Home' Forge : Vol 2 No 2 Contents of Australia. 'A big problem at the
moment is that you have got a
number of different players in the
area who have got different skill
sets, and they don't talk to each
other very well.' He says he often
sees non-statisticians trying to do
statistical calculations and 'making
a mess of it'.
'A lot of statisticians just go
away in disgust and write off the
analytic profession. There is some
justifcation for that. There is an
awful lot of stuff that is being
done by very enthusiastic and
bright people that is ignoring 100
years or more of good statistical
'It borders on the embarrassing.
They do some advanced things,
but they also labour over issues
that are statistics 101. It is not
entirely their fault, because the
statistics profession has been very
slow to pick up on these issues.
Statisticians have got to be more
IT-aware and IT-capable.'
Henstridge says there is often
a language barrier between IT
analysts and statisticians. 'They
use different terminology and
vocabularies, which are often
referring to the same things under
different names. There are some
techniques that statisticians should
learn from the IT people, and there
are a lot of things that the IT people
should learn from the statisticians,
because statisticians have been
doing this stuff for many years.'
Another challenge with big data
is to assess whether there will be a
good return on investment, given
that there is a level of uncertainty
about what will be found. To
that extent, big data investment
is similar to mining investment.
Will the company fnd a valuable
'resource' or not?
Bhattacharjya says companies
should not view data management
as a 'cost-prohibitive exercise'.
The tools are evolving, and many
open-source solutions are becoming
available. But she does believe that
organisations should start with their
'It is about having the right
strategies in terms of what kind
of workforce you need, what kind
of data sources you are looking
at, and how you prioritise to get
the value at the right time, for
whatever products and services
you are selling. Then it is iterative,
because strategy is also iterative.
If we had a market that never
changed, you would never have to
change your strategy.
'Strategy development is also about
looking at what your competitors
and suppliers are doing, and how
end consumers are behaving.
Then, if you want to execute your
strategy, it has to appear on your
dashboard. And that, too, has to
evolve', she says.
And the future? Bhattacharjya
says the speed of processing
will increase, and the volume
of data will amplify. What will
also emerge is greater variety. 'It
is not quite possible to predict
the variety of data that will be
relevant in 10 years' time, so
companies will need to revisit
their big data--related strategies
and make sure that these are
keeping up with emerging
technologies and platforms.
'The other issue, I think, that will
be different from a consumer 's
perspective is the concern over
privacy. Although we are used to
banks and the giant supermarket
chains mining our data, the
mention of big data still makes
the end consumer uncomfortable.
I believe this will change in 10
years' time as we get used to
better services as a consequence
of companies mining our data.
The regulatory environment will
also have evolved to address our
concerns by then.'
See over the page for insights from
Australian data businesses.
Big data facts
• Every two days, we create as much information as we did
from the beginning of human civilisation until 2003.
• We create 2.5 quintillion bytes of data daily.
• Ninety per cent of the data in the world today has been
created in the last two years.
• According to the International Data Corporation, the big
data technology and services market will grow to
$41.5 billion in 2018, which is about six times the growth rate
of the overall information technology market.
• The total amount of data being created in the world doubles
every 1.2 years.
• By 2020, the amount of digital information in existence will
be 40 zettabytes (one billion terabytes).
• Unstructured data from sources such as social media sites is
growing at a faster rate than structured data.
• According to EMA/9sight, marketing uses data analysis
more than any other function.
• Google processes more than 40,000 search queries per
• The speed of processing is the number one factor driving
big data investments.
cover story // 23
Links Archive Vol 2 No 1 Vol 2 No 3 Navigation Previous Page Next Page