2018 has been a year of significant change in the data industry. Think back to the end of 2017: GDPR was incoming, and we’d never heard of Cambridge Analytica. With the introduction of GDPR, and huge data breaches from the likes of Facebook and Aadhaar, we are now more aware than ever of the value of our personal data. The risks are such that we’re now seeing the advent of ‘burner’ credit cards which are valid for only one transaction thus protecting the user from fraud.
The public has lost a great deal of their naivety concerning the treatment of data and these themes naturally extend to the world of data science, which was itself rocked this year by revelations of biased models. For many the stand-out case was Amazon’s recruitment algorithm, which was found to be prejudiced against women candidates. The reality, however, is that Amazon did their due diligence, found the issue and scrapped the model, whereas there will be many more algorithms in the industry that have been unintentionally trained on data containing bias and never challenged.
With companies waking up to the risks of the ‘super’ algorithm best-practice guidance is now emerging. Organisations like ORCAA are leading the way, offering certification and an algorithm audit service to verify the quality of models. We are also seeing the development of systems that bypass these concerns. The first stage of Unilever’s new recruitment process requires candidates to complete a series of games on their phone which assess their skills fit for the company. It’s this dataset, rather than their CVs, which is processed by the recruitment algorithm.
In contrast to the protections around personal data, public data has never been so open. Platforms such as Kaggle and the UK Government’s Open Data Portal facilitate the democratisation of data. Data journalism is flourishing in this environment. Driven by the improvements in visualisation tools and UX, journalists are now able to tell their stories to a wider audience than ever before and in greater detail.
We need only look to some of the winners in the Kantar Information is Beautiful awards to see powerful examples of data-led reporting. Both the National Geographic’s “What Happens to the Plastic We Throw Out?” and Thomson Reuters’ “Life in the Camps” are excellent examples of hard hitting journalism that present the complexity of macro and micro elements in a clear unified narrative.
If the increased visibility of data journalism has opened people’s eyes to the creative aspects of data analysis, then industry trends have confirmed data’s place at the heart of any successful business. As the dust settles after GDPR we see a public more empowered in their relationships with companies and with higher expectations of the services they receive.
The new year will undoubtedly bring us further technological advancements at the forefront of research and design but primarily 2019 is about consolidation. In this changing data landscape, we must ensure that strong foundations have been laid so that we can take advantage of the innovations of the future. This is our opportunity to prove our quality and pass on the benefits of dynamic reporting and machine learning to our data literate customers; ensure that they have the information they need, when they need it, and in an engaging format. The market is ripe for local innovations.
2018 may be nearly over, but it has paved the way for stronger businesses and in future years will surely be marked as a turning point in our relationship with data.