4 ways that cloud analytics disrupt traditional decision support

ampelosaurus big size small intelligenceRecent research has revealed that huge Sauropods such as the Ampelosaurus had puny brains. The brain, shown in purple in the picture, was tennis-ball sized while the beast itself could be 15 meters from snout to tail. I was reminded of one of my early engagements in my consulting career. I was asked to audit the reporting practices for a bank’s real estate management division. It was not a flattering review. Every manager who had come through the department over the years had tweaked the reports to his/her style and added to the compendium. There was no oversight on relevance, usability and currency. What’s more, paper copies were printed and circulated to the next layer of management on a regular basis. Nobody read them but the circulation was a matter of rote. In many respects the decision support function for that bank resembled the Ampelosaurus – a beast with a large mass and a tiny brain. That bank was not alone in this kind of bloat. Most companies register the need for decision support and react by building armies of analysts rather than intrinsic process intelligence. This problem is compounded with the arrival of Big Data. A 2011 report by McKinsey Company’s Business Technology Office stated that the need for data-analytics talent would exceed supply by 50 to 60 percent by 2018. Cloud analytics is a path out of this morass, especially for SMB’s that lack the formal processes for offshoring and/or in-house skills development. There are four essential ways that cloud analytics enhances (and possibly disrupts) the traditional services model.

The traditional decision support layer comprises business intelligence software and analysts who support the requirements of the business users. In this model the decision support layer scales proportionately to the business requirements, and despite the best intentions such a model risks going down the path of the Ampelosaur.

With cloud analytics the decision support model delivers a more sustainable program via

  • Integrated ETL processes: Time and effort not expended in manual data pulls;
  • Embedded analytics: The analytics is integrated into the business process so the business user does not need to solicit analytical support. e.g. Infernotions’ warranty process or the marketing spend allocation feature in the Polytab solution
  • Integrated (self-service) reporting: Reporting is on-demand and via a web based interface. This is the basic BI intelligence that is now delivered over the Internet.

DECISION SUPPORT MODELS

 

The four ways that cloud analytics is a more sustainable decision support model are

  1. Cloud analytics scales cost-effectively: Scaling the business needs 5x does not mean the decision support layer should scale up 5x as well with a proportionate increase in costs. Because cloud analytics relies on automation and the on-demand model, it is not dependent on the growth of the team size. As an example, when we added telematics to the Polytab rollout for a retail customer, the underlying data mart was transformed, but the number of analyst hours increased by 1/4 analyst hours. A conventional model would have required the addition of up to 3-10 analysts.
  2. Intellectual capital remains in-house and perpetuates without incremental effort: Knowledge of business, processes, operational strategies is critical intellectual capital that is best retained within the organization. Cloud analytics embeds this knowledge into the technology via the user interface. This is superior to training the decision support analysts whose attrition means the loss of intellectual capital as well as the incremental cost of training.
  3. Decision support is timely with less dead-time: At the very basic level, cloud analytics offers the capabilities of basic Business Intelligence tools delivered over the web. However, because the model resides on the web, there is less dead-time due to system upgrades or analysts’ hand-holding. Reports are delivered on-demand (like with Polytab) and intelligence is embedded into the user interface.

Free webinar - marketing intelligence automation for multi-channel retailer4. Cloud analytics delivers process intelligence: A marketing manager wants to allocate his/her spend for the next quarter to maximize the return on direct mail/email campaigns. A warranty manager wants to get an early warning on issues that are hiking up claim costs. These types of decision support should be embedded into the marketing and the warranty processes respectively. The traditional model would necessitate a query and report process to get the decision support. With cloud analytics, it is possible to embed and automate the analytics into the business process.

Further Reading: “Does On-Demand Business Intelligence Make Sense? David M. Raab, Raabssociates.com.

Leave a Reply

Your email address will not be published. Required fields are marked *