In the run up to ConTech 2018, industry leaders Sam Herbert and Vincent Cassidy explore some of the key issues for publishers looking to become more data-driven..
Director of Academic Markets, Institution of Engineering and Technology
Co-chair ConTech 2018
Co-Founder, 67 Bricks
Co-chair ConTech 2018
1. What is the difference between a data-driven publisher and a traditional publisher?
A data-driven publisher captures, creates, stores, manages and most importantly uses data to support business processes and product development. A traditional publisher is content driven and might use technology and data to make its products more accessible. But a data-driven publisher has data at the heart of every process they undertake. I think all publishers are, or should be, considering how to be more data-driven. Hopefully in time there will be no need to distinguish between the two.
A data-driven publisher puts a greater emphasis on the systematic gathering and management of data as a core business activity. They then understand, manage and use that new data as a core business asset.
2. Why is this an important evolution for publishers to make?
Thanks to apps, smartphones and the internet, end users experience high quality product and interaction from B2C suppliers on a day to day basis. This has fuelled high expectations about what information publishers can provide them with. Static content is no longer enough to keep them happy. Publishers have to transition to be data-driven to stay in business, and that transition is fuelled by data about your content, data about your customers and data about the market.
3. What benefits are there to the publisher in the short and longer term?
In the short term publishers can make cost savings and internal efficiencies by becoming more data-savvy and using AI and machine learning to speed up and automate time intensive jobs like categorising content and identifying peer reviewers.
In the long term, it is about having the right skills and capabilities to enable you to prepare for a data-driven future and new business models. This can give you commercial advantage and future-proof your business. For example, at the IET, we are using a data-centric approach to build an analytics solution alongside Inspec, our leading A&I service. Appreciating the value of our content as data is allowing us to build new products and services.
4. What benefits are there to customers or readers?
Content solutions will be more personalised and focused on specific user goals and will become richer, more intelligent information. For example, our client SAGE Publishing wanted to improve the quality of engagement between the sales team and their institutional customers in order to best serve their customers. Previously this involved manually creating reports containing lists of SAGE products that fit the organisation. This is a very time consuming process. But they now use a new tool that takes a text based description of an educational course and instantly returns a list of SAGE content relevant to the course.
5. What are the trends you are seeing across the wider publishing industry in this area?
Some publishers are tackling being data-driven head on with big transformation programmes; changing the whole focus of their business. This involves major decisions at board level and a considerable amount of work (and pain) to reach the point of change. But that approach is not suitable for everyone and we are seeing a lot of publishers dipping their toes in to improving how their content is used and trying spot solutions without thinking about broader evolution. That enables them to try and learn as they go and often works well; as long as the big picture is that data is the driver.
6. What steps should publishers who want to become more data-driven take?
Publishers need to recognise where they are on the path towards being a data-driven organisation and then plot a journey that both delivers in the short term and achieves bigger picture goals in the longer term.
Publishers need to think in terms of capabilities; both human skill sets and technology capabilities. Start with smaller projects that develop skills, deliver some success and put in place knowledge on which to make future decisions. It needs to be an iterative ongoing approach, but with a strong and bold direction.
7. Do publishers need to consider creating a taxonomy or ontology?
Not necessarily. Taxonomies and ontologies are useful in specific circumstances and can bring platformed data to life, but they are no longer always the right way to travel. Modern content processing techniques have developed sufficiently to replace the need for them in many cases.
8. We often hear the term semantic enrichment used, what does it mean and how does it support the goal of becoming a data-driven publisher?
Semantic enrichment is the application of modern content processing techniques like machine learning, AI and natural language processing to add structure, context and metadata to content. It can be a key component in creating the data rich view of your content that you need to become a data-driven publisher. By creating a ‘semantic fingerprint’ of your content items you can drive personalisation features like relatedness or personalised email notifications that deliver high spec, relevant content recommendations.
9. AI and machine learning are grabbing the headlines, how do they fit in to this discussion?
Publishers can use data about their content and users to drive machine learning algorithms that deliver internal efficiencies such as allowing them to spend less time on menial tasks and reducing workload. They can also use it to power new features for end users, such as making personal recommendations or using automated bots to quickly and intelligently respond to queries.
10. Are there any standout areas of best practice or innovation you would like to share?
Organisations like GE Digital are changing their entire mindset and reimagining themselves as a totally data-driven, company. The GE Digital group started on the journey by capturing data on jet engines to increase internal efficiencies and has now evolved into a data services company. This has leds to big changes and a massive competitive advantage.
11. Is there anything you think the publishing community can learn from or be inspired by from other industries?
Some key players in the legal industry are making large efficiencies by utilising new technologies and changing their mindset. FMCG has always been more customer knowledge focussed and aviation is hugely advanced in customer experience and satisfaction. There are definitely some lessons to be learned that can cross sectors.
12. If you could give publishers who want to become more data-driven one piece of advice what would it be?
Consider a future where everyone and everything is more connected, where algorithms underpin everything that happens and where data is the fuel that feeds those algorithms.
Set a course that puts your organisation on a solid footing in that future. Work up a data strategy before you think up a technology strategy.
13. Finally, if we were to have this conversation in 10 years’ time what do you think we would be discussing?
As we get closer to seeing an AI that matches human intelligence, I think we will be asking – do we want to replace humans in the workplace? Do we want a workforce largely run by AI?