The Data Analytics Blog

Our news and views relating to Data Analytics, Big Data, Machine Learning, and the world of Credit.

All Posts

#PostTruth – What Does It Mean In The World Of Data Science?

January 9, 2017 at 4:19 PM

Silouhetter of Pinocchio and the definition of Post-truthIf I was to sum up our purpose at Principa, it would be “to help clients make informed decisions using data, analytics and software”.  As information grows, so the opportunity to make better decisions increases.  Data helps you understand your customer better.  That’s our mantra.  That’s our ethos.  That’s why we are.

Our long march towards “better decisions” through the clever use of data could not have been juxtaposed better (worse?) last year with the news that post-truth was the Oxford English Dictionary choice for Word of the Year.  This should be no surprise to anyone who at least offered a fleeting glance to two of the chief stories of 2016 – the Brexit vote and the US Presidential Election.

It appeared suddenly that data was irrelevant in political decisions - just look at the Washington Post’s extensive list of fabrications made by Donald Trump during the election campaign or Michael Gove (a key evangelist for Brexit) proclaiming that Britain had had "enough of experts".

Is evidence now passé? Has our existence as data scientists been scuppered by two 2016 elections in the US and Britain?  Do we at Principa now suddenly have an existential crisis?!

Not New

Actually this idea of post truth is not a new concept.  If you’re interested and have some time it’s worth reading up on 17th Century philosopher Sir Francis Bacon’s “Idols the Mind” (a decent summary is available here where he addresses why we are disposed to such thinking).

More recently, though, “post-truth” is the odour from the rot that is the post-modernist movement that has been with us for a few decades now.  A significant part of the postmodernism movement is a rejection of objective reality and rationalism often claiming this to be a product of sociological and political interpretation.

Some of you may remember the incident in the video snippet below from the University of Cape Town #Fallist Movement’s call for #ScienceMustFall in October 2016.

Postmodernism advocates for a more pluralist approach and sees science as just another narrative. The arguable cultural partialities in certain scientific narratives notwithstanding, science is an unbiased, self-correcting and valuable process that can be applied to help understand the world around us.

Science has been battling postmodern thinking for a while.  So where does that leave us data scientists?

Data Science

Despite the shock and horror of Post-truth 2016, I don’t think we as data scientists have a crisis at all, but rather a number of key challenges that we should address and continue to address.

  1. The value of what we do is typically quantifiable. Do you believe opinion or gut trumps the data?  You can test it to determine so.  Will a new model provide the lift that the analyst suggests?  Run a randomised champion/challenger test with pre-defined metrics to measure over a predetermined period of time.
  2. For many, just like the British and American electorate, consuming the flood of new information (data) feels like trying to drink from a fire hydrant. It’s easy to give up, but adopting appropriate techniques and technology in a well-planned methodical manner, will bring value to your business. Principa advocate the 3D approach
  3. If you’re utilising data to make decisions or looking for value in data, conduct your analysis or run your models always with key scientific principals in mind (for example: randomisation, statistical significance, champion/challenger testing, cherry picking, confidence in your data, sampling, confidence intervals, boot-strapping, extrapolation, inferences, etc.)
  4. The last point is education of the cynics and the uninformed. We need to be evangelists for our trade and honest proprietors too.  Information is the new wealth, but only if translated into wisdom and data science enables us to do this.


As the world navigates itself through fake news, echo chambers and post-truth, we data scientists will continue to work to harness the value in the data out there. Unlike politics, fortunately the decisions we make in business are not popularity contests. Even if a debate around the board-room becomes political, we will still have the ability to test our models and hypotheses through sound data science. Embrace it!

Truthseeker - logical fallacies

Thomas Maydon
Thomas Maydon
Thomas Maydon is the Head of Credit Solutions at Principa. With over 17 years of experience in the Southern African, West African and Middle Eastern retail credit markets, Tom has primarily been involved in consulting, analytics, credit bureau and predictive modelling services. He has experience in all aspects of the credit life cycle (in multiple industries) including intelligent prospecting, originations, strategy simulation, affordability analysis, behavioural modelling, pricing analysis, collections processes, and provisions (including Basel II) and profitability calculations.

Latest Posts

Shift happens: Top tips on Scorecard re-alignments

Principa employs a variety of best-practice credit scorecard building techniques including mathematical programming, regression modelling, optimal segmentation-seek genetic algorithms and reject inference parceling, amongst others. Through our credit risk scorecards businesses can look to improving their credit risk decisioning by 5-30%.

The Retaining and Reviving of Customers

What is the true value of losing a customer?

How Quick-Step Machine-Learning Models will help you through COVID-19

In a previous blog, we looked at assessing your credit models and the challenge of building and deploying models representative of the COVID-19 crisis. At the crux of the challenge was that the fact that: