If Noah had built HANA, what would SAP's cloud strategy be?
- SAP's cloud strategy muddles its messaging around HANA and cloud, leaving co-founder Hasso Plattner cast as Noah and HANA as his Ark
The analogy came to me as I sat listening earlier this week to SAP executives presenting an overview of the company's cloud strategy to industry analysts. In SAP's messaging, the tsunami of data from the Internet of Things looms large. Hence the notion came to me of HANA as the ark built to survive this data inundation — which casts Dr Plattner in the role of Noah.
The trouble is that SAP is muddling up two sets of messages. There's a tendency to talk about HANA and cloud as if the two are interchangeable. But putting both into the same narrative makes it harder to understand the merits of each. Here's my take on the various elements of SAP's product strategy as presented this week.
HANA as data platform
There's a cloud angle to HANA but it's not the core of what HANA is. By positioning a lot of the marketing of HANA around its two cloud incarnations (more on those later), the vendor is making it sound like a cloud platform. It can be but that's not what it's about.
HANA was conceived before cloud really got on the radar at SAP. Its primary mission was always about processing and analyzing data in a much faster and more instantly flexible way than previous database platforms. That's partly thanks to its in-memory architecture, although that technology has become a more or less de-facto requirement for transactional databases these days. More important is its ability to readily instantiate new data views rather than being tied to a predefined table structure.
Of course there's a lot more technical detail to it than that, but the key takeway is that HANA is about handling massive volumes of rapidly changing data. Which is why the Internet of Things is so often emphasized in the marketing — it's the most tangible expression of the need for enterprises to overhaul their data processing capabilities.
But there are plenty of other factors that SAP could also talk up. I've written recently about the headache of two-speed IT many enterprises face. At some point, organizations are going to have to upgrade their core systems to operate with the same agility and speed as their outward-facing systems in digital marketing, e-commerce and mobile. HANA is SAP's answer to that quandary.
Yet instead of leaping aboard, customers are holding back, leaving Dr Plattner looking like an isolated visionary fuming against the stubborn complacency of the crowd. I suspect that a large part of the problem is the conflation of cloud messaging with HANA messaging.
HANA and the cloud
It's not unreasonable to talk about HANA and cloud in the same breath — but there needs to be more clarity about how the two interact. There are two dimensions to this.
The first dimension is simply that a lot of the pressures I mentioned above that will lead enterprises to require more agility and performance in their core systems is caused by the cloud, in its most generic sense. We are moving towards a digitally connected future, when doing business will mean having information and resources readily available on demand, iteratively adapting to changing requirements. HANA is designed to be the engine that drives this capability into SAP's application portfolio.
HANA doesn't have to be in the cloud to fulfil that role, so long as it runs somewhere with the network infrastructure required to connect into resources that reside in the cloud.
The second dimension is that running HANA on a cloud platform is a really good way of helping you to a) make all those connections b) scale out to cope with all of the flexible, on-demand processing we discussed earlier and c) more easily upgrade with new application functionality as SAP makes it available. That is why the preferred environment for HANA is one of the two cloud platforms, HANA Enterprise Cloud (HEC) and HANA Cloud Platform (HCP).
But let's be clear because the nomenclature is now getting very confusing. Either of these cloud platforms can run in an organization's private datacenter and may actually turn out not to be very well connected or scalable or upgradeable, depending on how the organization has chosen to implement it. Just having 'cloud' in the name doesn't automagically allow it to inherit all of the miraculous attributes of a true cloud application.
SaaS as the vehicle for growth
OK, I could go on about HEC and HCP and how no one really groks the difference between the two, mainly because at heart they're both the same thing (ie HANA), but implemented differently. Instead let's run with that thought about true cloud applications.
One of the most interesting contradictions within SAP's cloud messaging is that the company is rightly proud of its cloud application user base and revenue numbers. It trumpets claims that it is the number one cloud applications vendor in user numbers and number two in revenue.
Whether you accept those rankings or not, none of it is anything to do with HANA. We know that 95 percent of cloud revenue results from SAP's acquisitions of SaaS vendors, primarily Ariba and Successfactors, more recently bolstered by Hybris and Fieldglass.
As Den Howlett already reported from Tuesday's presentation, last year's cloud revenue of €1 billion is projected to grow at 35 percent a year and thus to have tripled by 2017. It seems pretty clear that virtually all of that growth will be from those same SaaS product sets.
Talk about clouding the message. The muddling of HANA and cloud is bad enough. But now we realize that all of SAP's claimed cloud market leadership rests on an acquired, cloud-native product portfolio. Will that gloss rub off on HANA? SAP clearly hopes so.
In fairness, those products are now in the process of moving across to run on HANA. Ariba is already there, Hybris is moving customers across now and Successfactors will follow. Tim Minahan, CMO for SAP Cloud and Line of Business said on Tuesday that putting Ariba onto HANA yielded a massive 50-fold speed improvement, which in turn grew usage five-fold because customers found they could run reports in minutes that previously took too long to be worth running.
But the key point I feel is that SAP's cloud revenues and pretty much all of its projected growth in cloud is coming not from applications that customers implement and run themselves, but from ready-to-run applications that SAP implements and runs as a shared resource for its customers. Yet there is little sign of that model making any headway with customers for its core application portfolio. Indeed, SAP actively discourages it, preferring to push a model in which each customer owns their own separate, dedicated instance.
It seems like SAP is telling its customers to do one thing while it does something else entirely.
Building on networks
Yet I still see light at the end of the tunnel for SAP because of one really powerful set of assets that the company is starting to realize it owns. Primarily in Ariba, but also in other fields such as the SAP Financial Services Network, the company owns and operates highly entrenched digital networks. In a digitally connected world, these are hugely valuable assets.
As Chakib Bouhdary, SAP president business networks, put it in Tuesday's briefing:
In the 21st century, we're going to work into a digital world where we're going to see more digital commerce, more global commerce and what I call the network-enabled ERP of the future.
No business can be successful without collaboration with many other businesses that serve their customers.
The Ariba network can quote some compelling numbers: 1.5 million trading partners with sixty thousand new businesses joining every quarter and more than 100 million invoices processed in the past year.
The potential to profit from digitally unbundling hidebound paper-based enterprise procurement processes is huge. Bouhdary quoted as an example an ROI of $80-100 million achieved at ExxonMobil from adopting digital processes:
Most of the investments in this space, the ROI is in months ... Once you start to get the benefit of it, this is a very sticky business model.
Is it churlish of me to point out that these assets are powerful precisely because they are shared, networked resources rather than operated for the benefit of a single enterprise in isolation? There is a truth hiding within all of this but I don't feel that talking about porting existing ERP systems to HANA Enterprise Cloud does very much to elucidate it.
SAP needs to sort out its messaging muddles. Promoting HANA as a cloud platform obscures the main advantages of adopting the technology. Especially when most of SAP's cloud business continues to flow from its cloud application portfolio and its business network assets.
Disclosure: SAP is a diginomic premier partner.
Image credit: Ark © www.photostock.am - Fotolia.com