Advanced Analytics, Customer Churn and the Appliance of Science

Advanced Analytics, Customer Churn and the Appliance of Science

In 2010 Eric Schmidt (former CEO of Google) said “Every two days now we create as much information as we did from the dawn of civilization up until 2003." That’s something like five exabytes of data. According to IBM, the build out of the “internet of things” will lead to the doubling of knowledge every 12 hours. Let that sink in for a moment.

We take the digital era for granted these days, we’ve normalised its existence but when you step back and think about its impact, it’s as remarkable as it is overwhelming.

With the collective knowledge of the entire history of civilisation available for dissection, human behaviour has been documented in its entirely.

We’ll leave the philosophical ramifications of all of this to others - this is a B2B article on advanced analytics after all, but it’s worth taking in the bigger picture of just how much data is out there.

If we leave aside the focus on big data and the internet of things and apply advanced analytics on just a tiny speck of this information - your customer database - the insights gleaned from their behaviour will be decisive in the future success or failure of your company.

Customer Intimacy

To start with, let’s get the most obvious learning out of the way - retained customers are way more valuable than new ones, due to the costs of acquiring new customers. Adobe once found that it takes seven new shoppers to equal the revenue of a single repeat customer.

So if your focus is on retention campaigns, then your focus needs to be on your existing customer base. The development of programs to improve customer experience has been a direct result of this understanding. By delivering an exceptional experience, customers will not defect – or so the theory goes. But despite throwing millions of euros at “experiences”, customers continue to defect. If anything, they leave even more quickly and easily than ever before.

So what has gone wrong?

Net Promoter Score

Customer experience is a nebulous concept, but there has to be a way to assess its success. And so the famous “net promoter score” (NPS) was born. For a while marketers felt they had a good way of understanding satisfaction levels by simply asking customers what they thought.

Surveys were sacrosanct.

But there is a problem with surveys and the NPS regarding churn prediction – what customers say and do are two different things. According to a report published in Bloomberg Businessweek 60% of defecting customers describe themselves as 'very satisfied' just before they leave.

To make matters worse, the evidence of their impending defection has always been available – if you know where to look.

The Appliance of Science - Applied Analytics

Your existing customer database is a veritable goldmine of data for analysing customer behaviours. Every interaction between brand and consumer creates a digital footprint, an indication of intent – if you know how to read them.

Applied analytics provide a way to spot trends and patterns based on past behaviours. By classifying and categorising customers based on commonalities, you can drill down into those behaviours and better understand customers as individuals.

By following the behavioural trail you can identify indicators of intent. A customer may not say they are leaving, but their behaviour provides clues about what they are thinking. Has there been an increase in calls to customer support? A use of increasingly negative language in their emails? A reduction in their use of your service? All the behavioural indicators are there in plain sight - but only if you know what to look for and how to analyse it.

Taking these indicators and comparing them to the behaviours of other customers, you can predict their next move.

And here is the thing - you can identify, understand, and predict behaviour right down to the individual.

You can uncover how any one customer feels about your service and your offering and confidently predict how likely they are to leave, when are they likely to leave, why are they likely to leave, and what offer will make them happy to stay.

Act Early, Reduce Costs

With refinement your analytics will begin to identify these behaviours much more quickly, allowing you to act earlier. The sooner you act, the easier it is to recover the relationship – and the cheaper the incentive you need to offer. Your analytics will even reveal which retention incentives have had the greatest success for similar customers previously, further increasing your chances of a positive outcome for both parties.

Instead of issuing surveys that can be ignored, or which capture inaccurate sentiment data, analytics use the actual behaviours of your existing customers to make extremely accurate inferences and predictions. Statistical patterns provide actionable insights in a way that the nebulous NPS scoring system cannot, which means that your attempts to improve customer experience will always be more effective because you better understand each customer as an individual.

Fads come and go, but predictive behaviour modelling is just that...predictable. All the answers are there, but very few have the expertise or the tools to spot them, track them, report on them and recommend actions.

CLICK HERE TO DOWNLOAD OUR EBOOK

Will my car pass the NCT?

Will my car pass the NCT?

EDIT 7/8/18: Our NCT work featured in the Sunday Independent: https://www.independent.ie/life/motoring/car-reviews/which-car-is-best-of-the-test-37185356.html 

In Ireland, every car over 4 years old requires a roadworthiness certificate, which it gets by passing the National Car Test (NCT). If you're buying a used car, it's important to know how likely that make and model is to pass the NCT - and if it fails, on what part of the test.

To help you find out this information, Idiro has analysed the results of the last 5 years' NCT tests - and we offer you two tools:

The NCT checker

Idiro has created a simple NCT car checker tool, available online. You’ll find it at www.Idiro.com/NCTchecker.

Just enter the make, model and year of the car in question to learn all about how these cars perform in the NCT. If your car is quite rare, like the Mazda MX-5, then we recommend that you select all the years of NCT tests. Otherwise, just leave 2017 ticked.

As you can see, Mazda MX-5s from the year 2000 have a 66.3% failure rate across 5 years of tests - just slightly worse than the average of 65.7% for all cars of that age. However, look at the detail - the MX-5 does much better than average for some elements (no failures for suspension!) but much worse for others (four times worse for emissions). That will help you know what to look for when buying a used car, and help you prepare your car to maximise its chance of passing the NCT.

Idiro's NCT checker - input form and results

Our handy NCT checker works on phones, tablets and PCs. Kudos to my colleague John Grant for building it.

Exploring all of this year's NCT results

For people who would like to dig deeper into the NCT results, we have produced an interactive dashboard as a demonstration of our data analytics skills.

CLICK HERE TO OPEN THE 2017 NCT DASHBOARD

Again, it uses data published by the Road Safety Authority covering 2017 NCT tests. Data has not been published on retests, so our dashboard covers the first NCT test that each car underwent in 2017.

For practical purposes, the data is filtered to show only the twenty most popular makes of vehicle tested in 2017, and for each of these, only models with at least 1000 tests in the year. This was necessary because to show every make and model of car tested would make the dashboard so complex as to be unusable. As a result, from the total 1.4 million tests carried out in 2016, 1.1 million tests are represented in this dashboard.

However, if you do want to look at makes and models of cars that are not shown in this analysis, you can download the full dataset from the RSA.

How to use this dashboard

Pro tip: To reset all your filters and return to the original screen, click your browser’s refresh button.

The RSA provides the ‘Year of Car’ of each car tested, which we understand to mean the year of first registration. You can filter by the age of cars using this slider. For example, if you want to see test results for all cars registered in 2010 or before, you simply drag the end buttons in the slider over to the desired year.

This is an interactive dashboard, so as you change one parameter, all of the graphs adjust to match your selection. For example, this bubble table shows the top 20 most popular makes of car with a 2010 or older registration. The bubble's size indicates how popular that make is, and its colour indicates the make's pass rate - from deep blue (high pass rate) to deep red (low pass rate).

Test volume and pass rate per make

In this example, we can see below that for vehicles registered in 2010 and older, Toyota is the most popular make and has a high pass rate.

Now let's click on 'Toyota'. As you can see, this changes all the charts in the dashboard - they now only show the details of Toyota models.

Toyota selected

Pro tip: To compare different car makes, hold down the CTRL key while you click on each make that you want to filter in the bubble chart “Overall Popularity & Passing % of Model”.

Now let's examine the different Toyota models. In the next graphic to the right, ‘First time pass rate by model’, you’ll see the pass rate of each Toyota model.

First time pass rate for Toyotas

In the table on the far right entitled ‘Most Popular Model & Age’, you’ll see each model in the Toyota range. The Prius is the Toyota with the highest pass rate, but it isn’t the most popular Toyota - as you can see, the Corolla is the most popular (as you will see, Corollas have been around since 1980).

As you scroll down, you can see the ‘First Time Failure By Year’ graph which shows the number of cars tested (blue bars) and failure rates (red line) for each year of registration. As you can see, younger cars are much more likely to pass the NCT. To look at failure rates over time in each category within the test, you can filter by category in the drop-down menu.

To the right you'll see the ‘First Time Failure By Category’ table, which shows the percentage of cars that fail each category within the NCT test. This image displays what caused Toyota cars to fail their NCT.

As you can see, the dashboard allows you to dig deep into the 2017 NCT test results. Here again is the link to the dashboard:

https://public.tableau.com/profile/idiro.analytics#!/vizhome/NCT2017Top20Makes/NCT2017-20MostPopularMakes

This dashboard works best on PCs, rather than mobiles. Kudos to colleagues Paul Goldsberry and John Grant for building it.

 

We do hope you find these tools useful. To discuss how Idiro's analytics skills can help your business, drop us an email at info@idiro.com.  To download the source data from RSA.ie, click here.

 

Analysis of NCT test results helps car buyers choose wisely

Analysis of NCT test results helps car buyers choose wisely

 

Today Idiro has published a data visualisation dashboard (here) allowing you to explore the 2016 NCT test results. 

Update 21/8/17: The Sunday Independent ran a story yesterday on Idiro's dashboard - see here http://www.independent.ie/life/motoring/car-reviews/put-your-car-to-the-test-36049372.html 

Update 30/8/17: Our dashboard has been picked up by many other media including RTÉ: https://www.rte.ie/lifestyle/living/2017/0830/900977-is-this-the-worst-car-model-for-nct/

About this dashboard and the data

This interactive report is published by Idiro Analytics as a demonstration of our data visualisation capability. The data used is published by RSA.ie and covers NCT tests conducted in 2016. The data covers the first NCT test that each car underwent in 2016 - data has not been published on retests for cars that fail.

The data is filtered to show only the twenty most popular makes of vehicle tested in 2016, and for each of these, only models with at least 1000 tests carried out. This was necessary because to show every make and model of car tested would make the dashboard unusable. As a result, from the total 1.4 million test carried out in 2016, 1.1 million tests are represented in this dashboard.

If you wish to examine the data further or want to look at makes / models of cars that are not shown in this analysis, the full dataset is available for download at RSA.ie.

How to use this dashboard

Pro tip: To reset all your filters and return to the original screen, click your browser’s refresh button.

Pro tip: To compare different car brands, press CTRL + click on the type of makes you want to filter in the bubble chart “Overall Popularity & Passing % of Model”. 

 

The RSA provides the ‘Year of Car’ of each car tested.  You can filter by the age of cars using this slider. 

If you want to see cars from 2010 or before that were tested in 2016, you simply drag the end buttons in the slider (at the top) over to the desired year.

As the year changes, so do the interactive maps. This bubble table shows the top 20 most popular car models with a 2010 or older registration tested in 2016 that passed first time.

The bubbles indicate how popular that model is, and the colour of each circle indicates the pass rate - from deep blue (high pass rate) to deep red (low pass rate).

For example, we can see below that for vehicles registered in 2010 and older. Toyota is the most popular and has a high pass rate in 2016.

In the next graphic to the right, ‘First time pass rate by model’, you’ll see the different Toyota models that were tested.

In the next table ‘Model Popular Model & Age’, you’ll see the most popular models in the Toyota age range of the models tested.

Although the Prius has the highest pass rate in the Toyota model, it isn’t the most popular make, Corolla is the most popular.  As you scroll down the interactive map, you can see the failure rates of each car that was tested in the NCT and what the cause of the failure was.

 

You can filter what cars failed on by category or you can choose to see total which will show all the categories on the ‘First Time Failure By Year On’ drop-down menu.  The image below displays blue bars which indicate the units tested in each year and the red line graph shows the failure percentage rate. Remember that pass/fail thresholds can vary according to the age of the car.

As you scroll across, you’ll see the ‘First Time Failure By Category’ table. This table shows each category that the NCT test each car.

This graph displays what caused Toyota cars to fail their NCT. 

 

 

Here's the link to the dashboard: https://public.tableau.com/profile/idiro.analytics#!/vizhome/NCT2016Top20MakesResults/NCT2016-20MostPopularMakes

To contact Idiro about this blog post or about how Idiro's analytics can help your business, drop us an email at info@idiro.com.  To download the source data from RSA.ie, click here.