Author Archives: Stuart G. Hall

About Stuart G. Hall

Making a positive difference one day at a time. #London #Leicester

A quick example of thinslicing – to find the data and to act on the data

Posted on by
Reply

1. Consider this excerpt from Wikipedia on the Friendship paradox, as way of a quick mathematical -based example of ‘thinslicing’, that helps predict disease epidemics:

The analysis of the friendship paradox implies that the friends of randomly selected individuals are likely to have higher than average centrality. This observation has been used as a way to forecast and slow the course of epidemics, by using this random selection process to choose individuals to immunize or monitor for infection while avoiding the need for a complex computation of the centrality of all nodes in the network.[5][6][7]

2. Then consider that this is probably what happened in one New York community, prior to the full impact of HIV, to quote one study from Dr Sam Friedman:

In the period from 1976 to the early 1980’s, seroprevalence in New York rose from zero to about 50%…The epidemic then entered a period of dynamic stabilization…Although mathematical models have suggested network saturation may have been an important part of the stabilization process (Blower, 1991), the sociometric analysis of drug injectors’ networks conducted during the research for this volume suggest that the extent of network saturation may have been quite limited.

Behaviour change probably made a major contribution to the stabilization of seroprevalence. In spite of a popular image that would suggest that either “slavery to their addiction” or “hedonistic, selfish personalities that ignore risks and social responsibility,” drug injectors in New York (and indeed, throughout the world) have acted both to protect themselves and others against the AIDS epidemic. Thus, by 1984, before there were any programs other than the mass media to inform them about AIDS or to help to protect themselves, drug injectors in New York were engaged in widespread risk reduction…Furthermore, observations on the street confirmed this by showing that drug dealers were competing with others for business by offering free sterile syringes along with their drugs as AIDS-prevention techniques.

BTW if you’ve stumbled on this post and wonder what it all means, join the club. I am still working on myself, but there’s something here about ‘thinslicing’ as an outsider – in this example finding who to immunize in an epidemic; and ‘thinslicing’ from an insider perspective, in this example, who with little information people figured out how to take precautionary measures.Hence the title addition – to find the data and to act on the data..

‘Thinslicing’ connects the data, to the behaviour that creates it

Posted on by
7

If you’re here for the two examples of companies that improved customer service by allowing people (customers) to talk to people (employees), highlighted in red – and the 2nd example is in the 3rd comment. You can ignore the stuff about thinslicing:-)

To explain why I like the term thinslicing first take a look at the cool piece about data interpretation written today by Lithium’s Dr Michael Wu, including this neat illustration:

Then consider this, that my response to reading this blog post clarified a key thing I have been trying to say. Firstly, that I’ve come to term the business objective of finding the “interpretable, relevant and novel” in data as Michael terms it – through a combination of art and science – namely that of thinslicing.

thinslicing

But now I’ve made the next step. Identifying the value of thinslicing lies in the elegant and powerful way the term thinslicing connects the approach to data analytics to the behaviour that creates that data – namely with the thinslicing of online consumers who “tend to ignore most information available and instead ‘slice off’ a few relevant information or behavioral cues that are often social to make intuitive decisions,” as Brian Solis puts it. 

But perhaps it would help if I made clear what I don’t mean by thinslicing as a strategic tool, is that summed by nicely in these two paragraphs written by Bob Thompson on the CustomerThink community:

“Despite our best efforts to collect and analyze data, good business decisions will always include elements of judgement, intuition or just plain luck. Many day-to-day decisions are made with little or no thought, because the option selected just seems “right.” Gut-feel decisions might be examples of what Malcolm Gladwell called “thin-slicing” in his provocative 2005 bestseller Blink.

“However, the best decision can sometimes be counter-intuitive. For example, the financial services firm Assurant Solutions wanted to improve its “save” rate on customers calling in to cancel their protection insurance. The industry’s conventional wisdom, which resulted in 15-16% retention rates, was to focus on reducing wait time to boost customer satisfaction. But data analysis found a solution that tripled the retention rate: matching customer service reps with customers based on rapport and affinity.”

What I mean is the approach to data as you outline above which I categorize as thinslicing, coupled with the way consumers make purchasing decisions – which like good business “will always include elements of judgment, intuition or just plain luck”.

In other words by thinslicing, rather than using intuition to make decisions, I mean adopting a strategy which is based on the understanding that by connecting the means of analyzing the data with the way the data is created by customers.

The question then is why? While it may be clever to see a way which logically connects the way to analyse data with the way it’s created, why is that potentially so useful to a business? Now there’s a good question. The obvious answer is that by aligning the analytic method used by your business, with the way the data is created by your customers, you are going to produce better results in terms of both better quality actionable recommendations which also produce an increase in ROI. How does that sound?

Update: so there’s a nice response from Dr Michael Wu on that question of linking the too together, the way you approach the data, with the way its created, that connects the two ends of the spectrum together:

Good data scientists must know everything that happen to the data, from its creation, all the way to the point where they get their hands on the data. It is actually a pretty standard practice for hardcore financial/business analysts. Not only you need to “connecting the means of analyzing the data with the way the data is created,” you must know everything that happen to the data along the way, until the data reaches you (or the analyst). Only then can you be certain that your analysis is not biased or confounded by something before you get your hands on it. In statistics term, only then can you know the confidence interval of your result.