It’s difficult to overstate the impact data science has made on marketing over the past few years. A variety of highly complex techniques and algorithms have added unprecedented precision to targeting, messaging and attribution. It has opened the door to new tools such as recommendation engines, predictive analytics and AI chatbots.
Marketing has never been so alive with possibilities for businesses and consumers. However, this brave new future is predicated on one important factor – do we actually know what we’re doing?
This is not a flippant statement. It covers a host of challenges inherent in the increasing complexity of data science-driven marketing. First, how can businesses be sure that they are using the right tools? Second, how do we know that we are using the right data in the right way? And, finally, how do consumers know how their data is being applied?
At the root of these issues is transparency and clarity. As data science was, and in many ways still is, an academic pursuit, the language used to explain how different techniques work is often beset with technical jargon.
This terminology can hinder how marketers and business leaders make decisions about what is best for their companies. For example, many businesses are keen to leverage the power of predictive analytics. After all, who wouldn’t want to predict the future? However, some businesses won’t have the right datasets or data infrastructure.
Others may misinterpret what the results actually tell them and then proceed to make poorly informed business decisions. Or to borrow a word from data science – they are “confounded”. Then there’s the final group that risk using the outputs from their predictive analytics to launch a marketing campaign that scares their customer base by revealing just how much they know about them.
This is just a snapshot of the obstacles that can emerge through data science being needlessly opaque. Yet, the problem can go deeper to include a lack of understanding about data itself – namely, how it can be ethically collected, stored and used?
Knowledge is often siloed due to a communication gap between data science, IT and marketing teams, and executives. Whether these issues are systemic – poor processes – or deliberate – linguistic gatekeeping – they can add up to serious problems.
Thankfully, there is a very simple solution: education and simplification. It is incumbent on all the players in data science and marketing to demystify their services for businesses. This can often mean data scientists having a much more commercial mindset.
First, every solution offered needs to not only consider the stated aims of a business, but also recognise their resources, brand identity and technical experience. Instead of looking purely at the data science techniques that will be needed, focus much more on the outputs and how the company could in practice use what they will learn – how will it impact their customers? Then, it’s about explaining in layman’s terms what will actually happen, everything from the processes involved to the limitations and likely benefits. Crucially, alternative options outside of a purely data science approach should be offered.
For companies selling data science services this may sound counterintuitive. However, a more consultative, transparent approach will make everyone happy. It enables the client to make an informed decision, unencumbered or misled by technical jargon, and ensures the company does not over-promise and under-deliver. In short, everyone knows what they are getting. This is why we find training and upskilling those working in-house such an important part of the work we do – it enables us to all get more out of it.
Of course, this is not a one-way street, it’s also up to executives and marketers to get educated and to also clearly translate their needs. They cannot rely on their supplier or in-house data science team to be the only source of knowledge. After all, even with the best of intentions, their goals will not always be completely aligned because they may not be privy to information held in the wider business; nor should you expect your data scientists to be marketing experts.
There is a wealth of resources available online – blogs, webinars, videos, events – that seek to demystify various data science approaches. In my experience, the simplest way is to ask. Data scientists often have a background in teaching, thanks to their post-graduate degrees, and their passion for the discipline means they are eager to share. Don’t be afraid to push them to explain concepts in simple terms.
The more you understand (and I promise it is easy to understand when explained well) the better able you are to be demanding and get the vital information you need to make better business decisions.