How SAP Has Quietly Changed Strategy on HANA and SAP Oracle

Executive Summary

  • SAP’s original plan for dominating the database market with HANA was overly optimistic. The software was given a massive marketing and PR push, and the company aimed for a penetrate and radiate approach, in which customers would adopt HANA and then purchase follow-on database products.
  • In May 2017, however, SAP came to an agreement to continue supporting Oracle database products through 2025. This suggested SAP was quietly distancing itself from the aim of dominating the database market with HANA.
  • Factors driving the shift away from HANA included that the product didn’t deliver on its marketing claims and that it failed to receive projected market adoption.

Introduction: Changes in Policy

SAP has been communicating that it will replace Oracle DB with HANA on a mass scale. You will learn about SAP’s original plan with HANA and then discuss the recent unpublished change in policy regarding SAP Oracle, which says a lot about what SAP has internally realized about HANA and its future.

Considering The Original Plan for HANA

SAP’s original plan for HANA was optimistic in the extreme. It was championed by Hasso Plattner, and the plan was so extensive that describing all of it would be its article. However, in included the following:

  • Books: Four books written by Hasso Plattner around in-memory computing.
  • Marketing and PR: A massive marketing and PR push for HANA, what must have been one of the greatest efforts and expenditures in the history of enterprise software.
  • Exaggerated Benefits: The proposal of enormous benefits to HANA, some of which are cataloged in the article When Articles Exaggerate HANA’s Benefits.
  • The Illusory Concept of Simplification: The proposal that because of HANA’s simplified model, that HANA would massively reduce the complexity of SAP as well as the business processes on projects.
  • Artificially Limit S/4HANA to HANA: Here SAP took its best-selling ERP system and in the new version, called S/4HANA, limited it only to HANA. This was always an anti-competitive move as there was never any true technical reason to perform this limitation. But as usual, the Federal Trade Commission as well as the European Trade Commission were not paying attention and did not step in. The concept was that many customers would migrate to S/4HANA and then they would be forced into purchasing HANA. But that approach has not yet worked out as S/4HANA adoptions have been incredibly low, which is covered in the article, How SAP Controls Perceptions with Customer Numbers.
  • Penetrate and Radiate: SAP’s had a follow-on strategy after getting HANA into the data layer. SAP has all types of follow-on databases and products that intended to pitch to customers after they had purchased HANA. One of the issues with HANA is that SAP does not divulge all of the components that are necessary to run HANA when one buys a copy of HANA. The customer finds out about complementary products after they have made the purchase. But secondly, SAP had positioned lagging databases, such as SAP IQ, formally Sybase IQ to perform data archiving for HANA. The argument was going to be that SAP IQ communicated much better to HANA, and this would, the plan went, force out database companies that were much better at archival, in favor of SQP IQ.

SAP’s Plan to Grow Into the Database Layer

As Oracle had acquired its way into the application layer using its database market share, SAP intended to grow (rather than acquire) their way into the database layer. In this way, they intended to push Oracle out of “their” accounts. For all software vendors, you should be on notice. When you sell software into an SAP account, they believe they own that account.

SAP’s move into the database layer was based on inaccurate information. Oracle acquired its way into the application layer. This is something I am not a fan of for several reasons:  as acquisitions lead to concentrations of markets, and concentrations of markets invariably lead to less competition and lock-in.

This lesson needs to be relearned by every generation. But Oracle did have real applications that had real value propositions. So while I disagree with whether Oracle’s massive acquisition spree should have been approved by the Federal Trade Commission (which seems to spend most of its time eating donuts and taking long naps), it was not in itself deceptive, except that Oracle promised to add value to its acquisitions, when it really didn’t.

The Long-Term Problems with Software Aquisition

Here is why software acquisitions are in most cases not a good idea for the government to approve. They relate to:

  • Market Inefficiency
  • Market Concentration

Market Inefficiency

Acquisitions are typically made by companies that are dominant in one space. Oracle in databases. SAP in large company ERP. Microsoft in operating systems and office suite. These “excess” profits are then used to purchase often companies in related areas. However, this results in lower profits. Over half of Oracle’s profits still come from databases. Three-quarters of Microsoft’s profits come from Windows and Office (not XBox, phones, etc.).

Acquisitions (and even growth into non-core areas) means a decreasing efficiency of the enterprise as bureaucracy increases and the additions are less effective than the core. About the only company that seems to counter this rule is Amazon. Amazon seems to be able to move into anything they do and be highly effective at doing so. They even migrate into areas like AWS that have a higher profit margin than the area they started in (online retailing).

Market Concentration

Acquisitions lead to concentrations of markets, and concentrations of markets invariably lead to less competition and lock-in. This lesson needs to be relearned by every generation. But Oracle did have real applications that had real value propositions.

So while I disagree with whether Oracle’s massive acquisition spree should have been approved by the Federal Trade Commission (which seems to spend most of its time eating donuts and taking long naps), it was not in itself deceptive.

However, SAP’s statements about HANA were always inaccurate.

The Recent Change in Policy on Oracle

SAP’s plans for HANA were so extravagant that they fell into the category of fantasy. I was working through a consulting company several years ago and attended a meeting where several men from SAP came over to educated us on HANA and S/4HANA. What was pitched as an educational presentation became one of the most aggressive sales pitches for HANA and S/4HANA or any other product for that matter that I have ever witnessed.

Taking a Deep Drink of SAP HANA Kool-Aid from SAP Marketing

At one point, one of these highly confident men stated that “there was going to be no innovation in SAP outside of HANA and S/4HANA.”

“there was going to be no innovation in SAP outside of HANA and S/4HANA.”

This is a ridiculous statement, as it implies that no other SAP products would receive any development (at SAP, “innovation” is a euphemism for “development.” And of course, ceasing all development on all SAP products aside from HANA and S/4HANA would be a very bad, and very illogical move. It is difficult to overestimate the hyperbole that surrounded HANA, and to some extent still encompasses HANA. HANA took over SAP conferences, it took over the marketing messaging of the SAP consulting companies, with each attempting to out-HANA one another.

At one consulting company that I was contracted to, a salesperson with an amazing pipeline of HANA deals was brought into the company without checking anything about him, the desire for some HANA expertise was so high that one. But probably other consulting companies would grab any person with HANA on their resume, and at very favorable terms to the resource. On LinkedIn, virtually overnight, people added HANA not only to their experience to their job titles. There were HANA Solution Architects and HANA Marketing Specialists.

The SAP ecosystem could simply not get enough of HANA!

The Problem with Not Making Sense

There was just one little problem, and it took me a few years to figure it out. But HANA simply did not make any sense. I began investigating HANA from multiple dimensions several years ago, first because the hype around HANA smelled strange. I reminded me too much of the Netweaver marketing program that preceded it. In fact, I compared HANA to Netweaver in the article Has SAP’s Relentless HANA Push Paid Off? 

“Information about the database reseller contract between SAP and Oracle (May 30 2017)

The following text has been agreed upon by SAP and Oracle:

SAP and Oracle have agreed to a long term extension of SAP’s global reseller and support relationship. For more than twenty years, SAP and Oracle have worked together to provide customers with a supported SAP/Oracle environment, running SAP applications and an Oracle database. During this extension SAP customers can continue to acquire Oracle licenses from SAP and Oracle to support their SAP business applications, and SAP and Oracle will continue to offer support for the combined Oracle/SAP Offering.”

This is one of the longest partnership announcements SAP has ever signed. SAP sales people were until recently telling customers that support for Oracle would stop in 2017. This contract between SAP and Oracle changes that.

How SAP is Underemphasizing This

Here is the entire text of the SAUG, the SAP Australian User Group article on the contract.

“SAP and Oracle have agreed to a long-term extension of SAP’s global reseller and support relationship. For more than twenty years, SAP and Oracle have worked together to provide customers with a supported SAP/Oracle environment, running SAP applications and an Oracle database. During this extension SAP customers can continue to acquire Oracle licenses from SAP or Oracle to support their SAP business applications, and SAP and Oracle will continue to offer support for the combined Oracle/SAP offering.

Between now and 2025, SAP will continue to offer integrated support for all Oracle/SAP environments (full time and runtime).  SAP will continue its current resale practices Oracle runtime licenses for the first part of the extension period, through December 31, 2023.  For the final two years of the extension, between January 1, 2024 and December 31, 2025, SAP will sell Oracle runtime licenses only to existing Oracle runtime customers.”

That is 144 words. One hundred and forty-four words on a significant change in contractual terms that all but spells out a massive change in SAP’s HANA policy. There is no introduction to the article, and it is the identical language as in the contract announcement. It as if SAUG is ashamed to write it.

The Future of HANA?

Here is another problem SAP has. Without a large installed base of HANA, they are at a disadvantage regarding much more popular databases regarding the number people reporting things, the “community.” My view is that SAP exaggerated HANA hoping to “fake it until you make it.”

But without that growth, the marketing push is fading, and it will become more obvious that HANA is not a fast growing product, and so its inertia becomes stalled.

Does this look like a growing database? According to DB-Engines, which ranks database popularity, HANA has been declining for eight months. It should be recognized that SAP pumped a very large amount of marketing dollars and company energy into pushing HANA, actually to the detriment of other of SAP’s offerings. This is why HANA’s future does not look good. HANA is out of runway on applications where it adds some reasonable value and is now receiving blowback from SAP’s exaggerations versus real-world performance. IBM and Oracle are now increasingly pushing back on SAP — even though both companies have relationships with and depend upon SAP for revenues.  

Taking the Creme off the Top of the Coffee with BW Conversion to HANA

As you may have read in my recent article, SAP already took the creme off the top of the coffee when it sold into BW. After the BW the value proposition becomes considerably weaker as HANA is just an analytics database masquerading as something more.

Now SAP is locked into this strategy, but it does not make any sense.

I have focused on proposing that SAP is legacy. Its new products have turned out to be poor values had had a weak implementation history on accounts, as is covered in the article, How SAP is Now Strip Mining its Customers. It’s time to keep the ERP system, but stop buying so much from SAP. Our view on this is so emphatic that it is a primary theme on the Brightwork home page.

Of course, that is radical to SAP customers. But really, none of what they said would happen with regards to the benefits of these new non-ERP systems worked out. It was in total, a massive misallocation of resources into a failed strategy. But the customers that did this can’t admit the error. It is interesting that decision makers are very rarely held accountable for bad decisions, and how bad so many of these decisions are when analyzed in detail.

As I mentioned in my article last year, the other teams within SAP will not like to be discounted heavily and then charge full price for HANA. It gives the HANA guys unfair share of recognition within the company.

Who Was Wrong and Who Was Right?

Well, obviously SAP was one. SAP claimed they would be the #2 database vendor in the world back in 2017. According to DB-Engines,

  • They have the 13th most popular database in the SAP Adaptive Server.
  • They have the 19th most popular database in HANA.
  • They have the 49th most popular database in SAP IQ (formerly Sybase IQ) and the 52nd most popular database in SAP SQL Anywhere.

Does that add up to the #2 spot? I need someone good at math to help me out with this one. Does anyone have Bill McDermott’s number? His data is always so reliable.

It seems reasonable for customers to ask SAP sales people to apologize to them for telling them that SAP would not support Oracle after 2017. That looks quite dishonest, especially after that date has now passed.

Hyping HANA

In the many articles that followed on HANA, I was repeatedly told by people that worked for SAP or for large SAP consulting companies that I did not get it. That HANA was going to be a huge success, and I simply lacked the vision. I was exposed to many discussions, by people that knew little about databases or comparative database designs, that HANA was the future. This is what it is like working in SAP. When you work in SAP, you are expected to mindlessly and loyally support that notion that SAP is the best at everything. And that the mere existence of an SAP product, means that this product should be used — because it is SAP. The lunacy of people talking about things they did not understand reached a fever pitch when I was condescended to by recruiter about how HANA would be so great and that I was not getting it. That is right people, a recruiter — that is a person who never touches much less sees any of the systems that they place people for, who worked for Dickenson and Associates, explained to me that I needed to get on the “HANA train.” He explained that his only reason for supporting HANA was not to staff HANA resources, but was related to making the world a better place. (I can’t make this stuff up)

How To Deflect That You Were Wrong About HANA

A little while ago I began to notice a pattern in the rhetoric with people I would debate, which is the topic of the article How to Deflect That You Were Wrong About HANA. This described an argumentative style that moved away from the main proposals of HANA when challenged on them, to stating that HANA was “so much more” than a database. Then the topic was changed to the HANA Cloud Platform or to HANA Studio, etc.. SAP had simply come up with a number of products and had slapped the name HANA on them. This created a lot of confusion but allowed HANA proponents to essentially move the topic in other areas, and to declare that HANA was a “platform.”

Who Got it Wrong?

  • Hasso Plattner: Hasso was the main driving force behind HANA. Hasso is so wealthy and powerful within SAP that no one can “No” to him. However, Hasso greatly overestimated the benefit that something like HANA would bring to SAP. Hasso thought he had found the magic pill for giving SAP a new life. However, he had SAP invest big into databases at the exact time when less expensive databases and open source databases were about to erode the margins and sales of the proprietary database vendors. This is covered in the article How Accurate is Seeking Alpha on the Decline of SQL? SAP clearly invested in the wrong thing by going into the database, and then by bringing out such a premium priced product. Hasso was wrong on HANA not only in that the market did not respond as Hasso predicted, but most of what Hasso said about HANA was simply incorrect.
  • Bill McDermott: Bill McDermott is more PR spokesman for SAP than CEO. When Bill McDermott speaks, I never get the impression he has any idea what he is talking about. Bill is an excellent salesman, but essentially he excels as explaining things in a way that make it seem as if the objective the listener wants can be achieved very simply, but to do this, he dispenses with any detail that complicates the messaging. For example, Bill stated to a Fortune Magazine interviewer that one could do both transactions and Big Data all in a single database. However, connecting HANA to Hadoop is not using one database, is it? For those looking for career advice, this is how you can make over $120 million per year without having to do much work. It’s a nice gig if you can get it.
  • Vishal Sikka: Vishal Sikka was for a time was the “architect of HANA,” something which is a bit doubtful given his level in SAP. Vishal Sikka is almost indecipherable when speaking. I analyzed Vishal’s statements in great detail about HANA and concluded that he was massively exaggerating the overall reason for and benefit of HANA. After reading his comments, I would never listen to Vishal on any topic as long as his name is attached to it. Interestingly, Vishal left for Infosys in 2014, which may have been for reasons unrelated to HANA’s progress. However, for someone who is thought of being so intelligent, HANA makes little sense. But at SAP, at that level, you do what Hasso tells you to do. If HANA was in part his vision, one does not need a Ph.D. in computer science to recognize that you don’t put an ERP system on what is nothing more than a gussied up analytics database.
  • Bluefin Solutions: Bluefin Solutions, more than any other consulting company went all in on HANA. They wrote articles, lead by John Appleby, that to me, clearly simply copied and pasted from information provided to them by SAP. In one article John Appleby stated that “SAP is finished on Oracle.” This is a curious statement as the evidence he provided related to S/4HANA. John Appleby should know that SAP is about a lot more than S/4. Therefore even if Oracle were finished on S/4 (which I don’t think it is, as covered in the article Why SAP Will Have to Backtrack on S/4HANA), it would not mean that all SAP applications are finished on Oracle. Four and a half years after Appleby made this statement, I wonder if he could guess what the majority of SAP applications use as a database today (hint, it’s not HANA). This is why consulting companies that are partners with SAP are unreliable sources of information on SAP. And this is particularly true of public statements released in the media. After Appleby was promoted at Bluefin Solutions, Steve Chaflen took over the position of HANA promoter from him Appleby. Not only John Appleby and Bluefin Solutions but all consulting partners of SAP release inaccurate information about SAP as they compete to see who can more intensively shine the shoes of SAP in return for special consideration.
  • Gartner: Gartner receives probably around $100 million from SAP per year. All of this undisclosed. The primary analyst assigned to SAP is Donald Fienberg who is also Gartner’s liaison to SAP. This puts Gartner in a bit of a problematic area as there was no line drawn between the analyst work Donald Fienberg was performing and his role as essentially an SAP relationship manager for Gartner. Donald Fienberg made repeated inaccurate and promotional statements about HANA, such as the following “Now SAP can set up a database to do my transactions and analytics with everything virtual. People have wanted to do both in the same database for years, but have not been able to do it because of the discrepancy between the two types of data.”These guys (SAP) are first, period. That’s a good thing. They have got a two to five-year head start on everybody else..” Oracle 12c, which has this ability Donald was describing was released six months after this article was written. IBM DB2 Blu (which offers the same ability) actually preceded HANA. And Microsoft SQL Server had a column store index in 2012, therefore before Donald Fienberg made this statement.
  • Forrester: Forrester was paid by SAP to write a report that predicted a reduced TCO for HANA, which SAP then took and marketed as proving that HANA “did reduce TCO.” This deftly changed a forecast into an actual. Forrester made some inaccurate assumptions in the report including the use of a runtime license for HANA. This is not a production license, and there was no reason for Forrester to have used this except — to reduce the cost of HANA artificially. This is covered in the article How Accurate was Forrester’s TCO Study for SAP HANA?  
  • ASUG: A usual, ASUG served as simply outlet for SAP marketing proving once again that they provide no filter to anything SAP says or does.
  • Fortune Magazine: Through a series of what appears to have been paid placements, Fortune provided SAP with an outlet for SAP to say anything it wanted on the future. Brightwork will be sharing an analysis of these articles, and how wrong they were in the coming months.
  • IT Leaders Who Purchased HANA: IT leaders that supported moving to HANA were tricked. Even in HANA’s “golden case,” which is moving BW on HANA, there was never any reason to do this. SAP customers could have stayed with SAP Oracle and received a far better database with better performance than HANA at a lower price. I believe similarly this may apply for DB2 and SQL Server, which also can store data in a columnar store, but I simply have not investigated the technical comparisons as I have with these databases as I have with SAP Oracle as is covered in the article Which is Faster HANA or Oracle 12c. However, Bloor Research recently performed, a study that while funded by IBM, I have reviewed and considered it to be quite accurate. In this study, which I have analyzed in the article How to Best Understand Bloor’s Research on HANA, and have given it one of the highest accuracy ratings of any publication I have analyzed. Those IT leaders cost their companies and directed monies towards SAP’s vanity project that should have been used elsewhere. And this is no small thing. I have visited companies that needed investment quite badly in neglected areas and instead decided to use that money to invest in HANA. What a monumental waste of resources!

This is a highly truncated list. Aside from individuals who were wrong, every single large SAP consulting company, Accenture, IBM, etc. all of them promoted HANA, and all of them got it wrong. Their support of HANA was so extensive that to analyze all of it would require a book.

This is why I have often said that if I want false information, I can always find it at a big consulting company. Whatever SAP marketing happens to be saying at the time, the major SAP consulting partners all in on the “vision.”

Who Got it Right?

  • Larry Ellison: Larry Ellison criticizes anything reflexively that is not good for Oracle. But you have to hand it to him; he made some very valid criticisms of HANA. Probably one of his best lines was in 2015 when he said:“For years SAP offered up lots of benchmarks for its database systems, Ellison said. But when SAP introduced HANA, its in-memory database, the company didn’t use any of its existing benchmarks —“ and they had a dozen of them,” Ellison said. They said they were going to write a brand new benchmark for HANA.“Now, call me cynical,” he continued, “but I guess this was to make HANA look good. It was a benchmark that allowed HANA to distinguish itself. And we decided to run the same exact benchmark on Oracle. And we ran more than twice as fast as they did.”By the way, SAP has not certified our benchmark. We’ve sent it to them…over and over and over again. And for some reason, they don’t want to certify.…The facts are, the entire SAP cloud runs on Oracle, not HANA. SAP Ariba in the cloud runs on Oracle, not HANA. SAP SuccessFactors in the cloud runs on Oracle, not HANA. SAP Concur runs on Oracle, not HANA. I don’t know what HANA runs, but it’s not [SAP’s cloud applications].” Every one of these statements by Larry turned out to be true. SAP did play around with the benchmarks, in my view because the HANA benchmarks for anything but read access are poor. This is something I have verified even with people inside of SAP. And secondly, the acquired applications are not moving to HANA. Right again Larry.
  • Peter Goldmacher: In the article How Popular is SAP HANA?, I observed that Peter Goldmacher pointed out “In a research note, Cowen analyst Peter Goldmacher did a deep dive into SAP’s HANA results. SAP has said that HANA’s two-year license compound annual growth rate through fiscal 2013 has been about 120 percent. If that growth rate is true, Goldmacher noted that “the other 90% of SAP’s license business, apps and business intelligence, is growing at a combined two-year CAGR of roughly 2%, materially below category growth rates.” Peter was right. As with CRM growth many years before, SAP was simply shifting revenue to HANA from other sources. This is virtually the same thing that SAP did with CRM beginning 15 years go. SAP pushed back hard on Peter, issuing a nonsensical statement about what percentage of HANA’s sales were stand-alone deals. Every time HANA was criticized, SAP pushed back in the media through a spokesperson. And, every single time I checked what the SAP spokesman said in defense of HANA, it ended up being inaccurate.
  • Brightwork Research & Analysis: We wrote the most comprehensive and detailed critiques of HANA available on the internet. As more information increasingly comes from the field, it turns out that even we missed areas of weakness of HANA. Our research into HANA is verifiable on my LinkedIn profile article page as well as the SAP HANA (and S/4HANA) sub-blog.

Conclusion

A huge number of people have been proven wrong about HANA. HANA fatigue has hit not only customers but SAP as well. Other groups inside of SAP are tired of playing second fiddle to the HANA teams.  

The word from inside of SAP under brackets of “you did not hear this from me, but ..” SAP now realizes that it cannot come anywhere near approaching the uptake of HANA that they had hoped.

  • This is the reason for extending its relationship with Oracle.
  • This is why SAP does not want to talk about it and why they want it kept as quiet as possible. 

HANA will become just another of SAP’s offerings now. It will not become, as Bill McDermott once said, the “basis for everything that SAP does.” More information is getting out that HANA is just not as advertised, and I predict from here on out, SAP going to lose far more arguments on HANA than it will win. 

As HANA’s time in the sun has come to an end, in a future article, I will cover how SAP will have to reverse itself on HANA. It will be tricky, as it must be done in a way that will not offend Hasso Plattner. Hasso must be handled very delicately. He must be told that his vision was an absolute genius, but SAP customers simply lacked the ability to see his fantastic long-term vision.

Brightwork Disclosure

Financial Bias Disclosure

This article and no other article on the Brightwork website is paid for by a software vendor, including Oracle and SAP. Brightwork does offer competitive intelligence work to vendors as part of its business, but no published research or articles are written with any financial consideration. As part of Brightwork’s commitment to publishing independent, unbiased research, the company’s business model is driven by consulting services; no paid media placements are accepted.

HANA & S/4HANA Question Box

  • Have Questions About S/4HANA & HANA?

    It is difficult for most companies to make improvements in S/4HANA and HANA without outside advice. And it is close to impossible to get honest S/4HANA and HANA advice from large consulting companies. We offer remote unbiased multi-dimension S/4HANA and HANA support.

    This article is free, we do not answer questions for free. Filling out this form is for those that have a budget. If that describes you, just fill out the form below and we'll be in touch asap.

References

https://www.g3g.com/hubfs/White_Papers/Projected_Cost_Analysis_of_the_SAP_HANA_platform.pdf?t=1455200742977

https://www.saug.com.au/news/saps-global-reseller-and-support-relationship-with-oracle-extended

http://www.oracle.com/us/solutions/sap/introduction/overview/index.html

http://www.businessinsider.com/sources-why-vishal-sikka-left-sap-2014-5

https://www.forbes.com/sites/oracle/2015/10/29/larry-ellisons-world-view-7-observations-straight-from-the-hip/

http://www.computerworlduk.com/it-vendors/will-oracles-customer-base-migrate-sap-business-suite-on-hana-3420138/

http://blogs.gartner.com/donald-feinberg/2014/09/28/in-memory-dbms-vs-in-memory-marketing/

*http://www.simba.com/blog/donald-feinberg-and-vishal-sikka-explain-about-in-memory-and-sap-hana/