Oz Analytics

The Business Case Just Gets Better

I'm not blogging as much as I used to as Doclens is taking up much of my time, and when it's not Doclens then it is my consulting activities which have now gone global. So the best way of keeping up with my comings and goings is to subscribe for free to Doclens.

Despite this, I will blog every now and again and today is one of those days ...

A couple of years ago, Bain published an infographic called 'The Who, Why And How Of Big Data'. The gist of the message being communicated was that Financial Services companies that spend big on big data were twice as likely to be outperforming their peers in just about every metric that matters: speed, quality, and financial return.

My consulting activities and running Six Degrees of Data have thrown-up a lot more data on this and it has become pretty clear to me that the Bain infographic remains true - but the business case is even stronger today. Investing in big data and analytics returns a healthy 2 - 5 times. Of course, there is still one critical qualification to this: you've got to be in an organization that really 'gets' data and analytics. Brain dead companies defending their incumbent market share still waste huge amounts pretending to invest in data. Those few projects that they succeed at are still unlikely to be transformative.

That's the sad truth and it is hard to conclude otherwise. Take a look at the many articles on Doclens for the stats to back this up.

Here is the Bain infographic from 2013:

You can read more of my original analysis on Doclens.

8 Weeks Into Our Beta Launch

It's been sometime since my last blog entry. Apologies, but I have a good reason: I launched my beta website doclens.com. It's now a couple of months on and time to communicate what the experience has been like.

First off, what is doclens?

Our mission is to make data driven narratives by discovering insights from reports and articles.We do this by:

1. Using the online research and opinion published by thought leaders such as consultancies, think tanks and academic experts. We estimate 50,000 – 100,000 items are published annually in our beta target market of analytics and big data people in the financial services industry. We automatically monitor this research as it is published.
2. Our DocLens platform automatically reads, understands and analyses this research and links it to profiles of the people and organisations involved.
3. We then curate the information and publish the best 1% to our website – making it possible for you to keep up with best practices, be the most informed person in your organisation, and make smarter decisions. All in just minutes of your time.

We turn information overload into personal strategic advantage:

Doclens is a lean startup. We have 4 people involved - with only one currently working full-time on the venture. We are also self funded. So resources are incredibly tight, but as the lean movement has taught us, it is amazing what can be achieved.

So here is a summary of the first 8 weeks of Six Degrees of Data as a 'trading entity'.

The launch has been a success. Being a data scientist, I measure success against a set of short-term targets we agreed on prior to launching. These are:

• 100+ subscribers by the end of 2014
• 5,000 article reads within 8 weeks of launch
• Average visit time > 2.5 minutes.

I'm pleased to say that all 3 targets were exceeded.

Whew ...

What else can I say about the experience so far? Well here are some of the highlights, in no particular order of importance or time sequence:

• For 3 of the 8 weeks we conducted A/B tests using 4 social media networks to advertise, using a total of 14 different messages. Twitter has been the outstanding performer for us.
• That said, we have also been actively tweeting and posting about the content being published on doclens each day. Judging by the number of retweets, favourites, mentions and clicks we are getting, this is maybe even more effective than paid advertising. I'm not yet sure how I can test this in practice (given our resource and budget constraints).
• Our customers are of course central to everything we do and some of the organizations they represent are:

• There is no particular geographic concentration in our hundreds of users - at least nothing unusual that I can detect. The USA, Western Europe and Australia feature heavily, as does Scandinavia, New Zealand and Singapore. Our users come from over 80 countries now.
• 95% are data and analytics professionals. Good news as this is are target market.

So what hasn't gone well or smoothly? A few things, that's for sure:

• Social media advertising. LinkedIn, Google and Facebook have not performed well so far. Advertising on Twitter was about 5 times more effective than these other channels. I still don't know why although my 600+ followers on Twitter may be a factor. I need to run some more A/B tests to resolve this.
• Customer engagement. People just aren't commenting on the content we are posting. Perversly, they seem to love the content and we get lots of visits. Our most popular articles have now been read thousands of times each. The silence is puzzling and I will probably survey our customers soon to find out what is going on.

Longer-term (6 month) targets were also set and these are:

• 1,000+ subscribers
• 20,000 article reads
• Average visit time > 5 minutes
• CAC < 40 cents
• Launch of our analyse-your-own-document service.

CAC = Customer Acquisition Cost. In our case, the cost of advertising on social media.

So these are the basic measures we are using. They are useful to have and us help focus on the things that matter.

It has been a slog, with many, many hours expended. And many more to come. But it is fun, we are learning every day and I wouldn't be anywhere else at the moment. I'll let you know if this is still how I feel in about 20 weeks time.

Do you know what is bigger than Big Data?

From episode 10 of my Naked Analyst Channel on YouTube.

I think I do - and it is the ‘appification’ of analytics. What I mean by this is the reduction of a complex analytic activity such as market segmentation, down to a single button on your computer interface. Very much like the Apps on your smartphone, tablet or increasingly your desktop.

That’s what it looks like but the impacts are more profound. That’s because it makes it possible for analytics to be successfully done by people who may not understand how it works, but do understand the ‘why’ and ‘when’ they need to do it.

For example, a marketer in a company can access more sophisticated views of their campaigns without the need of a specialist analyst. Appification extends the range of analytic things that a non-specialist can do.

This appification is made possible because of three things that have emerged in recent years:

1. The rapid increase in the number and sophistication of APIs (Application Programming Interface).
2. The rise of open source analytic platforms like R (http://www.r-project.org). These platforms have created vast libraries of algorithms freely available to anyone who knows how to use the platform.
3. The shear number of people and organisations involved in creating open source analytic platforms like R.

The last enabler needs a little further explanation. R is a free software programming language and software environment for statistical computing and graphics. It contains thousands of packages (10,000?) specializing in topics like econometrics, data mining, spatial analysis, and bio-informatics. Nobody knows how many R users there are, but a reliable estimate (see http://spatial.ly/2013/06/r_activity/) puts it in the millions. Many thousands have helped R develop over the years.  I think that this sort of large-scale self-organising open source effort is beginning to teach the world how to use analytic algorithms.

The above is all supposition, but I can back this up with evidence. Here are 4 examples of algorithm markets - or at least they exhibit varying degrees of ‘algorithm marketness’.

• Dataxu - senses and reacts in real time to changing consumer behavior. Openness is at the technical core of DataXu. The DataXu platform is a flexible technology stack with open APIs that make it easy to integrate and extend functionality.
• algorithms.io - delivers machine learning for streaming data. The Algorithms.io cloud platform makes it easy to use machine learning algorithms to classify streaming data from connected devices. Turn the Internet of Things into the Internet of Action
• Snapanalytx - aim to provide predictive analytics for all and make them more accessible and affordable.
•  Algorithmia - are building a community around state-of-the-art algorithm development, where users can create, share, and build, on other algorithms, and then instantly make them available as a web service.

In conclusion, there is still one issue needing resolution before algorithm markets take-off: How will the world’s business people get data into and out of these algorithm apps? I’m not sure yet, but I think the answer will be more apps. Apps that themselves appify the transformation and loading of data into and out of algorithm apps.

Confused? Well don’t be, like most new and shiny things, what we are talking about is just the next generation of ETL - something business intelligence people like myself have been building for the last 20 or more years.

What’s old is new again? Maybe, but from my perspective the future looks exciting for analytics.

My YouTube channel contains tips from a 25 year veteran of the analytic profession.

What Are Analytic Marketplaces?

The Naked analyst Episode 9 looks at the best places for analytic competitions.

Here are my top 4 sites. They range from the the well known to some more specialist sites. There are 3 marketplaces I know of, and a 'world championship' currently running.

Kaggle

Kaggle is the world's largest community of data scientists. They compete with each other to solve complex data science problems, and the top competitors are invited to work on the most interesting and sensitive business problems from some of the world’s biggest companies through Masters competitions. Their most famous competition to date was run for netflix where a prize of $1 million was awarded.  Details of the winning team and how they did it can be found here.

Innocentive

Claims to have a global network of millions of problem solvers, a proven challenge methodology, and cloud-based innovation management platform combine to help their clients transform their economics of innovation through rapid solution delivery and the development of sustainable open innovation programs. Clients have included Astra Zeneca, Booz Allen Hamilton, Cleveland Clinic, Eli Lilly & Company, NASA, Nature Publishing Group, Procter & Gamble, Scientific American, Syngenta, The Economist, Thomson Reuters, and the US Department of Defense.

Topcoder

Also claims to be the world’s largest Crowdsourcing Development, Design, and Data Science Platform. The [topcoder] community gathers the world’s experts in design, development and data science to work on interesting and challenging problems for fun and reward.

Texata

The Big Data Analytics World Championship is run by the Professional Services Champions League (PSCL) and involves two online qualification rounds and a Live World Finals event held in Texas, USA in November 2014. Participants from more than 100 countries will compete individually in two online qualification rounds.

They also run other ‘world championships’ such as ModelOff (Financial Modeling World Championships) and LoyaltyGames (Loyalty and Gamification World Championships).

 

My YouTube channel contains tips from a 25 year veteran of the analytic profession.

The Top Websites For Big Data

The Naked analyst Episode 8 looks at the best sites to learn about the world of Big Data.

Here are my top 6 sites. They range from the easily digestible to the totally immersable and possibly overwhelming.

1. Dataversity (http://www.dataversity.net/category/data-topics/big-data/) daily updates of the latest Big Data News, Articles, & Education - not too overwhelming with 1 or 2 items each day.
2. Planet big data (http://planetbigdata.com) is an aggregator of blogs about big data, Hadoop, and related topics. In terms of volume of news, it is a step up from Dataversity with up to half a dozen items daily.
3. SmartData Collective (http://smartdatacollective.com/all/8731?ref=navbar) a commercially run and moderated business community for business intelligence, predictive analytics, and data professional bloggers.
4. BigData Startups (http://www.bigdata-startups.com) is a European site that is making an effort to track the ‘market buzz’ - especially emerging companies. Hundreds of companies are profiles so it can be a bit overwhelming for those new to Big Data.
5. Data Science Central (http://www.datasciencecentral.com) is a good online resource for big data practitioners and quality is maintained by content being actively moderated. This saves it from being overwhelmed by marketing crap.
6. R bloggers (http://www.r-bloggers.com) is more for practitioner Analysts and may be overwhelming for those new to Big Data - but it is a fantastic resource for anyone using the R language to do real stuff. And besides, a list of Big Data sites just has to include one covering the R community because it is such an important open source movement. Don’t let anyone tell you that R is just a fringe technology. A huge list of organisations use it - from the biggest multinationals to the smallest startups.

Finally, if you are using a particular tool or solution - then find a good source of information for that and get into the habit of checking those sites regularly.

If you use all or some of the sites I have mentioned then you will quickly absorb a lot of new information about Big Data - but don’t forget to look at these sites critically as you’ll get a lot more of of them that way!

 

My YouTube channel contains tips from a 25 year veteran of the analytic profession.

The Analytic Career Part 2 of 2

What is a career in analytics? What sort of work is involved and what should you expect?

Here are 4 things I think you can expect:

1. Your work will be project based. To me, it's a warning sign if it isn't a project. Your employment status may be permanent, part time or contract. That's a separate issue
2. You will spend most of your time translating business needs into analytic solutions
3. The higher you get in your role, the more political you have to be
4. Most people don't get what you do. You will need to balance educating them with simply advocating that they trust you based on your track record.

Tips from a 25 year veteran of the analytic profession.

 

The Analytic Career Part 1 of 2

Just published on my YouTube channel:

What is a career in analytics? What's it like and what are the key features to expect?

Here are 4 features I think you can expect:

1. You won't have just one career but several
2. Don't look to your employer to help too much. It's up to you
3. If you are not learning on the job AND having fun - leave
4. Learn or be replaced by cheaper alternatives,

 

In the 2nd and concluding part of this discussion I will talk about what sort of work is involved and what you should expect as an analyst working in a commercial organisation.

Tips from a 25 year veteran of the analytic profession.

Shiny New Toys

Just published on my YouTube channel:

 

The Naked Analyst Episode 5

This video contains some advice on how to address Big Data opportunities. Believing vendor and pundit hype is not always healthy. Test their assertions and make use of your own analytic brain - it's good practice!

Think also about the impact you can have on your colleagues when you build a solution using your current toolset. It may not be perfect, but if you can deliver a 70% solution for $0 then that's impressive.

But don't get too confident - under promise and over deliver is always better than the other way around. Start with a small pilot or test to prove viability if you can.

And just to make life difficult - there are a large number of solutions that are possible only with the new Big Data tools. Again, thinking for yourself and knowing who to trust is the best strategy of all.

Tips from a 25 year veteran of the analytic profession.

The Difference Between Analytics and Big Data

Just published on my YouTube channel:

The Difference Between Analytics and Big Data (The Naked analyst Episode 4)

 

How to Find Your First Job as an Analyst

Continuing on my YouTube Channel covering everything you wanted to know about being an analyst, business analyst, data scientist, and any other data-using role in business today.

Here is part 2 of my tips on how to land a job in analytics:

 

Part 1 was published a few days ago here and you can subscribe to the channel here as well.

As mentioned before, I've also put together a related playlist on YouTube of a few videos that include those made by others on similar topics that I have found on the net. 30 minutes of watching this whole list should give you some insight into how you can go about finding that first job. Good luck!