Shortcuts


The increasing relevance of predictive analytics

With technology advances outpacing most people’s understanding and even needs, identifying which solutions should form part of your company’s future can be a minefield

The increasing relevance of predictive analytics

With technology advances outpacing most people’s understanding and even needs, identifying which solutions should form part of your company’s future can be a minefield. In reality, the key to the majority of the trends and developments is data. Now that it is possible to capture almost infinite amounts of data about people, mining the data in helpful ways can generate insight that, if used well, can lead to beneficial change. For marketers, the ability to use data to predict how customers might behave in future scenarios has the power to deliver significant improvements to marketing campaign success. But it’s not necessarily as straightforward as it might sound.

Achieving success with data in this way is dependent on understanding predictive analytics and its role in making businesses more customer-centric and successful. Tangible success stories are leading many organisations to place it at the centre of their future strategies. Predictive analytics is the use of data, statistical algorithms and machine learning techniques to identify the likelihood of future outcomes based on historical data [SAS]. 

Predictive analytics fuels 35 percent of the purchases consumers make on Amazon.com, according to research firm McKinsey & Company. It has the power to transform marketing from mass communication to many, with a slim success rate, to personalised, targeted communication delivering the right message, to the right person, at the right time and in the right format! Maximising marketing ROI is often the name of the game.

The principles of predictive analytics

First, it’s necessary to build predictions of the needs and preferences of your customers. This can be done with historical data, business intelligence and even some reasonable assumptions and is no longer the preserve of statisticians. Modern systems assist the prediction modelling process and are easy to use. However, the real magic happens when you deliver campaigns built on these predictions and feed the actual results back into the system. Next time, the predictions will be more accurate, thus the results will improve and hence you have an infinite cycle of progress. What’s clever about modern technologies is that a lot of this can happen automatically thanks to machine learning.

As consumers continue to be bombarded with marketing messages left, right, centre, up, down (you get the picture), those brands that offer relevant, personalised messages are much more likely to be noticed. Data-driven campaigns will by definition provide a target customer with something they might want or need at an appropriate time, improving its chance of success. And even if it doesn’t result directly in a sale, the relevance will make it more likely to result in some kind of interaction or feedback, giving you a greater understanding of that customer for future campaigns. Customer perceptions of an organisation can also be greatly enhanced. Rather than leaving them irritated or feeling bombarded with irrelevant offers, they will consciously or subconsciously appreciate it when an organisation only contacts them with something really suited to their needs. Their perception of your brand is therefore more likely to be positive.

Better business intelligence

However, predictive analytics is not just useful for marketing. Building predictive modelling into other aspects of business management enables leaders to make highly objective decisions based on facts and data, rather than gut feel or assumptions. A good example is when making decisions about pricing. Predictive modelling lets you see the implications of different prices in different scenarios on revenues and can help an organisation to be highly responsive to market demand changes, thus maximising returns.

In sectors such as banking or telecommunications, predictive analytics can also be employed to detect fraud. Data mining algorithms based on predictive models can identify potential security breaches, thus triggering alarms or service halts to protect customers, as well as save the organisation from losing out.

Success stories

John Muir Health began using healthcare predictive analytics in 2016 to monitor sepsis. They saw a significant drop in sepsis almost immediately once clinicians began receiving alerts to inform them, for example, that based on a patient's information aggregated from various systems, the patient was at risk of sepsis and that a clinician needed to take immediate action.

In telecommunications, churn is a key challenge. In this sector, predictive analytics is commonly used to measure the ‘churn propensity’ of a customer to then deliver suitable incentives to encourage them to stay, based on how profitable they are.

At Consulteer, we were able to assist one of our customers in improving the response rates of their marketing campaign by three times compared to their traditional approach. We implemented a realtime platform that analyses the customer’s behaviour and suggest the marketer the best multidimensional segment for a given campaign. The platform learns from previously run campaign and thus continuously improves the outcome.

It’s not all plain sailing

The many potential benefits of predictive analytics mean it’s sure to become a key tool for business success in the future. However, the road to widespread use and acceptance has a few bends up ahead. While employing predictive analytics does not mean employing statisticians anymore, a lack of understanding and resistance to change is sure to hamper its adoption. Many modern organisations are also not yet structured optimally to take advantage of its many benefits. Data and departmental silos are not being broken down as quickly as technologies are advancing, so availability and sharing of data will continue to be a barrier for some.

Something else that is yet to catch up with the technological advances is regulation. Increasing concern about privacy and a virtual ‘Wild West’ means that no-one knows what the future of online data capture and use will look like. It might be that buying habits or online search habits might become classed as personal information that organisations are restricted from using. A more pressing concern is the security of data, with many high profile leaks leading to nervousness among some organisations about moving towards data-driven business models and among consumers, who are becoming more cautious about what information they share.

Predicting the future

Like with many emerging technologies, predictive analytics has vast potential across broad sectors. Its appeal to marketers in predicting the success or ROI of campaigns is clear, but as mentioned earlier, it has the potential to provide specific benefits to consumers and organisations in healthcare and telecommunications, among other industries, including real estate. By predicting, for example, a potential increase in house prices in a specific location or the likelihood of homeowners to be thinking about moving, estate agents can personalise their communications to beat the competition to the highly coveted prize.

So while there are clear hurdles to overcome, I can confidently predict that the future will be even easier to predict!

At Consulteer, we specialise in Digital Transformation and would love to hear about what stage of that epic journey you are at, to see how we could help guide you along.

Stay ahead of the crowd

Subscribe to our monthly newsletter with insights to the latest trends in Digital Transformation


To the top
Close