaS vendors were concentrated in simpler applications (CRM being the defining SaaS application, with Salesforce the defining SaaS vendor) that could allow all or nearly all of the customers to work off of a single database instance (in many cases) but most importantly a single application instance. This eventually came to be known as multi-tenancy. SaaS vendors also provided free trial access to their software and easy cancellation terms. The concept was that rather than following the standard and expensive sales process where viewing the system was controlled by pre-sales demo teams, the prospects could test out the application often for free. I became enamored by the Arena PLM application in 2008 following this exact process.

Everyone reading this article right now uses some type of cloud application, be it Gmail, a cloud invoicing program, Facebook, etc.. We often don’t think of them as cloud/SaaS, we just think of them as “websites.” However, they are in fact cloud/SaaS applications.

In the rush to embrace the cloud, application categories that have never been historically “cloud” inherited cloud expectations. Wall Street, with its cadre of technically shallow analysts, has yet to get the memo that not all enterprise applications are equally straightforward for migration to the cloud.

One of these non-historically cloud application categories is ERP. And one of the inherited cloud expectations became pushed upgrades.

Upgrading Cloud ERP Applications

Diginomic published the article “Workday Bows to Enterprise, Reins in Release Cycle” in 2013. The following quotations from the co-CEO Aneel Bhusri caught my eye.

“I just want to crank out new functionality but our customer groups said no, we need more time to do these updates,” he explained.

It’s a totemic shift for Workday, which has hitherto made a selling point of the frequency of updates compared to the once-every-three-years-or-worse cycles of its on-premise competitors. Release 23 next August will be the last of the current regime, after which Workday will fall in with the same twice-yearly release schedule and Summer and Winter naming system as cloud CRM giant

The shift is a reflection of a paradox that many cloud application vendors face as they win increased adoption among established enterprises. For all the agility that such organizations aspire to, there are limits to the management resources they can devote to supporting new capabilities as they’re introduced. Workday has had to throttle back the frequency of updates to match those constraints.

The irony is that — although Workday presumably intends to maintain the same rate of innovation under its sparser release cycle — those looking on from the faster-moving, consumer-led end of the cloud industry may see it as evidence Workday is becoming more like the traditional enterprise software vendors that it claims to supplant.

In another move designed to ease the update cycle further, Workday has added a 4-week preview phase to the release process. As well as having 2-3 weeks for technology testing of a sandbox release instance before rolling out the production instance, customers will in the future have the option of letting users access a preview instance where they can try out new features or test new business processes.”

Facing the Reality of Cloud Updates

The Diginomica article on Workday demonstrates the challenges faced when upgrading an ERPish system (I say “ish,” as Workday does not have supply chain functionality) in the cloud.

If we can all recall, the original idea of SaaS was that the upgrades would be performed in a way that is seamless for the user. Multi-tenancy combined with a single code base for all customers.

Why is this Wall Street analyst so happy? Well, he thinks that all enterprise systems can be as easily migrated to the cloud as CRM. This makes him see reduced operational expenses for the software vendors he tracks as far as the eye can see.

Does the Cloud = Money?

This is a reason that Wall Street bid up the price of cloud companies and is promoting SAP and Oracle to overstate their “cloudiness.” This tantalizing prospect is the idea of many customers shared over a far smaller investment than with on premises software. This profitability for cloud vendors promised by Wall Street has yet to appear. The most profitable vendors continue to be the major monopoly on-premises vendors. (Salesforce has weak profitability, as does Azure. The only impressive profitability of any size comes from AWS.)

While some simple software categories, notably CRM, works like that, ERP systems do not work this way. This Diginomica article back in 2013 on Workday’s upgrade process gave us a hint of these complicating factors. Moreover, all of this is quite common for cloud ERP vendors.

The following quotation is from Rushabh Mehta of ERPNext who is very upfront about their cloud upgrade process.

“We push minor upgrades seamlessly without the customer knowing, but a major release takes 2-3 months for us to push across the customers… unlike GMail or CRM, ERP has enormous implications on things like finance, payroll, etc., so any change in configuration could be disastrous, so customers are naturally wary of pushing aggressive upgrades.

There needs to be a fine line here. Some customers can get a longer window so they can schedule any potential disruption.”Rushabh Mehta

Unequal Cloud Capabilities

SAP internally developed products that don’t “do the cloud” very well. Oracle’s long-term problematic unifying development project, one of the longest in the history of enterprise software, was called Fusion, which is now Oracle ERP Cloud (in part). However, both of these applications have far less functionality than their on-premises counterparts and also come with the negative implication of SAP and Oracle’s cloud.

Unfortunately for the major monopoly enterprise software vendors, it is not that easy to be good in the cloud. If all vendors could be, they would be. Furthermore, it is one thing to bring out a good cloud CRM system, it is quite another thing to create a cloud ERP system. It can’t be something a vendor is doing because it is trendy, but this seems to be the case with both SAP and Oracle.

Back when the cloud was just starting off, SAP and Oracle did everything they could to undermine it. SAP and Oracle speak of the cloud now because Wall Street has informed them they have no alternative. SAP and Oracle, however, have zero interest in upsetting their on-premises software business, which has been amazingly profitable for both of them.

Taking a Step Back with SAP and Oracle Cloud

This brings up the question of why a company would want to get what amounts to a weak host with S/4HANA Cloud and Oracle ERP Cloud (i.e., choosing SAP or Oracle as your hosting provider) versus just putting the much larger S/4HANA “on premises” edition on AWS.

The historically on-premises vendors have tried to propose that they are the equal of historically cloud vendors. It has become more than apparent at this point that cloud capabilities are a distinct capability. AWS gets it. Google gets it. SAP and Oracle do not get it.

It’s the difference between doing something because it is trendy, versus doing something because you believe in it.

What SAP Has Been Telling Customers About S/4HANA Cloud Development/Upgrades

SAP warned the audience in several S/4HANA Cloud presentations given at SAPPHIRE 2018 that S/4HANA Cloud has so much “innovation” that customers need to accept the risk of having what SAP calls,

“20% more S/4HANA Cloud functionality every quarter.”

SAP presenters in S/4HANA Cloud told audience members to only focus on S/4HANA Cloud for the most innovative areas of their business.

This is a euphemism for “get ready for instability.”

Is S/4HANA Cloud Ready for Production?

It’s difficult not to interpret S/4HANA Cloud as a development box. All one has to do is review the number of changes/improvements along with the frequency of improvements. It sounds like a curious statement as S/4HANA Cloud + S/4HANA on premises is touted by SAP as being used by more than 1,500 customers (SAP combined the numbers of S/4HANA Cloud and S/4HANA on premises because it would be embarrassing if it reported the S/4HANA Cloud customers by themselves.). Moreover, the application has been out for several years. But, if we strip away the name for a moment and all of the SAP marketing literature to look at the underlying facts, it is not actually a difficult determination to make.

A Question for IT Directors

If one were to approach an IT director with the following question…

“How do you feel about going live and running the business with a system that will have an automated push every quarter with a large amount of untested functionality by a development team trying to create product that was released too early?”

Simply by the nature of the releases, they would designate the S/4HANA Cloud as a development environment.

All of this brings up the question, who could possibly live with S/4HANA Cloud?

SAP has a number of slick S/4HANA Cloud videos, none of which correspond to the actual S/4HANA Cloud product. If you want to find out what is not true about S/4HANA Cloud, be sure to listen to Sven Deneken. There is no lie about S/4HANA Cloud that Sven Deneken is unprepared to tell. I can’t be sure, but Sven may moonlight as a standup comedian. 

I would not recommend S/4HANA Cloud, in its current state out until 2019, for production environments.


Because there is no way such a system can be relied upon, and the overhead of dealing with pushed updates is too high. The costs imposed by the system, even if the small footprint of functionality is ignored momentarily, would put a company’s business at risk.

Could there be some reason almost every company on this slide of S/4HANA Cloud implementations is an SAP consulting partner? How many of these companies would claim to have gone live with S/4HANA Cloud for marketing purposes, while using something else to run their business?

Do I have to say it? 

Notice Anything Odd About the S/4HANA Cloud Customers?

This is one reason why so many of S/4HANA Cloud customers are small and why there are so few of them. Nearly every S/4HANA Cloud customer can either be traced to an SAP consulting partner or an insubstantial entity that implemented the application to raise its marketing profile in some way. (The overall exaggerated nature of S/4HANA implementations (on premises + cloud) was covered in our research, A Study into S/4HANA Implementations.)

What SAP is calling innovation with S/4HANA Cloud is really just a euphemism for development. This development is so extensive because the application was announced far too early (to impress Wall Street) and it’s still going through its primary development phase. 

Buying a Ticket on the S/4HANA Cloud Rollercoaster

SAP is projecting a high degree of disruption for S/4HANA Cloud going out at least until 2019. However, the distinction between S/4HANA Cloud and Workday and SuccessFactors is that S/4HANA Cloud has few customers using the system for production. For this reason, you don’t hear very much about customers running into problems with S/4HANA Cloud upgrades. Moreover, this brings up an important question.

To see why, let us start with several of SAP’s statements:

  • SAP proposes a lot of changes to S/4HANA Cloud on a quarterly basis.
  • SAP proposes that there are over hundreds of S/4HANA Cloud customers.

If you search, you cannot find complaints about upgrade issues with S/4HANA Cloud (on SAP message boards, articles (even if sanitized), etc.). Given the instability of S/4HANA Cloud, we should see complaints. That lack of these complaints gives us a clue about the S/4HANA Cloud uptake/usage, even if I had not performed the detailed research into the subject to already know.

Oracle’s Cloud Updates

Oracle’s cloud ERP is very similar to SAP’s cloud ERP in many respects. Like SAP, Oracle ERP Cloud is a small percentage of the overall ERP business for Oracle. Still, Oracle ERP Cloud has been around much longer than S/4HANA Cloud and it has quite a few more live customers. Unlike S/4HANA Cloud, Oracle does not push Oracle ERP Cloud customers to the latest update.

Let’s review Oracle’s Cloud ERP upgrade policies.

  • Oracle ERP Cloud has upgrades every other quarter.
  • Oracle ERP Cloud also has monthly patching.
  • Oracle ERP Cloud  is tested and cutover like any on-premises application.

Experienced Oracle ERP Cloud  consultants know that changes to the application to meet new business requirements are different than on-premises, with which the upgrades are generally less frequent, and therefore the system is more stable and controllable. Oracle keeps funneling money into Oracle ERP Cloud as noted by the following quotation from Rimini Steet.

“Currently Oracle is pouring vast funds into Cloud ERP with little innovation being built for EBS. It’s unlikely that there will be a new feature in a release in the next 5 years beyond 12.2 that will be critical for your business. If EBS 12.2 has some important new functionality you want, upgrade on your own timeline and when there is a real business need.”

However, this is not translating into that much benefit to the application, or to its ease of upgrading. From a marketing perspective, Oracle has been fighting this reality.

Asserting the Opposite of What is True with Against Type Marketing

In June 2018 Oracle introduced “Soar,” which is ostensibly a way of automating the upgrade process of Oracle applications, including Oracle ERP Cloud. Oracle and Ellison began with making very similar claims around their database and have now moved to make the same claims around their SaaS applications.

As was covered in the article, How Real is Oracle’s Automated Database, I determined Oracle’s automated database claims to be not only inaccurate but impossible. This is due to the number of Oracle databases that customers want on previous versions of the Oracle database and that have dropped Oracle support.

It is also curious how automation impacted both Oracle’s applications and databases in such close timing to one another, especially considering the fact that the technology for automating database and application upgrades (only one aspect of automation) is not the same. This is referred to as “against-type” marketing. It is where you take a known weakness and assert the opposite.

SAP did this with their “Run Simple” marketing program. Before the Run Simple campaign, SAP was for decades the most complex software to implement with the highest TCO. Therefore, SAP’s Run Simple program simply asserted the opposite (as did every SAP consulting partner, aka “Team Parrot”). Oldsmobile did the same thing with their “Not Your Father’s Oldsmobile” to try to dispel the fact that Oldsmobiles were boring.

Against type marketing programs do not have much of a history of success. They are most often attempts by marketing to change perception, without the changes being incorporated into the actual product. The campaigns usually boil down to sales reps being tired of hearing something about the product that is true, and marketing deciding to assert the opposite. 

Oracle Puts in the Effort to Improve and “Automate” Updates (Through its Marketing)

Oracle puts most of its emphasis on improved updates and automation in marketing rather than anything real. The reality from the field is that Oracle Cloud products have historically not only been difficult to upgrade but have substantial downtime, as explained by Mark Dalton of AutoDeploy.

“Every Oracle Cloud customer has to deal with a maintenance window that starts out as 1 hour and runs into multiple days. If their systems are autonomous, why all this manual work? And why would any customer want this disruption to their business?”

While Larry Ellison is putting batteries in his presentation “reality distortion field generator” and making ludicrous statements to please Wall Street (which is easy, as Wall Street analysts don’t work with the actual software), AWS and Google Cloud Services are making true strides in upgrading and downtime reduction. The level of documentation on the topic of upgrades and downtime available at AWS and GCS is far beyond anything that either SAP or Oracle have. The reports from the field on AWS and GSC is also far beyond anything SAP or Oracle is capable of.

How Lock in Differs from On Premises ERP to Cloud ERP

On-premises software is the traditional enterprise software model where the software is taken by the implementing company after purchase and installed on the company’s servers, rather than the model where the software is “rented” and the software vendor hosts the software. The on-premises model is presently the dominant form of ERP software, and of course, has been since ERP was first introduced. However, many vendors provide SaaS or cloud-based ERP, and they use the term ERP even though their solutions are more limited in scope than on-premises ERP. This is not necessarily a bad thing, but it does confuse the issue somewhat when I want to use the term ERP; any number of SaaS-based ERP vendors could say that the conclusions in this book do not apply to them, and they would be right.

How the Lock-In Level Differs

SaaS vendors don’t have the same ability to lock in customers as do on-premises vendors. They are still relatively new and do not have the same history as on-premises ERP solutions, so it cannot be expected that the research I will quote will apply to them. In addition, my project experience is 100 percent with on-premises solutions, so when it comes to SaaS ERP, I do not have the same personal experience upon which to draw. SaaS solutions currently represent about 4 percent of all enterprise software sales, but because the software is less expensive, this represents a higher percentage of “seats” served than this 4 percent value would indicate.


S/4HANA Cloud is a glorified development box with a quarterly push of changes that cannot be relied upon for a company that is interested in doing something other than using the implementation to sell implementation services to other companies. S/4HANA Cloud contains a shadow of the functionality of S/4HANA on premises, which is itself a shadow of the functionality in ECC.

As for Oracle, it is misleading customers as to the upgrading required for the Oracle ERP Cloud. Ellison believes he can market his way around Oracle ERP Cloud’s limitations, using monopolistic profits to shovel money at Gartner to obtain a preposterous score, above that of Workday.

Oracle and SAP are examples of vendors that have historically offered on-premises ERP but which are now pushing cloud ERP. The concept of cloud is that the lock-in is less, but many on-premises vendors only have experience with locking in customers. And if one looks at the ERP offerings of many vendors, they do not match the cloud. Secondly, ERP systems are far more sticky than other types of applications like CRM. Therefore, the cloud receives far fewer of the benefits from the supposed lower lock-in of cloud systems.

Not all vendors are equal in the cloud. Cloud ERP works better with companies like Workday or ERPNext, which started in the cloud and have a long history of providing updates while minimizing disruption for their customers. This makes both SAP’s and Oracle’s cloud ERP a liability for current and potential customers. SAP and Oracle’s cloud ERP is a great story for Wall Street, but not much of a story for customers.

The Problem: A Lack of Fact-Checking of Oracle

There are two fundamental problems around Oracle. The first is the exaggeration of Oracle, which means that companies that purchased from Oracle end up getting far less than they were promised. The second is that the Oracle consulting companies simply repeat whatever Oracle says. This means that on virtually all accounts there is no independent entity that can contradict statements by Oracle.

Being Part of the Solution: What to Do About Oracle

We can provide feedback from multiple Oracle accounts that provide realistic information around a variety of Oracle products — and this reduces the dependence on biased entities like Oracle and all of the large Oracle consulting firms that parrot what Oracle says. We offer fact-checking services that are entirely research-based and that can stop inaccurate information dead in its tracks. Oracle and the consulting firms rely on providing information without any fact-checking entity to contradict the information they provide. This is how companies end up paying for items which are exorbitantly priced, exorbitantly expensive to implement and exorbitantly expensive to maintain.

If you need independent advice and fact-checking that is outside of the Oracle and Oracle consulting system, reach out to us with the form below or with the messenger to the bottom right of the page.

Financial Disclosure

Financial Bias Disclosure

Neither this article nor any other article on the Brightwork website is paid for by a software vendor, including Oracle, SAP or their competitors. As part of our commitment to publishing independent, unbiased research; no paid media placements, commissions or incentives of any nature are allowed.

Search Our S/4HANA Cloud Content


There are three different types of multi-tenancy

1. Shared application, a separate database
2. Shared application, shared database, a separate table
3. Shared application, shared table (pure multitenancy) – Bezemer & Zaidman

Cor-Paul Bezemer, Andy Ziadman, Multi-Tenant SaaS Applications: Maintenance Dream or Nightmare?, Association of Computing Machinery, Sept 2010

Thanks to Ahmed Azmi and Mark Dalton for feedback on the article.

Workday Bows Enterprise Reins Release Cycle, Phil Wainewright, Diginomica, 2013

Something that I did not know, as I do not study marketing outside of software marketing, is that against type marketing occurs when the brand is in decline. Examples of this are Oldsmobile, Buick (Is That a Buick?).


Rimini Street White Paper Five Upgrade Strategies or Oracle E Business Suite Customers to Consider

“Business Disruption — Upgrading to EBS 12 in a complicated process that could tie up your IT team for 12–18 months and have unintended consequences for the business. Sometimes an upgrade “backfires,” leading to unpleasant and expensive results. One CIO of a mid-market light fixture company commented around their Oracle upgrade: “Long story short, we went through the pain of doing this upgrade only to find that not only there’s no functionality there that we could use, but, also in essence, we had to upgrade our hardware because performance fell back and we’re still not at the performance we were before!”

The Real Story on ERP Book

ERPThe Real Story Behind ERP: Separating Fiction From Reality

How This Book is Structured

This book combines a meta-analysis of all of the academic research on the benefits of ERP, coupled with on project experience.

ERP has had a remarkable impact on most companies that implemented it. Unplanned expenses for customization, failed implementations, integration, and applications to meet the business requirements that ERP could not–have added up to a higher Total Cost of Ownership for ERP were all unexpected, and account control, on the part of ERP vendors — is now a significant issue affecting IT performance.

Break the Bank for ERP?

Many companies that have broken the bank to implement ERP projects have seen their KPIs go down— but the question is why this is the case. Major consulting companies are some of the largest promoters of ERP systems, but given the massive profits they make on ERP implementations — can they be trusted to provide the real story on ERP? Probably not, however, written by the Managing Editor of SCM Focus, Shaun Snapp — an author with many years of experience with ERP system. A supply chain software expert and well known for providing authentic information on the topics he covers, you can trust this book to provide all the detail that no consulting firm will.

By reading this book you will:

  • Examine the high failure rates of ERP implementations.
  • Demystify the convincing arguments ERP vendors use to sell ERP.
  • See how ERP vendors take control of client accounts with ERP.
  • Understand why single-instance ERP is not typically feasible.
  • Calculate the total cost of ownership and return on investment for your ERP implementation.
  • Understand the alternatives to ERP.


  • Chapter 1: Introduction to ERP Software
  • Chapter 2: The History of ERP
  • Chapter 3: Logical Fallacies and the Logics Used to Sell ERP
  • Chapter 4: The Best Practice Logic for ERP
  • Chapter 5: The Integration Benefits Logic for ERP
  • Chapter 6: Analyzing The Logic Used to Sell ERP
  • Chapter 7: The High TCO and Low ROI of ERP
  • Chapter 8: ERP and the Problem with Institutional Decision Making
  • Chapter 9: How ERP Creates Redundant Systems
  • Chapter 10: How ERP Distracts Companies from Implementing Better Functionality
  • Chapter 11: Alternatives to ERP or Adjusting the Current ERP System
  • Chapter 12: Conclusion