logo
07 November 2017

Big Data is in the eye of the beholder…

Everyone’s heard of the inkblot test, where an individual is presented with an ambiguous image and asked what it is they see. Typically what is seen varies from person to person depending on what element of the picture they see first and how the shapes relate to images they are aware of. They might see something positive or something negative, something promising or something threatening.

Everyone’s heard of the inkblot test, where an individual is presented with an ambiguous image and asked what it is they see. Typically what is seen varies from person to person depending on what element of the picture they see first and how the shapes relate to images they are aware of. They might see something positive or something negative, something promising or something threatening. 

A Positive View

We have a particularly positive way of viewing Big Data, unlike many who may see it as an ambiguous mass. Today we see the term mentioned in all different industries, but what exactly does it mean?

The term itself is quite self explanatory, and is used to refer to extremely large volumes of data that would require computational methods to analyse. This analysis would be used to reveal patterns and trends, which in turn can predict how things will evolve via modelling. The latter part of this is the most significant element, for any organisation wishing to use it to its advantage.

The beauty is what can be created as a result of various analysis techniques that can be applied to the dataset. This comes down to what a person sees in the data presented to them. The difference in our view between Big Data and the inkblot test is therefore that Big Data will definitely offer you a deep and true insight into your business, whereas the inkblot test’s inherently subjective nature has seen it fall out of favour in recent years.

Three Steps to Big Data

It’s not as simple as that though; there are three main factors that will lead to effective data analysis:

  1. How well do you know the subject matter of the data and how it’s reported?
  2. There may be elements of the data that would mean more to one person rather than another, depending on their experience. A further factor is how the data is laid out and in what order. It may be a bit like a word search, which if the layout becomes clear then the answer becomes more obvious. So really, a data set is only as useful as the person who is looking at it - it can become meaningless and redundant if not dealt with properly.

  3. Meaningful results and conclusions from the right team of analysts
  4. This is the next step, following the various specialist auditing techniques. Here you will end up with a suite of different reports and ideally you’ll employ experts in the appropriate fields, so they can pinpoint exact reports to run, in order to pull out all the relevant data to maximise findings and reduce time taken, making the whole process much more efficient. Projects we regularly engage in see us analysing in excess of 50m data items month. The best approach is to view this as 50m ingredients, to use or otherwise in your recipe to deliver the tastiest conclusions for your client.

  5. Presenting Effectively
  6. Finally the data needs to be represented in such a way that makes it simple, understandable and actionable for the relevant stakeholders. Reports and findings need to be tailored to the audience, otherwise the work can end up becoming wasted effort.

    The conclusion from this, and as per the title of the article, is that Big Data can only really be truly useful if handled and processed by the right people, who relish the opportunity and are confident it will benefit their client’s business.

Big Data Back in the Real World

We were recently approached by a client to analyse their telecommunications costs, being advised that this same piece of work had already been undertaken by another auditor, which had taken months, with no tangible outcome. We believed there would be opportunity for significant positives for the client, even though this had already been reviewed and so were pleased to accept the challenge.

We received the same data set that was presented to the previous company, which ran into the millions and millions of items. We immediately established a good ‘feel’ for the data, asked questions about it, and actually requested further elements that we knew would be available and be beneficial in the analysis process. At the end of the conversation our client was confident we understood the subject matter and therefore the context for the data set, such that a positive result was much more likely.

Our feel for the data meant that we immediately knew which areas to target and how to deal with it effectively. Indeed, despite months of analysis by another firm, within the first week of the project we were able to uncover a billing error worth hundreds of thousands of pounds to the client. The project is still ongoing, and numerous other findings have been presented, all with the aim of delivering a clean bill of health for our client’s telecommunications estate and the tools, techniques and confidence to identify and stop future errors in their tracks, in order to keep systems in peak condition.

 

 

PCMG are world-class experts in the fields of Telecoms, Energy, Water and Accounts Payable auditing and have been operating since 1993. 

We understand the data in these areas, are able to quickly identify target areas, and present findings that give successful end results to clients, which in the vast majority of cases will significantly benefit their bottom line.

If you would like to talk to us about how we can help you then contact us on enquiries@pcmg.co.uk.