Subscribe to the Teradata Blog

Get the latest industry news, technology trends, and data science insights each week.

I consent that Teradata Corporation, as provider of this website, may occasionally send me Teradata Marketing Communications emails with information regarding products, data analytics, and event and webinar invitations. I understand that I may unsubscribe at any time by following the unsubscribe link at the bottom of any email I receive.

Your privacy is important. Your personal information will be collected, stored, and processed in accordance with the Teradata Global Privacy Policy.

Move Fast – But Don’t Break Things

Move Fast – But Don’t Break Things
The past few years have seen the adoption of ‘agile processes’ in the retail sector and elsewhere. These approaches break processes down to small elements and then seek to ‘sprint’ to software solutions to solve them as fast as possible. This ‘DevOps’ approach to IT has unleashed innovations and brought IT closer to the business. But it has also introduced new risks.

AI and machine learning are moving from proof of concept to mainstream value drivers in retail as in other sectors. But the agile methodologies that were effective in creating prototype, departmental, or single-issue type projects may not be suitable for enterprise-scale data projects. Agile practices can deliver fast and compelling returns, but they can also lead to fragmentation, data silos, and unnecessary complexity. Senior executives need to enforce structure and common approaches to create the data-driven enterprises that they need to drive growth.

Break it and there will be consequences

Risk, and specifically data risk, is now one of the highest-level concerns of CEOs. Yet often the fragmented nature of data and analytics projects means that it is all but impossible to audit where data is, let alone assess risk.  Minor mistakes, oversights and failure to foresee consequences can have huge repercussions. The infamous data leak at BA was caused by a software error in a third-party’s system, but that did not protect the airline from a £183 million fine and significant reputational damage.  With multiple teams using multiple data sets, often containing sensitive and PII data, retailers are at high risk of similar consequences.

Fragmentation and complexity are the enemies

Many retailers will track detailed information about their customers in many channels. They will have clickstream data on visitors to their website, call answering, drops and resolutions in customer contact centres, and perhaps even analytics from consumer mobile apps. But most probably these are all separate data silos – and cannot provide a view of a ‘real’ customer. According to Salesforce, 27% of marketers feel this data is siloed and 40% duplicated.

In the digital economy, algorithms play an increasing role in automating responses and augmenting the work of humans in virtually every department. But they are only as good as the data they are trained on. To be effective at scale, to not only predict but organise to act and capitalise on intelligence in real time, small-scale analytics that provide limited snapshots of the ‘real’ customer must be eschewed. Instead, AI must learn from enterprise-wide data sets, and the insights must be available to the whole organisation, from shop floor to executive office to logistics and warehouses, at the same time.

Simplicity, governance, control

Simplification of the whole enterprise is a watch word for CEOs in many sectors, and retail is no exception. The explosion in volume, variety and velocity of data available to the business is complemented by the demand for fast, effective and straightforward systems and processes that simplify the chain from data to insight to action.

Closely linked to this is every greater scrutiny from customers, regulators and governments on the management of all aspects of the business, from security and provenance of supply chains, to financial transparency, to cyber defences.

The urge to create agile, low-cost analytics projects to prove specific business cases cuts directly against this need for simplification and clarity. Governance demands simplicity, and simplicity demands holistic, long-term planning. It also requires a mindset that sees data truly as an asset with a value and an anticipated return. Only senior management has the purview, the power and the vision to enforce this. To evolve into the retailer of the future, key decisions about how, where and why data is used need to be taken from the top-down.

Data is so vital to the success and the survival of today’s retailer that it is too important to leave to data scientists or marketers to manage. The successful retailer of the future will be dedicated to managing millions of ongoing, personalised conversations with customers, suppliers, staff and stakeholders in real-time. Small scale, fragmented, stop-gap solutions will simply not work.

To capture the real value of data, CEOs need to oversee it and manage it alongside other assets to maximise return. In the 21st century it is likely that data will quickly become the most valuable asset that any retailer has. Protecting it and leveraging it to deliver economic value at every point is fundamental to the CEO’s role.
Portrait of John Malpass

John Malpass

John is the Retail Industry Lead for Teradata EMEA. He has over 30 years’ experience in Retail across the food, fashion and general merchandise sectors. In that time he has worked with numerous retailers across the globe to deliver business value through the innovative use of data and analytics. 

John has held a variety of senior development and operational management roles where he has delivered successful business transformation programmes.  He has a passion for using data analytics to challenge traditional ways of working and deliver tangible value.

View all posts by John Malpass

Turn your complex data and analytics into answers with Teradata Vantage.

Contact us