To say the pharmaceutical industry is overloaded with data is quite an understatement. It typically takes 12 to 13 years of research and development for pharmaceutical companies to bring a new product to market at a cost of more than $1.5 million. That is going to leave quite some paper trail.
According to a Thomson Reuters whitepaper Big Data and the Needs of the Pharma Industry, the biggest opportunity for big data in the pharmaceutical industry lies in this area of drug discovery and development (41%), followed by understanding the market (26.5%). The key is to find a way of chopping up that Big Data into the small morsels we humans can easily digest.
But you have to the right ingredients to whip up a well-balanced data meal, covering all the different food groups. That involves integrating (and cleansing) external database content (45.5%), internal experimental content (27%), external social media content (15%) and internal document content (12%), points out the Thomson Reuters whitepaper.
That all sounds fine on paper, but in reality it’s a tall order for many pharmaceutical firms where departmental silos and legacy IT infrastructure means data is not open to all. While silos can have their place, pharmaceutical firms also need to create effective communication between those data sets.
Silos or not, it’s what you do with those data ingredients that’s important. Quantity is all very well, but you need to be able to find quality information and that requires analytics, and the best information comes not from looking at historical data, but by embracing real-time data and future forecasting.
Predictive analytics opens up the possibility for far more timely data for pharmaceutical firms. A McKinsey Global Institute report points out that predictive modelling of diverse molecular and clinical data could identify potential molecules that could be successfully used to develop drugs that can act on biological targets safely. It could also break down the dams between data silos enabling data to flow freely between different departments - linking together discovery and clinical development with outside partners such as contract research organisations (CROs). It’s only with this free-flow of data creating real-time and predictive analytics that real business value will be realised.
The biggest advantage of predictive analytics, however, is providing the opportunity to identify and ditch failing research projects earlier in the process, as well as spot the likely winners. It also opens up the possibility of reducing that 13 year product development cycle by three to five years, estimates McKinsey.
Yet, McKinsey also highlights the fact that there are as yet few examples of big pharmaceuticals doing this in practice and organisations are reluctant to plough big investment into big-data analytics as yet. According to Gartner, in 2012, only 13% of users were making extensive use of predictive analytics.
The benefits are clear, but what do you need to consider before you start down the predictive analytics path?
There’s no point embarking on predictive analytics without top-level backing. So, put some numbers together, work out the cost of making the wrong decision for a particular problem and potential revenue from getting it right. You also need to think about the kind of decisions you want to make and the parts of the business where predictive analytics could have the greatest effect.
Finding the experts
Choosing a tool is the relatively easy part of the whole process. What’s more difficult is finding experienced staff, who not only make the right product choice in the first place, but can adapt it to solve their specific needs. McKinsey in 2011 estimated that the demand for talented analytical staff could rise by 50 to 60%. 2018. You can avoid the fight for that talent by developing staff with the skills in-house.
Include the business
It’s also important to be able to present this information at the right time and in the right format to the right decision makers within the company. Your analytics team must not be shut away in an ivory tower concocting brilliant analytics that no one will use. The relevant business or commercial managers need to be part of the analytics team from the start, to ensure that the analytics created are on track with business objectives. That will also mean that the analytics are presented in a format that the decision makers can easily understand and use.