recent bookmarks

Twitter
Rubio is literally just stating the facts here, but in typical Rubio fashion he’s implying—without actually saying—…
from twitter_favs
10 days ago by chrchr
Twitter
Every day, in each timezone, we know when it’s the end of the school day because we see a flood of kids…
from twitter_favs
10 days ago by nureineide
Twitter
Every day, in each timezone, we know when it’s the end of the school day because we see a flood of kids…
from twitter_favs
10 days ago by indiamos
Twitter
Every day, in each timezone, we know when it’s the end of the school day because we see a flood of kids…
from twitter_favs
10 days ago by hyperfekt
Twitter
Nope. The more educated you are the more likely you are to lean left. Not because of bias b…
from twitter_favs
10 days ago by mikeschinkel
Twitter
Productive interactions among data managers at
from twitter_favs
10 days ago by ebuchholtz
Twitter
Kirsten Gillibrand (@SenGillibrand)
RT @colbertlateshow: Tonight: Senator Gillibrand believes the path to basic gun reform is getting money out of politics. #LSSC https://t.co…
US  Senators  Twitter 
10 days ago by AlexanderERyzhov
Twitter
Kirsten Gillibrand (@SenGillibrand)
A historic number of people got to the polls on Tuesday to raise their voices – especially first-time voters and young voters. This is our future. I couldn’t be prouder to see it in the hands of young people who are off the sidelines, fighting for the change they want to see.
US  Senators  Twitter 
10 days ago by AlexanderERyzhov
Erik | Applebee's Grill + Bar
I'm at Applebee's Grill + Bar in Bellevue, WA
from twitter
10 days ago by ethauvin
Let American firms run hospitals, urges free trade group | News | The Times
RT : WHOOMP THERE IT IS

The master plan to sell off the NHS that was poo-pooed as a conspiracy, now out in the open.
from twitter_favs
10 days ago by rtanglao
terryum/awesome-deep-learning-papers: The most cited deep learning papers
The most cited deep learning papers. Contribute to terryum/awesome-deep-learning-papers development by creating an account on GitHub.
academic  papers  deep-learning  list 
10 days ago by nharbour
ZEIT – Now 2.0
Introducing the next generation of our platform.
monorepo  javascript  deployment 
10 days ago by HusseinMorsy
Twitter
171 kids. Still in custody.
from twitter_favs
10 days ago by andriak
Twitter
RT : Thanks to the editorial board for this. We agree.
from twitter_favs
10 days ago by mathewi
Twitter
This is terrifying. Not just “normal” terrifying either.
from twitter
10 days ago by skyesidhe
Twitter
And that we’re now seeing a whole generation of kids become mini devops people in training because of Min…
from twitter_favs
10 days ago by hyperfekt
Twitter
RT : The Holocaust could not have happened at such speed and scope without IBM. Such brilliant technology and brilliant…
from twitter_favs
10 days ago by rtanglao
Twitter
I’ve had this thought for years that there is a whole generation of users who learned web design b/c of M…
from twitter_favs
10 days ago by rtanglao
Twitter
Excellent response to or anyone who has not delved deeply into a subject. Issues su…
from twitter_favs
10 days ago by maura
Twitter
But real talk, this task will be absolutely impossible without huge investments in modular prefab construction and…
from twitter_favs
10 days ago by burritojustice
Twitter
There have been a gazillion articles today about Tucker Carlson’s door being cracked, and that sucks, but ONE about…
from twitter_favs
10 days ago by girma
Twitter
There have been a gazillion articles today about Tucker Carlson’s door being cracked, and that sucks, but ONE about…
from twitter_favs
10 days ago by chetan
Twitter
It really isn’t far fetched to imagine Trump winning Election Night vote. Declaring himself winner and launching in…
from twitter_favs
10 days ago by chrisgonyea
Untitled (https://www.nytimes.com/2018/11/08/business/how-to-ask-someone-out-at-work.html)
Your Office Doesn’t Have Nap Time Because You Are Not a Child, from Choire Sicha’s new column
from twitter
10 days ago by becked
Twitter
اللهم اغفر لي ما قدمت وما أخرت، وما أسررت وما أعلنت، أنت المقدم وأنت المؤخر، وأنت على كل شيء قدير https://ift.tt/2l295ZZ http://twitter.com/development01/status/1060759116782751744
design  web-design  layout  art  webdev  graphic  graphicdesign  graphic-design  web-dev 
10 days ago by dcolanduno
Twitter
RT : For election night, I present to you the best thing I learned in my time as a (third-party) sysadmin in the House o…
from twitter_favs
10 days ago by rossgrady
Adam Driussi's analytics business Quantium booming 16 years after a 'crazy idea' | afr.com
Adam Driussi's analytics business Quantium booming 16 years after a 'crazy idea'
Share via Email
Share on Google Plus
Post on facebook wall
Share on twitter
Post to Linkedin
Share on Reddit
Adam Driussi, chief executive and co-founder of Quantium, sees enormous opportunities for data analytics.
Adam Driussi, chief executive and co-founder of Quantium, sees enormous opportunities for data analytics. Dominic Lorrimer
by Sally Patten
AFR BOSS magazine sat down with Quantium chief executive Adam Driussi, who helps clients unlock the potential of data.

Did you always want to be an actuary?

I was good at mathematics at school but, to be honest, when I was a kid my passion was sport. I wanted to be a rugby league journalist. I was quite deep into year 12 and I’d never even heard of actuarial studies. I was staying with a friend for the weekend and his mum gave him one of those aptitude tests. It came back with chemical engineer as first career choice and actuary as second. His mum said, “I’m taking you to the Macquarie University open day to hear what this actuary thing’s about.” So I went along with him.

Macquarie used to be the only university you could do it at. There were about 80 people in our year and I’d say 30 dropped out in the first year because they didn't enjoy it, or they just weren't good enough at maths to do it.

From insurance to retail, data analysis provides valuable insights.
From insurance to retail, data analysis provides valuable insights. Shutterstock
Even a couple of years into the course, I still didn't really know what it meant in terms of the types of jobs available.

Back then you would go and work in an insurance company or a consulting firm. I wanted to work in consulting.

What sort of work were you doing in the early days?

Quite a bit of insurance. Actuaries were just starting to work in general insurance – around claims reserving. They advised an insurance company how much it needed to set aside as a liability in its balance sheet to pay claims.

I found that a bit boring because you just do the same thing every quarter or six months and it wasn't actually changing the profitability of an insurance company. So I started doing more work in pricing.

Quantium works with giant US retailer Walmart on its forecasting systems.
Quantium works with giant US retailer Walmart on its forecasting systems. Alan Diaz
If you could help an insurance company with how it priced risk in the first place, you could create a competitive advantage for it and change profitability. When you phone up for a car insurance quote, what questions do you get asked and how do your answers change how much premium you pay? That’s ultimately customer pricing. It’s trying to come up with an algorithm that sets a premium for every customer.

That sort of algorithm interested me. It led me to think about customer behaviour and how a company could analyse customer data to help it make decisions about things like products and pricing.

You were with Deloitte, right?

I was at a company called Trowbridge, which was bought by Deloitte in 2000. I got made a partner at Deloitte when I was pretty young. Then a former client joined, and together we thought about how you could apply analytics thinking outside of insurance. We went to a client with some crazy idea on how you could analyse data to do something, and this guy’s reply was something like, “I despise actuaries.” He said, “You’re probably gonna tell when I’m gonna die and then try to sell me insurance.” He clearly wasn't into these ideas.

Driussi says there are many opportunities within the value chain for retailers.
Driussi says there are many opportunities within the value chain for retailers. Michelle Mossop
As we drove back from that meeting, I said to my colleague [Greg Schneider], who is now my business partner, “We should have a go at this.” It’s funny to think about it now, but one of the worst meetings I’ve ever had was the meeting that spurred us to set up Quantium as a business.

At Deloitte, we put our hands up to do work in any other part of the business. Another part of the company was working with Foxtel which wanted to look at customer data. It was the first time it had looked at its customer data and we were able to very quickly give them insights.

It was 2002 and I was 28 when we started Quantium. If I was starting a business now, I’d be more conscious of risk but back then I never thought for a second that it wouldn't work.I was much more focused on the opportunity than the risk, even though my wife was eight months pregnant and we had a new mortgage. We now have about 700 staff.

Which sectors do you do most of your work in?

Data can help supermarkets decide how much fresh produce to order, keeping customers happy and reducing waste.
Data can help supermarkets decide how much fresh produce to order, keeping customers happy and reducing waste. Bloomberg
Retail and FMCG (fast-moving consumer goods) are our biggest sectors and our business is now half-owned by Woolworths. There are a lot of opportunities within the value chain for a retailer. In a supermarket there’s everything, from the range of products it sells to the price of those products and which products it puts near each other. How much of the aisle should be soft drink? How much of that should be Coke? That is retail judgment, but it all has potential for data analysis to help retailers make decisions to achieve an optimal store layout.

We do forecasting work for Walmart in the US. They have about 1 million different items in store. When you’ve got 4500 stores across the US, that’s a lot of combinations to model to determine what Walmart needs to deliver to ensure no empty shelves and minimise waste. We’ve been helping Walmart with their forecasting systems, figuring out how much of that product they’re going to sell tomorrow or the next day, so that they can make sure they deliver the right amount. One of the big determinants for how much you can sell of a product is price – if we promote coconut sugar, should the promotion be 40 per cent off, or 30 per cent off, or two for one?

That’s why I find retail fascinating. There are just so many different points at which you can make decisions. Working with the suppliers, we can look at customer data to work out the products they should be developing.

What is the future for customer analytics?

"Anything where you can make it feel like a more personalised, relevant experience for the customer, that's where ...
"Anything where you can make it feel like a more personalised, relevant experience for the customer, that's where there's potential for data," says Driussi. Jessica Hromas
Retailers can go a long way to predicting what customers want to buy. If you were ordering your shopping online and you filled your basket with a bunch of items, we could use analytics to ask if you’ve forgotten to order a particular item. We could build a model to predict what we think you want to buy. Maybe you have actually forgotten something, and you say, “Thank you for reminding me.” That’s a win for the customer and it’s a sale for the retailer. That’s an example of where the analytics will get a lot smarter, predicting customer needs.

Which other sectors could use customer data better?

Life insurance could definitely be more personalised. Retail is great because it has regular interaction with its customers but with insurance, you might, if there’s no claim, only have two interactions a year with a client. The more frequent the interactions, the more opportunities to personalise.

We’re increasingly doing work in health analytics. We bought a small business a couple of years ago that was working in the health sector and now that’s one of the fastest-growing parts of our business. Government is another opportunity. It could be how they make major policy decisions, or measuring the impact of some of those decisions.

How fast are you growing?

We’ve grown about 30 per cent a year since 2002 and our plan next year is to grow just under 40 per cent. We’re 16 years in and we still think it’s like the start of something.

AFR Contributor
data-science  big-data  analytics 
10 days ago by enochko
Twitter
Gonna go live and stream some NES (probably Mario Bros 1 and possibly some BurgerTime?) over on…
from twitter
10 days ago by colonel2sheds
Old River Media Group
The latest Old River Media Group Edition! Thanks to
from twitter
10 days ago by ormg
Twitter
The latest Old River Media Group Edition! Thanks to
from twitter
10 days ago by ormg
Twitter
I can’t wait to cut articles out of this to put up on my crazy conspiracy wall and connect it…
from twitter
10 days ago by tehviking
Twitter
I can’t wait to cut articles out of this to put up on my crazy conspiracy wall and connect it…
from twitter_favs
10 days ago by mellowfish
Twitter
Could you do a future Function episode on the make vs consume issue? 🙏🏼 It’s like the difference between…
from twitter_favs
10 days ago by hyperfekt
Twitter
oh noe the tankies are upset about my post!!! forgive me, in that instant i had forgotten that the only acceptable…
from twitter_favs
10 days ago by sharp
Twitter
It's such a weird thing the way the media reports on countless voter suppression schemes before election day but th…
from twitter_favs
10 days ago by jeffc76
Twitter
Favorite tweet:

Sometimes I look at Firewatch screenshots again (in this case, from the Switch branch of the game) and am reminded that Jane, Olly, Paolo, Will, Ben and our contract prop modelers made the game look ridiculously nice. pic.twitter.com/YYU9XGQW2h

— Jake Rodkin (@ja2ke) November 9, 2018
IFTTT  Twitter 
10 days ago by Ryanvlower
« earlier      later »

Copy this bookmark:



description:


tags:



Copy this bookmark:



description:


tags: