What Percentage of Oracle Customers are on What Oracle Database Version?
Executive Summary
- What percentage of Oracle database customers are on which version of their database?
- We cover the implications of these percentages for Oracle customers.
Introduction
Oracle has been introducing increasingly sophisticated versions of its flagship database products in the past few years. Garnering little attention in the technology press, unfortunately, is the resistance of many large enterprises to migrate to those databases and the resulting waste of Oracle support dollars.
Our References for This Article
If you want to see our references for this article and other related Brightwork articles, see this link.
Lack of Financial Bias Notice: The vast majority of content available on the Internet about Oracle is marketing fiddle-faddle published by Oracle, Oracle partners, or media entities paid by Oracle to run their marketing on the media website. Each one of these entities tries to hide its financial bias from readers. The article below is very different.
- This is published by a research entity, not some dishonest entity that is part of the Oracle ecosystem.
- Second, no one paid for this article to be written, and it is not pretending to inform you while being rigged to sell you software or consulting services. Unlike nearly every other article you will find from Google on this topic, it has had no input from any company's marketing or sales department. As you are reading this article, consider how rare this is. The vast majority of information on the Internet on Oracle is provided by Oracle, which is filled with false claims and sleazy consulting companies and SAP consultants who will tell any lie for personal benefit. Furthermore, Oracle pays off all IT analysts -- who have the same concern for accuracy as Oracle. Not one of these entities will disclose their pro-Oracle financial bias to their readers.
Oracle Database Version Usage Percentages
The data for this chart was acquired from HG Insights. This is the data that was pulled from their website on June 1, 2019. This company verifies the usage of applications and databases, but not all information is confirmed to the most recent month. The verifications are performed in a cycle, and this is how these verification services work.
By taking the counts for each Oracle database version, we could construct the following pie chart.
Overall, roughly an insignificant 6% of Oracle customers have migrated to the most recent version of the Oracle DB.
The question is why.
Our observation is that the Oracle database has become overly bloated in its most recent versions. Most companies do not need the extra functionality or the extra bloat and expense involved with Oracle 12c. For example, Oracle 12c adds a column store, which was put in place to combat SAP HANA, but which there is little use for in the database, as we covered in the article How Accurate with Bloor Research on Oracle In-Memory?
Implications of the Slow or Stunted Migration
The evidence is that a minority of customers have shown any interest in migrating to Oracle 12c, even though it was released in 2013. It also means that each successive release of the Oracle database is less capable of migrating the customers on the previous versions to the latest versions of the database. This means that Oracle is discussing its database features at conferences and in the IT media that their installed base by in large does not see.
This has several important implications. And they are as follows:
- The Usage of Open Source Databases
- The Oracle Support Question
The Usage of Open Source Databases
When Oracle discusses their database, they nearly always discuss the most recent version. The standard Oracle sales tactic is to state that only the Oracle DB has (insert in the ____ ) functionality. This is often stated by Oracle about HA (High Availability) or RAC (Clustering), even though these functionalities are, in most cases, better and more efficiently deployed from the virtual machine instead of the database layer. Secondly, some of the functionality, such as those in the Autonomous version of 18, does not work. We covered this separate topic in the article How Real is Oracle’s Autonomous Database?
Other areas of functionality have a low overlap with the primary Oracle customer market. For example, Oracle 12c offers multitenancy as a main new area of functionality. However, SaaS vendors primarily use multitenancy, which are less likely to be customers than companies that use the Oracle database internally. The shallow penetration of Oracle illustrates this into the startup market, as we covered in the article How Appealing is the Oracle Global Startup Ecosystem?
Beyond these topics, most customers do not agree with Oracle that they need the most up-to-date features that the more recent databases provide. Oracle would have a challenging time explaining this if it was brought up in discussion. Oracle representatives state that they know their customers very well and that they have the best RDBMS in the world, and that they know how to develop their RDBMS to meet the needs of customers. Given that the vast majority of Oracle DB customers are not running anywhere near the more recent versions of the Oracle database indicates one or several of these assertions are not true.
To determine the extent to which a high percentage of Oracle customers are using versions from years ago, we have compiled the following table.
Year of the Database Version Introduction
| Database Version | Year of Introduction/Stable Release |
|---|---|
| Oracle 8i | 1998 |
| Oracle9i | 2001 |
| Oracle 10g | 2003 |
| Oracle 11g | 2007 |
| Oracle 12c | 2013 |
| Oracle 18 | 2018 (not enough customers to track) |
| Oracle 19 | 2019 (not enough customers to track) |
What is the Proper Comparison of Open Source Databases to Oracle?
Open-source databases are easy to upgrade and have no license or support liability. This means that when customers compare open source options, the comparison most of them would need is to compare the 2003 version of the Oracle DB (10g) to the 2019 version of PostgreSQL or MariaDB.
Why is this the proper comparison?
Because this is the version that the Oracle DB customers are most often using; however, this framework is nearly never used when discussing the Oracle database. Invariably the discussion switches back to the most recent Oracle database version.
Secondly, Oracle’s newer or more advanced database functionality is increasingly in direct competition with virtualization technology. Our view on Oracle’s approach is that they are stuffing their database with functionality that does not belong in the database in the vast majority of cases. Oracle does not have effective virtualization technology. Oracle VM is very poorly thought of and very infrequently used.
Secondly, virtualization is a primary strategy deployed to reduce the number of CPUs necessary to run the same number of database licenses. Therefore, as Oracle prices their databases per CPU, they have a built-in financial incentive to push CPU utilization down, not up. In part because of this, most of what Oracle has done to its customers is to provide inaccurate information about virtualization, even questioning companies’ legal right to perform software partitioning full stop when there is nothing in any of Oracle’s contracts that denies this right.
The Oracle Support Question
The support question regarding Oracle (that is, should companies continue with Oracle support or look for other alternatives, including taking support internally) and applies more to companies that use the Oracle database rather than the database along with the applications.
The Multiple of the Initial License Paid Per Database Version Since Purchased
| Database Version | Overpayment Multiples of Initial Net License Charge | % of Oracle DB Customers |
|---|---|---|
| Oracle 8i | 13% | |
| Oracle9i | 21% | |
| Oracle 10g | 29% | |
| Oracle 11g | 31% | |
| Oracle 12c |
This table shows the multiple of the initial license price that has been estimated to be paid by customers since the software was purchased, and what percentage of the customers (roughly speaking) have paid this support money to Oracle over the years. For Oracle 10g, a typical customer would have paid 3.75 times the license cost in support if they purchased the database version around the software release in 2003.
Two factors would reduce this multiple, and this is the deployment of 3rd party support and the second being companies that bring Oracle support internally.
- There is some leakage of the support payment. “Leakage” is the loss of support revenues to Oracle through companies pulling support internally or using 3rd party support. However, even the largest 3rd party support company, Rimini Street, has less than $250 million per year. While Oracle’s support business is roughly $18 billion per year. Although it should be noted that as Rimini Street prices its support at either 1/2 or less than 1/2 of what Oracle charges, each dollar of Oracle support that goes to Rimini Street (or other 3rd party support providers) cuts Oracle’s support revenue by over 2x the revenue of the 3rd party provider. However, we estimate that while Rimini Street provides support for other vendors, at least 80% of its revenues are from Oracle support.
- There are other 3rd party support companies, of course. Still, all of them significantly smaller. It is difficult to see third-party support combined with bringing support internally, reducing Oracle’s support revenue by even $1.75 billion per year.
Estimated Leakage From Oracle Support
| | | |
|
|---|---|---|---|
| 1 | Oracle Yearly Support Revenues | ||
| 2 | Estimated Leakage to 3rd Party Support and Support Brought Internal |
If this estimate is roughly correct, then Oracle still has at least 91% of the overall support market for Oracle support. Oracle does little to improve its acquired applications, and Oracle customers are not purchasing the most recent versions of the Oracle database.
This means that every year that passes, Oracle Support becomes an increasingly wasteful proposition. Support is being paid on items where the bulk of development is at least ten years old. Any software vendor would love to collect support revenues on stable old applications into perpetuity.
The Oracle Consulting Advice Question
Oracle Consulting Partners can be useful suppliers if the objective is to perform work in line with what Oracle wants the customer to do. However, as soon as the objective contradicts what Oracle wants, the Oracle Consulting Partners become a liability because they lack any independence from Oracle. They lack the right to provide information contrary to Oracle’s views, even if they wanted to.
A lack of independent advice from Oracle Consulting Partners is a particular problem as many customers seek to reduce their Oracle exposure. Oracle repeatedly scores as the vendor that customers would most like to minimize their exposure to in the future in every survey of this type.
- Oracle Consulting Partners wants customers to upgrade their Oracle database versions, not necessarily because it is good for customers, but because they can charge for making the change. They add skills and experience to their resumes.
- Oracle Consulting Partners cannot credibly advise on license matters or 3rd party support or on using open source databases. Every answer to every question from Oracle Consulting Partners is to maximize the customer’s money to the Oracle Consulting Partner and Oracle.
- Oracle Consulting Partners are conduits of information about the customer to Oracle. They have frequent calls and meetings with Oracle, particularly the sales rep, that the customer is not informed about occurring.
How Oracle Responds to a Potential Support Contract Loss
| Oracle Strategy | Description |
|---|---|
| Oracle provides inaccurate information regarding how this impacts risk which is intent on maximizing the level of fear at their customers, even though many companies have quite effectively pulled support internal or moved to 3rd party support. | |
| This FUD is communicated by Oracle cloaked as a genuine concern for the wellbeing of their customers, which is not credible as this is Oracle. Oracle shows an interest in their customers in two specific periods, when desiring to sell more software to them, and at the support renewal timeframe. | |
| The intent of Oracle is to reach out to people within the customer that are less aware of the real implications and benefits of moving off of Oracle Support and to try to sow seeds of concern among anyone who is susceptible. This is most effective when it can get to the most senior levels of the company, therefore, Oracle seeks to send their warnings to the widest possible distribution list, hoping that it will reach even distant decision makers. The intent is to get them to put pressure on those closer to the situation and who have the domain expertise to know what Oracle is doing. | |
| Letters are sent to legal to try to bring extra "gravitas" to the situation. These are clearly intended to make an impact, rather than be anything real. Oracle cannot sue customers for not renewing support contracts. But Oracle has an enormous number of attorneys and they use them for maximum effect. | |
| Oracle presents staying with their support as the "rational" decision, while the moving off of their support as "irrational" and dangerous to the company. | |
| And at no point during the Oracle histrionics does Oracle make it known that they have a greater than 90% margin on support. Oracle presents the entirety of opposition to losing support is their "concern for the customer." This is not about money or margin. |
Performing a Full Spectrum Analysis
Moving away from Oracle Support is a perceived risk that requires a great deal more investigation by customers than simply not upgrading the Oracle database. And part of that analysis requires fact-checking claims made by Oracle that have no other objective other than maintaining their high margin support business.