Analytics, analysed

Back to All Posts

Capita Customer Management 16th October 2017 5 minute read

Analytics, analysed

In our October edition of Intelligence, Capita Analytics Director Neil Mackin shares six thoughts on where we stand in the battle to turn Big Data into big value.

  • Pay us if we’re right

 One of the most exciting trends I see is moving away from thinking about analytics as a technology towards seeing it as a capability. For example, let’s talk to the C-suite about your key challenges, which might be improving the efficiency of service, or increasing your win rate in acquiring new customers. So let me take your data into my world and do lots of stuff with it, then I’ll present the results back to you and if they work, you pay me. This conversation is essentially ‘we think we can use an analytics capability to do something for you; pay us if we’re right’.

Compare that to the traditional Big Data technology thinking: we need to start understanding Hadoop, data ingestion, ETL; then find ways of presenting what we’ve discovered so that you can make changes to act on it.

  • The data mindset

Analytics is not black and white. It’s more about reducing the risk that the right thing will happen – but sometimes it still won’t. You may be trying to change customer behaviour in a certain way, but at the same time someone or something else may be pushing it in another direction, and so your experiment fails.

Young companies like Airbnb know that. They are data driven from the top to the bottom. They have a senior management that know how to challenge the business and make things happen quickly.

  • Beyond the laboratory

There have been huge strides in our ability to capture then process and transform data – the costs for storage and analytical techniques have fallen in recent years. But what hasn’t advanced so quickly is the organisational management of the analytics team in a way that aligns projects to what the business wants, then takes the output and makes people accountable for realising the value.

If executives say, “we need a model that helps us identify what we should say when a customer rings me up”, then I think once you’ve built that model there should be a degree of accountability. Their peers should hold those executives to account to actually put it to use and gain value from it. If you don’t do that, everything stays as a little idea in a laboratory, never becomes operational, and you never see the value of the data gathered.

  • Visualisation

In a way I think it’s disappointing that we still present what we’ve discovered as shapes and colours on a 2D plane. Presentation hasn’t changed much in the last few years. I can’t help thinking there must be better ways of doing this, perhaps combining data and video to go beyond the simple animation of a chart to become a landscape of data that I can explore and walk around in. For example, look at how Minecraft has helped young people who play it understand and immerse themselves in complicated 3D dimensions. Now imagine walking through your data in that sort of world.

  • Will the growth of data shift the customer relationship?

Another interesting example of how data could come together is the likes of Alexa on the Amazon Echo. If I’m speaking to Alexa in my kitchen to sort out something on my holiday, what if it could say, well, if you’re going to be away for two weeks, why don’t I sort out a better deal on your car insurance? At that point it starts to become a really useful life assistant.

If Alexa is my intermediary, acting as my concierge and dealing with all those companies for me, that means customer service is radically transformed. Companies won’t be talking to customers any more – they’ll be talking to the concierge – so how does that affect their relationship with you and me?

  • Where could Big Data go next?

Here are some of my suggestions:

  • Analytics as a Service, where you pay by outcomes. Ignore all the technology, just give me the data, I’ll find a way of making it create value for you, and we’ll share that value together.
  • Exploiting the computational power that is becoming available. For example, there are new deep learning approaches behind many of the advances we’re seeing in image and voice recognition. Behind deep learning is the idea of harnessing the computational power within GPU chips. They can carry out huge quantities of numerical computations in parallel, and therefore can be used in machine learning.
  • The Quantified Self, which involves bringing together all the ‘me’ data that I generate during the day and giving me some rich contextual feedback. ‘Neil, do you think you’re getting stressed? Do you think you need more exercise?’ A digital coach to help me be the best version of myself. But I want real control of how it can be used, and definitely control of whether my data is shared with anyone else.

To read the full interview, as well as a recap of our Innovation Breakfast event on the 26th September, read the October 2017 edition of Intelligence on our Intelligence report hub here.


Why fixing a customer’s complaint really is the least you can do

by Tracey Roberts 26/07/2019

5 minute read

Read the article

Why being alert to emotional context can boost customer satisfaction and benefit the bottom line

by Kerry Edwards 25/07/2019

5 minute read

Read the article

Keeping your promises? Here’s what your customers think

by Jo Causon 22/07/2019

5 minute read

Read the article

Contact us

News & blog

Read the news & blog

Capita PLC Head Office: The Copyright Building, 30 Berners Street, London, W1T 3LR

Registered in England No. 01336850. Part of Capita plc.
© Copyright Capita PLC 2019