I went to the Athens National Museum this week, where I saw the famous "Hoard of Troy" antiquities. One of the Princesses of Troy was Cassandra, who may have worn the very earrings I was looking at, in 1200 BC. She was famously cursed by Apollo to be able to see the future but not to be able to do anything about it. She even foresaw the destruction of Troy and warned the Trojans about the Greeks hiding inside the Trojan Horse, but could not affect the outcome.
Implementing an Intelligent technology can be like this. You may have reams of data, great algorithms, and fantastic visualisations. But without the ability to actually effect the change in your business you will remain cursed like Cassandra. Seeing the future but not being able to do anything about it.
Here are some examples where you need both the prediction, and the ability to actually take advantage of the insight.
• You see from your Customer Experience (CX) analysis that a price segment has low elasticity and could take a price rise without diminishing volume, yielding a higher margin. But you cannot easily update the thousands of pricing records that store channel pricing without trawling through dozens of macro-ridden spreadsheets.
• You get a signal from the market that there will be an uptick in demand for you product in the upcoming Chinese New year, but you cannot shift your production to take advantage. It takes too long to find additional capacity, order the components and rebalance your workforce. The opportunity is lost.
• You are able to learn from sentiment analysis that a supplier should be black listed because it is close to bankruptcy due to a fraud scandal. But you cannot stop RFXs, order and payment going out to all their subsidiaries for several weeks. Not only leaking cash but also putting your own reputation at risk.
At SAP we call this the Insight-to-Action lag. For an Intelligent Enterprise to be effective, you need a Digital Core as well as the Machine Learning, AI and predictive analytics.
Don't be Cassandra: Cursed with ability to predict but the inability to effect change
By James Marland
Keywords: Big Data, Procurement, Predictive Analytics