Financial forecasting is moving to the cloud, says Oracle

Profile picture for user pwainewright By Phil Wainewright February 17, 2015
Most new financial forecasting and planning systems will be cloud-based within two years, says Oracle's business analytics guru

Rich Clayton, Oracle business analytics, speaking in London
Rich Clayton, Oracle

Oracle has seen an unexpected surge in adoption of its cloud analytics products for finance in the past six months, according to Rich Clayton, VP of the business analytics product group. Interviewed during an event in London last week, he told me the vendor has already seen hundreds of customers sign up for its planning and forecasting cloud service since it became more widely available last quarter (including a whopping 76 from the higher education sector alone).

The rapid adoption suggests this category of software will be largely cloud-based within a very short timescale, he added.

We think in the next two years, for planning and forecasting, it will be predominantly in the cloud — 60 to 80 percent in two years. Today it's 5 percent. The core financials will be on-premise, but the forecasting and planning will be cloud.

The paradox of that is, the group that you would think would be the most conservative actually said, 'We're ready to go.'

Finance teams are being driven to adopt the cloud service simply because they can't afford to wait for this kind of management information any more, he explained. The need to keep pace with changing technology and the expectations of users today to have information on demand has ruled out conventional, on-premise IT as too slow and inflexible.

People want it on a mobile device, requirements continue to change, they've got to act faster. Standing up servers and configuring hardware is not a value-add.

The need for speed feels to me like the guiding principle. Almost to a fault, because what I see is that the pendulum has swung to the speed thing over integration. [Customers say] I don't care if it's integrated, I just want it right now.

Digital disruption

In his keynote presentation earlier that morning, Clayton had warned his audience that most of them would find the cloud essential simply to allow them to keep pace with the speed of product innovation.

We will be putting releases out probably every month. Our innovation cycle and your ability to consume our innovation cycle are diverging rapidly.

He cited food manufacturer Heinz as an early adopter of Oracle's cloud-based business planning product. It had rolled it out to more than 600 users within a four-month timespan and achieved a fivefold cost savings.

That is unheard of if you tried to do that in an on-premise environment.

Speaking to Oracle's current theme of 'digital disruption', Clayton also promoted mobile and of course big data and analytics as tools that enterprises must seize to be able to remain competitive in the face of a new generation of digital upstarts such as Uber, Airbnb and Amazon.

He said enterprises have to adopt new "methods, machines and metrics" to make use of an expanding universe of new data sources. But at the same time he acknowledged that Oracle's customers already have existing data stores and encouraged them to find ways to combine the two sets of data.

The value of information rises when we combine it with other sources of information. But the thinking right now is, 'Let's keep these two islands separated.' We believe you have to have a strategy for joining these two together.

Joining two worlds

This was a point I picked up later in our interview. How can Oracle customers bridge their existing transactional databases and data warehouses with the new breed of non-relational big data stores, I wondered?

Because we have the largest share of the relational database market it's assumed that we think that the entire information management platform is relational moving forward. When in fact, we have NoSQL databases, we have multi-dimensional databases, we have in-memory databases, we have a redistribution of Cloudera for big data and Hadoop, and so on.

Our perspective is, there is not one data management strategy in terms of storage of information, but there needs to be one architecture that joins these two worlds together.

Companies today, by and large, have broad understanding of SQL. What they don't have is people that know Spark and Pig and Hive and Cassandra and [the rest]

So our strategy is ... to put SQL on top of that. So that one analyst in a company could query and access data in Hadoop or NoSQL as if it were a table in an Oracle database.

My take

The NoSQL generation of big data vendors like to cast Oracle as the platform they are reacting against, which must make it difficult for the vendor to come up with a message that shows its willingness to play in both camps.

It has to tread a fine line between arguing for adoption of next-generation platforms such as cloud and mobile (which enables it to sell upgrades and new products) at the same time as making sure its customers don't decide to omit Oracle from those future platform choices.

That strategy seems to be working well in the case of its budgeting analytics offering. Oracle will be hoping it can repeat the success in other product areas.

Disclosure: Oracle is a diginomica premier partner.

Image credit: Oracle.