Our Articles

A Monthly Article from our Speakers

Current Article of the month

Articles of 2020

Articles of 2019

Articles of 2018

Articles of 2017

Articles of 2016

Articles of 2015

Articles of 2014

Articles of 2013

Articles of 2012

Articles of 2011

Articles of 2010

Articles of 2009

Articles of 2008

Articles of 2007

Articles of 2006

Articles of 2005

Articles of 2004

Articles of 2003

Articles of 2002

Mike FergusonData, BI And Analytics – Lifeblood to Survival of the Fittest

by Mike Ferguson

January 2018

 

As we move into 2017 never has a title given to a book published ten years ago in March 2007 been more appropriate.

The book of course was “Competing on Analytics - The New Science of Winning” by Jeanne G. Harris and Thomas H. Davenport. Without doubt competing on analytics is now central to the business strategy of almost every company looking to survive in business in the medium to long term. From my own experience, I would say that single view of a customer and data privacy are probably now the top two priorities I am seeing in almost every vertical industry in the European Union (EU) in 2017. Privacy is dominated by the EU General Data Privacy Regulation (GDPR) which is now written into EU law and enforceable from May 2018 onwards in every member country. It’s not long away and so it’s not surprising that many companies are now reprioritising to get this compliance work done in time. GDPR will dominate the EU data governance scene in the next eighteen months. It is primarily associated with personally identifiable information (PII), which is primarily customer and employee related. Customer data in particular is the one area we all think about and that’s important because, as I already said, the other top priority area is single view of the customer.

The need to really understand each and every customer has become critical to business survival and for one very simple reason. They are now in control. The mobile device, the Internet, search engines, social networks and review websites have made customers and prospective customers all-powerful. They can easily find competitive products and services to what you are offering. They can also easily compare competitor offerings with yours and check out how others rate you. And they can do it all while on the move. It is so easy nowadays for customers to shop around and be well informed before they buy. It really doesn’t matter whether this is business-to-consumer or business-to-business. The power that customers have at their fingertips is enormous.

So if you are just relying on ‘good faith’ to retain your customers simply because they have bought from you in the past, then you are probably in for a rude awakening. The digital world offers no mercy. It has provided findability, convenience, speed and automation. Loyalty has become cheap and customers will churn if they find a better product or service elsewhere. Never has the fight to survive been so intense.

Companies therefore need to do everything in their power gain deeper understanding about their customers to compete. They need a single customer view of each and every customer. The customer expectation is now that you have a real-time, always up-to-date, common understanding of them across all channels and can offer each one of them personalised products and services. As an aside, it seems strange that here I am writing about the need for a single customer view when I can remember using the exact same expression twenty years ago when writing about building a data warehouse. So what has changed?

The first thing to realise is that creating a single customer view means going way beyond a customers’ historical transaction activity that might be available in a typical data warehouse. Today we want to know way more than that. We need to know customer needs, desires, intent, opinions, relationships with other people, with places and things. We also need to understand who they interact with in social networks, who the influencers are in each of their social networks, what their on-line behaviour is and how it differs from the next customer. We need to understand product usage, and how all this changes over time. We need to create much more comprehensive enriched customer profiles by capturing and analysing new data to derive new insights and integrate what we find with what we already know. Also each ‘enriched’ customer profile needs to be linked to all the marketing activities associated with that customer so that you can determine if your campaigns are successful in retaining, increasing the value of and growing your customer base.

You can’t do this with BI alone. Query and reporting is not enough. You need analytics to be able to segment your customer base sensibly, to predict churn and track changes in churn scores over time. You need to be able to predict purchases in order to promote specific brands and to make the right recommendations in real-time and in batch to improve customer value through cross sell and up sell. You need to keep track every customer interaction in all channels over time so that everyone and everything is kept customer aware on a continuous basis. You need text analytics to understand opinion. You need graph analysis to understand relationships and find influencers in social networks.

If companies are serious about doing this then internal and external data needs to be captured, cleaned and integrated. Customer data cannot be inconsistent. If it is inconsistent across all channels then how on earth can you expect to successfully cross-sell or upsell to a customer?

Single customer view is about capturing all kinds of data and leveraging the analytics best suited to derive the insight needed. We need machine learning, text analysis, graph analysis and clickstream analysis. We need fast data streaming in on a continuous basis and real-time analytics to monitor, predict and respond in an agile way. And we need to do this in a self-service world where governance matters and where no one violates GDPR.

We need agility. Also, with all this work to do, we need to free up people working in traditional data warehousing to move them on to new analytical workloads. It’s twenty-eight years on since the first paper on data warehousing by Devlin and Murphy in 1988. We know how to build data warehouse systems and so it is time we introduced more agility into them and automated data warehouse development. Data vault design, virtual data marts instead of physical ones and data warehouse automation are all now needed to improve agility. We need high quality consistent and trusted data in a self-service world. It’s not about a self-service free for all. Consistent data about products and services, customer privacy preferences, customer channel preferences, customer contact information, social relationships, personal and corporate household relationships are all needed if we are to improve effectiveness. We also need to enrich customer data by getting additional attributes from their social media personas and from open government web sites.

There is a lot to do to create a single customer view. It is way beyond data warehousing. It is not surprising that the thirst for new data to analyse has skyrocketed in just about every corner of the enterprise. Profitable revenue growth is now dependent on new data and analytics to optimise the effectiveness of marketing, sales and channel activities. We just have to channel this energy and enthusiasm into multiple customer-oriented analytical projects while staying in control of governance and complying with GDPR.

We need to increase investment in analytics to provide actionable customer insight available simultaneously across all channels and integrate analytics into BI tools and applications.