All-purpose announcement ,,,,

I am, strangely, reminded of the late and great Peter Sellers' all-purpose party political speech, when I read the following announcement about an upcoming conference:

"Business Analysis is growing at a phenomenal pace. Organisations are increasingly recognising the importance of business analysis in achieving successful change and many are investing heavily in developing this capability. The skills and techniques of business analysts are invaluable in shaping and forming business change overall as well as making the most of opportunities presented by new technologies. With the backdrop of a global economy recovering from a deep recession, organisations are in a constant state of fast-moving change and can't afford to get things wrong. Business Analysis capability is key to identifying what's needed and developing solutions that equip organisations for the future." Just substitute almost any field of endeavour for the term 'Business Analysis' and you will see what I mean.

Work(s) in progress

Forschungswerk (literally “Research works”), a Nürnberg based marketing research company, has recently (via www.marktforschung.de) announced[1] the availability of a procedure for statistically testing shifts/changes in NPS (Net Promoter Score).

As many readers would know (i.e. all two of you), when people think about NPS as a metric, they basically fall into two camps: those who think it is rubbish, and those who think it is the bee’s knees.

Regardless, NPS is here and it will stay.  So the question naturally arises, how do we know what is a ‘good’ NPS result, but probably of greater importance, how do we know when a change or difference in NPS is statistically significant.

To recap, NPS is simply the difference between the proportion of respondents who give an answer of 9 or 10 and the proportion who give an answer of 0 thru 6 on an 11-point recommendation scale.

It can in fact be easily computed in the statistical package of your choice with the following recoding scheme …

0 thru 6 = -100

7 or 8 = 0

9 or 10 = 100

… and then calculating the average score[2].  This then gives you enormous flexibility to (say) crosstab NPS by almost anything you fancy (i.e. there is no need to compute NPS ‘offline’ for desired groupings of respondents).

Unfortunately, the standard statistical significance tests do not apply when comparing different NPS results.  For a start, it is not too difficult to show that NPS has twice the sampling variability of a simple proportion (which does, of course, bring into question its suitability as a tracking metric – oops, here come those letters again).

There is, however, a way out, and one that I suspect that the Forschungswerk people have taken; namely to utilise a permutation test approach.  This is kind of hard to explain, but essentially, a null hypothesis that there is no difference between two NPS values is equivalent to saying that the two sets of results come from the same (statistical) population.  So if they come from the same population, then clearly we could validly mix up both sets of respondents into one group, and redraw our two samples, with no real difference in results.

And that is just what permutation testing does, except that it does it many, many times.  For example, we could draw 500 or 1,000 new samples underpinning each of the two NPS values from the combined respondent group, and see on how many of these 500/1,000 occasions the two NPS values were more different than they were in the original two samples.  If it’s less than 5% of occasions, then we can say that there is a statistically significant difference between the two original NPS values at p < .05.

This same approach can in fact be used to test the difference between other, equally intractable (i.e. from a statistical point of view) metrics, such as CVA ratios.

Clearly this approach isn’t ideal, since most standard packages used by the marketing research community won’t do it.  But there are some Excel™-based routines available (e.g. the excellent and home-grown PopTools[3]).

[1] http://www.marktforschung.de/information/nachrichten/marktforschung/indextest-neues-tool-von-forschungswerk/backpid/2958/

[2] Source:  Dr. Tim Bock www.q-researchsoftware.com

[3] http://www.poptools.org/

Best biscuits in the world ...

There are reports that Woolworths is foundering, or if not foundering, at least lagging Coles, not the least because of the foray into 'big box' hardware.  http://www.theage.com.au/business/choosing-the-right-hardware-for-growth-20130726-2qpp2.html

[Have you been to Masters?  It's great.  There are no customers to get in your way, and it's laid out pretty much just like Bunnings, so it's easy to find your way around.]

But I think I have the answer, and it's not that Woolworths should dump the Masters initiative.

PUT KOOKAS COUNTRY COOKIES BACK ON THE SHELVES !!

These are the best biscuits in the world, they are made in country Victoria, at Donald, and they should be regarded as a national treasure.

http://www.kookas.com.au/index.html

And Woolworths has chosen not to sell them any more ?  Makes absolutely no sense.

Shame, Woolworths, shame.

My goodness, that's a breath of fresh air ... :-)

The UTS Advanced Analytics Institute are running a one day course entitled "Introduction to Advanced Data Analytics".  Details are to be found  via this link ...

https://shortcourses-bookings.uts.edu.au/ClientView/Schedules/ScheduleDetail.aspx?ScheduleID=1270&EventID=1105

This particular course isn't for me, but I do (sincerely) commend the organisers for not doing the usual, and charging 'commercial' participants much more than participants from academia.  I also commend them for charging significantly less for interstate participants.

Both are very good decisions. 

For more info on the AAI and its activities, see here ... 

http://cfsites1.uts.edu.au/aair/news-events/newsletters.html

Is Westpac there yet ?

"Dear Westpac ...just some feedback … Having received your kind invitation to 'upgrade' to a super-duper Black Mastercard, I phoned the 'exclusive' Black 1300 number to ask a couple of simple questions, and sat for 12 minutes while waiting for someone at this supposedly 'Black Card Priority Service' number to be available to talk to me, before I had to go and do something else and gave up waiting.  

I would have been happy to leave my number and for someone to call back, but I know (from previous encounters with the Westpac credit card enquiry line) that this is not an option that Westpac offers.  [Even the bog-standard ANZ service is better than this, with many enquiries being able to be handled via their online secure email system.]  

So on the basis of this first and (so far) only experience, I would have to doubt Westpac’s claims of ‘black’ signifying a distinctly superior card product." 

Of course, I wasn't happy to leave it there ... I took the trouble to fill out a product enquiry request on the Westpac website, giving details of the above experience, only to push the 'send' button and receive this message:

So then I went to the section of the Westpac website that enables me to give feedback/file a complaint.  And  not wanting to type it all over again, I attempted to paste what was essentially the above text into the feedback/complaints box, only to receive this message:

The upshot is that I really don't think Westpac has got it, even after all their years in the market place.

So what am I going to do about it?  I'm buying more Westpac shares.  If they are using this approach to boost profits, then I want my share of them!

POSTSCRIPT 1 - (16 August 2013): I did indeed buy more Westpac shares, and so far (in just two months approx.) they have gone up 12% ! 

POSTCRIPT 2 -  (20 September 2013): Since writing the above post, I've had occasion to use my nice new black Mastercard to pay a sizeable sum for an overseas holiday, in $USD.  The fees charged by Westpac for this transaction of $USD20,000 approx were $AUD600 !!!!   So I did a quick search to find out why.  Here are the salient points in relation to owning and using one of these cards:

• Annual fee $395

• Fee of 2.0% of the Australian dollar value of any purchases or cash advances/withdrawals when the applicable credit card scheme converts the transaction from a foreign currency into Australian dollars

• MasterCard^® Issuer Cross Border Assessment - 0.8% of the transaction amount where a transaction is made using your MasterCard and the merchant or financial institution accepting the card is located outside of Australia

• MasterCard^®Currency Conversion Assessment - 0.2% of the transaction amount for converting foreign currency transactions made using a MasterCard into Australian dollars

The upshot is that the total fees for a $USD transaction of the type I made were 3.0% of the transaction value.  And, of course, you can bet that Mastercard/Westpac used an exchange rate that was not unfavourable from their point of view.

Almost needless to say, my Westpac shares are now up 17% since my purchase of them in mid-June.

Another quick word ...

Back in February, I very briefly reviewed the Wordle™ idea http://www.wordle.net/ and Garreth Chandler’s new option  https://www.wordyup.com/ .

Since then I’ve been playing some more on the Wordle™ site and noticed a link to yet another alternative at http://textisbeautiful.net/ .

And this one is rather nice.  In the words of the authors, their approach … “deals in concepts, not words. A concept is a collection of words that is automatically discovered from the text.“  Further, it… “provides more visualisations than just the concept cloud (our concept cloud is visually similar to the Wordle output but uses more advanced text analytics underneath).”

So I thought I would try it out for the text on my own website, as I had already done with http://www.wordle.net/.  Here’s the Wordle™ result redone, for reference, followed by three alternative representations from http://textisbeautiful.net/ .

I honestly don’t know which I prefer.  The Wordle™ obviously emphasises individual words.  The textisbeautiful format emphasises what it deems to be individual themes, and there appear to be far fewer themes than words on my website.  Maybe I am shallower than I think?

Big data ? _THIS_ is big data !

This  http://www.kickstarter.com/projects/1461902402/a-bit-e-of-me?ref=card   is amazing.


Plus he does some rather cool data visualisations.


His website is http://myprivacy.info/ .