In-house vs outsourced – building an analytics function that hits the ground running

a man standing in front of a wall, looking at his options to outsource data analytics for his company
We live in a data-driven economy and failure to build a data analytics competence of some kind leaves you at a competitive disadvantage. And we know that businesses need to become much smarter about how they use data to retain or attract customers.

One of the first choices you face is how to build out your analytics function – do you want to build a team in-house, or partner with external experts, or even choose a hybrid model? This decision will have wide-ranging consequences for your ability to exploit your data in the future.

DIY data analytics

In theory, building your data analytics capability in-house has one major advantage – you can begin analysing your data almost immediately. Obviously, you will still need to deploy predictive analytics tools, but you can save time that would be otherwise spent identifying potential partners and agreeing service contracts.

But this course of action assumes you already have data science and analytics skills in-house. If not, you will need to hire suitably-skilled staff. And that’s where you start to run into delays and risks.

Paradoxically, you need data science experience in order to hire your first data scientist – otherwise you cannot properly evaluate their technical skills. It is also incredibly important to realise that you cannot simply bolt data science onto existing operations – you must change your culture to be able to act on the insights being generated by your data science team. Too often, businesses make this mistake and never realise the full potential of their investment.

Expensive skills shortages

Data science skills are in very short supply helping to drive salaries up. According to Payscale.com, the average annual salary for a senior data scientist is currently €70,318 – and rising. And you’ll need a broad range of skills that are rare to find in one person – according to research from McKinsey, “Best-practice companies rarely cherry-pick one or two specialist profiles to address isolated challenges. Instead, they build departments at scale from the start.”

Although you will realise significant benefits, building an in-house team to turn your company’s data into money will involve a substantial initial outlay.

Instead of making new hires, you could retrain existing staff. But this will greatly increase the time to get your advanced analytics program up and running – and longer still until you see returns on your data analytics investments.

Why you should consider outsourcing

Keeping analytics in-house creates a huge burden on time and resources – at least during the initial stages of building your data analytics capability. Over time they will deliver value, but many CFOs will baulk at the time it takes to generate a return on investment.

Partnering with an external provider offers a much quicker return on investment because the entire process is shortened. And because your partner already has a suite of pre-configured analytics tools, they can begin unlocking value from your data almost immediately.

Outsourcing can be a transitional process too. One way to get the best of both worlds is to outsource all of your predictive analytics functions initially while you build an in-house data science team. As those capabilities come on stream, you can then start bringing functions back in-house.

Using third party consultancy also helps you avoid the staffing issues inherent in trying to maintain operations in-house. Your business doesn’t have to attract suitably skilled data scientists, or deal with rapidly increasing salary demands.

Outsourcing can be implemented in different ways too. Hybrid outsourcing allows you to split responsibilities with your analytics partner for instance. Under this model you retain responsibility for some elements – for example, the underlying database infrastructure, while the outsourcer provides others, such as the hard-to-come by modelling and analytics functions. The hybrid model is fully flexible because no two scenarios or deployments are ever the same.

This allows you to maximise the use of your own staff resources and minimise outsourcing costs without limiting your analytics projects and obtain the skills you really need for data-driven operations.

Speed is everything

When it comes to improving the customer experience, speed is incredibly important. Giving people what they want, when they want it is a key aspect of all customer retention strategies.

As you roll out your data analytics program, speed needs to be a factor at every point – including before you even begin analysing data. The faster you can get your predictive analytics capability in place and generating insights, the quicker you can begin to realise a return on your investment. McKinsey even put a figure on operating profit improvements, suggesting “first movers” account for around 25% of the gain Why? Because they have more time to integrate analytics with workflows than their competitors.

In reality, if your business has never used predictive analytics tools before, choosing to implement data-driven strategies in-house could be a mistake. Any initial cost saving will be quickly cancelled out by the extended time it takes to begin generating actionable insights. Far better to outsource the work to the experts initially, and have your outsource partner train and gradually hand over hand over responsibility for analytics as it comes up to speed.

For more help and advice on finding the optimum mix between in-sourcing and outsourcing for your data analytics team, please get in touch.

Advanced analytics, customer churn and the appliance of science

A crowded street with people marked, indicating the importance of analytics and data
 
 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.

Speak to one of our analytics experts to see how you can use advanced analytics to improve your customer experience and reduce churn.

 

When it comes to buying houses, people in Dublin are clearly superstitious

Streets of Dublin on an article about Irish Housing Market

Who would have thought that in this day and age, the Irish people would still be suffering from this ancient affliction? The terrible problem of Triskaidekaphobia, or the fear of the number thirteen.


The Irish people, as a nation have achieved many great things, we’ve become one of the biggest technology capitals in Europe, we’ve produced some of the world’s greatest athletes and sports stars, we’ve lead the way in giving equal rights to every citizen, not to mention the musicians and actors that other countries would love to claim as their own, but we know they’re Irish in their hearts.


But alas, we still have trouble shaking the quaint “luck of the Irish” image that American tourists hope to see when they step foot in temple bar. The image of a superstitious nation who base decisions on old wives tales and mythology.
We may say to ourselves that this isn’t the case, that it’s just how the Irish people are portrayed on tea towels found in Carroll’s. But like everything in life, you can only really find the truth by looking at the data.


So that’s what we here at Idiro Analytics did. We are experts in data analytics for business. We looked at the data, to prove how far we’ve come as a nation, that we base our decisions on reason and logic and not on whether or not our palm itches (so we know we’ll be coming into some money). But unfortunately, the data showed us our true colours.


We looked at the price of houses in Ireland over the past six years. We took the data from the Property Services Regulatory Authority, showing every house sold in the Republic of Ireland since January 1st, 2010. We analysed housing data from every corner of Ireland, looking at the values, the locations, the house names etc.

And we found that when it comes to a large decision, such as buying a house, a lot of our nation are still as superstitious as ever. The value of properties sold in counties such as Dublin, Cork, Kildare, Cavan and Longford is significantly lower if the house is a number thirteen.


When analysing the average prices of houses we can see the drop in value for houses numbered 13 compared to their neighbours 12 and 14. It seems the Dublin population are slightly more superstitious 4.01% than the people from Cork 3,46%. In Longford, this drop in value is as much as 23.8%.


So all that hard work done by Brian O’Driscoll, all of those times he put his body on the line to dispel the unlucky nature of the number 13, it appears, have all been in vain.
This isn’t the case for the entire country though, the west of Ireland can be proud that they have bucked the trend. With counties Galway having an 8.67% increase in value for houses numbered 13 over their 12 and 14 neighbours, Mayo having a 3.28% increase, and Sligo having a massive 20.22% increase.


Some other insights we’ve pulled from the data are, that houses with particular words in the names have a higher average value. If you’re looking to buy a house with “Mara” in the name (refers to the sea) in Dublin, you might have to be willing to pay up to 115.18% on average more than houses named “An Tigin” (The Cottage).


The two most popular saints in Ireland to name a house after are St. Patrick and St. Mary, although we probably could have guessed that one. With the choice of over 10,000 named saints (it’s difficult to get a definitive ‘headcount’), the Irish people prefer to keep it traditional.


Idiro Analytics provide Big Data Analytics solutions to businesses across Ireland. We help businesses gain a better return on investment by helping them understand and use the data they already have.

About Idiro
Based in Dublin, Ireland, Idiro Analytics is an award-winning provider of analytics to businesses around the world. For an overview of Idiro’s analytics services, see our homepage www.idiro.com.

Media contact information
Simon Rees, Clients & Marketing Director, Idiro Analytics.
087 240 5999 – simon.rees@idiro.com

Why you should not buy an Irish lottery ticket until the weekend

Irish lotto

Buying an Irish lottery ticket for the weekend draw? Hold on a minute.

Here’s why you should not buy an Irish lottery ticket until the weekend.  This month changes take effect in Ireland’s national lottery game – they are adding two numbers, so we now pick from 47 numbers. The odds of each row winning will now be just under one in eleven million. Expect more rollover draws, bigger wins and fewer winners.  And if you buy a Saturday ticket on the previous Sunday, you have a bigger chance of being murdered before the draw than you have of winning the big prize.

Bear with me.  There were 52 murders in Ireland last year. Therefore, the overall odds of being murdered in any given week is one in 4.5 million (one person per week, out of each of Ireland’s circa 4.5M population).  There were 196 road deaths last year, giving you average weekly odds of being killed on the roads of one in 1.2 million.  Let’s not talk about how many die of heart disease…

Let’s say you buy a 2-line ticket on a Sunday for the following weekend lottery draw.  You have a higher chance of being murdered (all things being equal) by the weekend than you have of your numbers coming up for the big prize on Saturday.  Furthermore, you have a much higher chance of dying on Ireland’s roads than of winning.

And worse, if you win the lottery, someone else may have the same numbers, meaning you have to share the prize money, reducing the benefit. If you die next week, however, it doesn’t matter how many others share your fate – there is no upside to dying in company.

Here’s the good news: your chance of being murdered or dying on the roads before the draw falls steadily as you get closer to Saturday – but your odds of winning Saturday’s lottery stay the same, at eleven million to one.

So keep your Saturday lotto draw money in your pocket, at least till the weekend!

p.s. your chance of winning the big Euromillions prize is one in 116,531,800 – so buy that ticket as close to the draw as you can. Good luck!

Some research tidbits for Christmas

Over the last couple of weeks a few interesting research items on social psychology and social network analysis have crossed our desks – so we have compiled them into this collection of research tidbits for Christmas.  Enjoy!

In-flight influence

First up, a study that shows how the decisions of people around us influence our decisions, even if we don’t know the people.  This elegant piece of analysis, written up in this working paper and covered by the Washington Post (albeit with a misleading headline) shows how our decisions about whether to purchase in-flight food and drink are influenced by those around us.   Because the study had access to reservations data, it was able to exclude groups travelling together, and control for parameters such as seat choice.

The research found that people sitting near other purchasers were 30% more likely to make in-flight purchases.  If this is the the level of influence that strangers hold over us, how much more is our behaviour influenced by those who we care about?  Answer: in Idiro’s experience, lots.

The same Washington Post article referred to an interesting piece of research demonstrating the power of peer pressure in schools.  Message to all parents: make sure your kids are in classes with people cleverer and more diligent than them.

A link analysis of languages

Multi-lingual Wikipedia editors: which languages?
Multi-lingual Wikipedia editors: which languages?

There are plenty of studies showing how which languages are spoken by the greatest number of people, which languages are economically the most powerful – but which languages serve as the pivots between other, less popular languages?  To put it another way, if you speak a minority language (like Welsh) and want to understand it written in another (e.g. Kikuyu), which other languages are necessary to make the link?  In this case, most Welsh speakers know English, as do many Kikuyu speakers – so the answer is simple: just English.

Quartz published details of an interesting MIT study looking at this in depth, using three data sources: multi-lingual Wikipedia editors, multi-lingual Twitter accounts, and book translations.  The data is displayed in an interactive website but it’s worth watching this video, as it’s a complex enough study.

One can criticise the data sources, of course (for example, the great firewall of China restricts Chinese Twitter usage) but nevertheless it’s a fascinating topic.  Here in  multi-cultural Idiro, the most common hub language is English (of course), followed, we observe, by Russian.

How many friends?

How many people do we have contact with through our mobile phones?  Idiro’s researchers took a week’s worth of connection data from a European mobile phone network, and counted the number of different phones that each person had contact with over a week.  We then plotted the distribution of the number of contacts each phone had – in other words, the total number of links per person.  As the graph shows, a number of phones were (as one would expect) used rarely or not at all that week.  A few users made over sixty unique connections in a week, and a large number of people made between 5 and 15 connections.  We compared Christmas with an average summer week. and found – no surprise – that people make more connections over Christmas week, as we renew old friendships.

Distribution of the number of mobile contacts per person
Distribution of the number of mobile contacts per person

Finally, here is a study

by Hill and Dunbar demonstrating that Christmas card networks are (or were, when we used to send Christmas card to all our friends) a reasonable approximation of Dunbar’s number – 150.

Merry Christmas to all, from the Idiro team

Sex, teenagers and Big Data

A good friend said to me recently that Big Data and analytics is a bit like teenagers and sex; everybody is talking about it but very few are actually doing it. I think he may need to update his knowledge of teenage behaviour but I got his point nonetheless.

 

The rush to Big Data has the usual hallmarks of other past industry hot trends i.e. lots of hot air and hype. Additionally, there are a lot of definitions of what Big Data actually is and what differentiates it from, say, your bog standard Oracle BI/data warehouse.

So what’s my definition of Big Data? If pushed I’d say something similar to the following: “Big Data is the discipline of analysing vast volumes of structured and/or unstructured data with a view to generating insights and predictions that improve business performance”  (OK, I know that’s not very inclusive of non-business activity but you get the general idea).

My gripe about some soi-disant Big Data companies is that all they have done is moved their dashboard reporting tool to Hadoop (if even that). I can understand the temptation to rebrand an existing BI tool as a Big Data platform but it would be unfortunate if anyone actually fell for that.

Here in Idiro we like to differentiate between BI and predictive analytics – there are many companies offering BI tools of varying levels of sophistication. However, there are far fewer suppliers of predictive analytics platforms  (and even fewer still who provide predictive social network analysis like ourselves). In essence, BI tells you what did happen (i.e. after the horse has bolted) and predictive analytics tell you what will happen (while the horse is still happy in the barn). A smart company will use both.

But back to the definition of Big Data…some would argue that Big Data is all about analysing unstructured data such as blog postings, tweets and other such rubbish. Sorry, yes, I know there is useful information in there but there’s a lot of junk too. We prefer not to discriminate against data and believe that any data can form the input for a Big Data platform.

A word of caution lest anyone think that by installing some Big Data platform that all their problems will be solved. The analytics generated by any such platform need to be used to change business behaviour – this is probably the biggest challenge to the successful deployment of analytics within a company. Often there is political resistance within a company to the use of analytics that makes the Israeli-Palestinian problem seem like a walk in the park. Simply put, people and processes need to change if a company is going to capitalise on its investment in analytics.

As for the aforementioned teenagers, I think that when it comes to the adoption of behaviours that they find “beneficial” they exhibit a lot more openness to change than some large companies who desperately need to reinvent themselves. Big Data may or may not be a panacea for all a company’s problems but, once we step away from the buzz and the hype, what we see is that companies small and large, who intelligently leverage analytics for business really do get the edge over their competitors. Call it Big Data, call it analytics, the important thing is to call it right.

Aidan Connolly
Email: a.m.connolly@idiro.com