Sunday, January 8, 2012

5 Predictions for Online Data in 2012 (Mashable)

Jones-Dilworth, Inc., a public relations consultancy focused on bringing early-stage technologies to market. In the spirit of the new year, I've determined my online data predictions for 2012. And because I'm attempting to make this an annual experiment, let?s begin my looking back at how my 2011 predictions shook out.

[More from Mashable: How to Access the Best New Features in Google Analytics]

1. Personal data management matures into an industry. I?ll admit the prediction was largely a fail. Ambitious startups like Singly + The Locker Project got off the ground, but did not pick up steam. Many new, related startups are in the works, but none have even come close to the goal of managing one's data the way he manages his health or wealth, for instance. An end-to-end platform may emerge, but it will take time.

2. The flood gates of corporate data open widely. This prediction has come close to being true. It certainly has become important to mine, parse and manifest corporate data for internal and external uses alike. The predominant question has indeed shifted from ?Should I make my data open and available?? to ?How can I do it best?" But at the same time, a lot of people are still sitting on the sidelines, waiting to see how things develop.

[More from Mashable: 5 Predictions for Social Media Law in 2012]

3. Big data gets regulated. Facebook agreed to submit to independent privacy audits as part of its long-gestating FTC settlement. I?d argue that we?ll see more regulation in 2012. Congress has taken an active role, and the big Internet companies have only further increased their lobbying spend. Controversies surrounding Carrier IQ, Apple and SOPA have accelerated public interest.

Of course there?s good regulation and bad regulation. Anyone doing business in or with data is going to have to understand how government works, and play an active role.

4. The trend itself gets old and tired. Another outright fail. It is certainly true that everything is becoming, or has already become, data-driven, but we haven?t yet had the hangover I predicted. If anything, we?re full-steam ahead, and as ebullient and ambitious as ever.

This is good and bad -- escalating investment of both human and real capital will spur innovation and speed up the inevitable. But the pace also makes us more vulnerable. Data for data?s sake, or data for self-justification is an ongoing risk.

5. Data scientists become the new community managers. I think it?s fair to say that this happened, by a large margin. Check out this chart from Indeed that chronicles the explosion of data scientist jobs. CMSWire explores the particulars in more detail.

Last year aside, what?s going to matter in 2012?


1. Future Tense Analytics -- Hello, McFly?


For a long time, were analytics defined by the past -- what happened a day ago? A week ago? A year ago?

The arrival of real-time technologies sped things up quite a bit. It?s now commonplace to ask, what?s happening? What?s trending? What?s changing? Today?s best data-driven technologies and strategies orchestrate insights in the present tense.

Naturally, predictive analytics tackle the future tense. This isn?t a new notion, but the underlying technologies are finally mature enough for predictive use. Keep an eye on places like Decide.com. In 2012, future-tense capabilities will become widespread, and in high demand.


2. Mixing and Matching Data Sets


One of the things that comes right after wider availability of data is the desire to put that data in context. For instance, this type of mashup culture is what put Mashable itself on the map back in 2006 and 2007. For other brands and enterprises, 2012 will be all about crashing diverse data sets together, and seeing what clicks.

I?m actually a dissenter when it comes to companies like Factual or SimpleGeo?s heavy emphasis on location as a single data vertical. BlueKai's ad targeting are other examples of early leaders in a particular domain.

It?s not at all that these companies aren?t cool or useful, but the most interesting things happen when you mash up data sources. For example, Umbel can provide incredible insight into your audience and its habits, but it?s even more interesting when merged with usage logs, comment streams and share counts.

The year 2012 will be about putting corporate data alongside open public sources -- internal data meets external data.

This is why I continue to be a big fan of data marketplaces like Infochimps, a company I advise, which has bundled data sets and API surrounding key themes, instead of verticals. The most actionable insights come from a diversity of data.


3. The Data Scientist Talent Grab


A few years ago, I remember reading job descriptions for Ruby on Rails programmers who should have ?5+ years of experience.? This was funny, as Rails was only three years old at the time.

That is what happens when a certain skill set spikes in demand. Employers assume market maturity when there is none, hiring gets very competitive, salaries rise, the very best practitioners give their employers an unfair advantage, and the market is flooded with wannabes (some of whom prove their mettle, some who do not).

Training and professional development initiatives have taken center stage, and the land grab is on. McKinsey predicts that by 2018, the U.S. alone will face a 140,000 to 190,000-person shortage of professionals who have deep analytical skills, as well as a shortfall of 1.5 million managers and analysts with the know-how to analyze big data to make effective decisions.

We?re even seeing venture firms like Greylock hire top 5% data scientists to work across the entire portfolio, as one solution to the problem. And companies are employing new means of data science outsourcing (see: Kaggle, Mu Sigma).

Since it?s all still a relatively nascent field, I urge you to reinvest in the talent you already house. Knowledge of your own business is just as important as knowledge of data science. There are requisite skills of course, and it takes a certain kind of person, but in my own experience, there are bound to be many qualified individuals who are hungry and game.

Soon there will be a much wider gap between the novices and the experts. However, don?t think of data science chops as just something you can hire for, and be done with. It?s a transformative trend, something that will eventually cut horizontally across your entire organization.

In 2012, data science will appear in job descriptions left and right, and will become the most in-demand skill set in the technology sector.


4. Data-Driven Everywhere


We generally think of data as the domain of large B2B companies that sell enterprise software, or of advertising networks and targeting systems that have an easy way of assigning a dollar value to their work. But the legal, pharma and law enforcement sectors have also been natural fits.

In 2012, we?ll see data-driven innovation and data-centric design emerge from the unlikeliest of places -- think data-driven fashion, crop insurance, Hollywood box office and even humanities education.

Essentially, don't rule out the possibility of data-driven strategies anywhere. Even if a business doesn?t run on data, it absolutely produces very interesting exhaust that can be used advantageously.


5. Data-Driven Non-profits


Data has a higher purpose too --- it?s not just a pageview multiplier, or a CPM lift. ?Data can help us understand how we are doing as a society and as a culture.

Here I?m not talking about non-profits that use data to grow. I?m talking about non-profit initiatives that are built around data in and of themselves. Two great examples are CommonCrawl and the Earth Dashboard Project. Common Crawl builds and maintains a free and open crawl of the web and all its data, with the purpose of forwarding important research that requires large-scale analysis. The Earth Dashboard Project is an effort to build a living report card of sorts for the entire globe. It will be installed at the UN Headquarters in New York City as a way of reminding delegates and visitors of the bigger picture, our interconnectedness, and what we?re really working toward every day.

There is a lot to be done with data, not nearly all of it profitable. In 2012, we?ll see five to seven new data-driven non-profits start to bite off important chunks of value that sit outside of corporate and governmental concerns.

Image courtesy of iStockphoto, nadla

This story originally published on Mashable here.

Source: http://us.rd.yahoo.com/dailynews/rss/internet/*http%3A//news.yahoo.com/s/mashable/20120105/tc_mashable/5_predictions_for_online_data_in2012

lady gaga marry the night video pac 12 championship game pac 12 championship game bobby valentine bobby valentine al franken al franken

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.