IoT
IoT
Cloud // Software as a Service
Commentary
1/29/2015
03:52 PM
Doug Henschen
Doug Henschen
Commentary
Connect Directly
LinkedIn
Twitter
Google+
RSS
100%
0%

Data Analyst: Does Everybody Need To Be One?

Microsoft Power BI and IBM Watson Analytics attempt to make data analysis accessible to all. But don't confuse access to tools with reaching correct and valuable conclusions.

10 Cloud Analytics & BI Platforms For Business
10 Cloud Analytics & BI Platforms For Business
(Click image for larger view and slideshow.)

Will the new "freemium" Power BI, introduced by Microsoft this week, single-handedly democratize data analysis? I doubt it, but it will accelerate change in an already fast-evolving analytics and business intelligence market.

The point here is that free and easy-to-use tools are great, but be careful not to confuse widespread access to tools with reaching correct and valuable conclusions.

The stated goal of offering a free service level on cloud-based Power BI is to "open up the floodgates to all business users," according to Microsoft. The new Power BI also offers a free data-visualization tool called Power BI Designer. But free is actually nothing new in the world of business intelligence.

"All the large BI vendors have been introducing free versions of their product," said Francois Ajenstat, vice president of product management at Tableau Software, in an interview with InformationWeek. The examples date back many years.

[ Want more on analytics and BI? Read 5 Analytics, BI, Data Management Trends For 2015. ]

The entire Microsoft BI platform, for one, is the combination of free services (Analysis Services, Reporting Services, and Integration Services) included with Microsoft SQL Server together with SharePoint and free plug-ins -- PowerPivot and Power View -- for Microsoft Excel. MicroStrategy has introduced free introductory versions of its Visual Insight, Mobile, and cloud offerings. SAP has free ways to use Lumira data-visualization software. And IBM offered free access to Cognos Insight before introducing the freemium-model IBM Watson Analytics service last year.

Tableau's free option is Tableau Public. Ajenstat said that more than 70,000 users have registered, downloaded the vendor's desktop software, and posted data visualizations on the Tableau Public site -- generating more than 300 million views in the process. But Tableau is about more than visualization, he insisted.

"We help people answer deeper questions from their data by making it easier for them to interact with their data and ask new questions," he said. "You discover the real meanings in data by testing hypotheses and chasing hunches. It's not just about showing an end result; you have to show how you got there and defend your answer."

This point gets to the difference between having a tool and drawing correct conclusions from your data. Ajenstat's it's-about-the-process view trusts that people are intelligent, have a basic understanding of their data, and will explore and rigorously test their hypotheses.

A Tableau Public visualization of New York City graffiti data.
A Tableau Public visualization of New York City graffiti data.

There's also a possibility that people will bring biases to an analysis, that they may not consider all relevant possibilities, and that they may not have enough information. Say a company wants to get to the bottom of a shortfall in sales. A marketing leader might look at the data and argue for a larger marketing budget, noting the sparse pipeline. A sales leader might conclude that more salespeople are needed, seeing low engagement and close stats compared to available opportunities.

"Both of those could be great insights related to the problem, but they may not be the cause of the problem," said Marcus Hearne, a business unit executive leader of IBM Watson Analytics, in an interview with InformationWeek. "It could be that a competitor suddenly cut prices by 50%, causing customers to not really take any interest in the company's products."

Making the analysis intelligent is just as important as making analytics accessible and easy to use, said Hearne. IBM Watson Analytics does this, he said, first by surfacing many possible interpretations of data in a ribbon of

Doug Henschen is Executive Editor of InformationWeek, where he covers the intersection of enterprise applications with information management, business intelligence, big data and analytics. He previously served as editor in chief of Intelligent Enterprise, editor in chief of ... View Full Bio
Previous
1 of 2
Next
Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
Page 1 / 2   >   >>
donaldamaccormick
50%
50%
donaldamaccormick,
User Rank: Apprentice
2/9/2015 | 12:56:46 PM
The data-discovery fallacy
Doug, you hit the nail on the head when you say "Organizations have to realize that arming everybody with tools ... won't necessarily unleash a wave of data-driven decisions or, more importantly, correct decisions."

The current cry of "everyone is an analyst these days", is very wide of the mark.

Data-discovery is a key BI technology, and in the hands of experts and power users adds huge value in organization, However simply passing on the same tools to operational end users (who make up the vast majority of an organization) is neither productive or advisable.

Data can help almost everyone in an organization do a better, more productive job. However, expecting them to dig for it themselves is the wrong way to go about it.

I always fall back to two questions :-

1) What do you want your sales teams doing a) selling or b) analysing data ?I have yet to hear the answer "b".

2) When you want a weather forecast (which is a big data, predictive problem) what do you do a) get one with a couple of clicks from a no-training-required web site or b) fire up your data-discovery tool and point it at your weather data lake? Again I have never hears the answer "b" (except from meteorologists, but they are the analysts in this version of the story)

A lot of this comes back to the hi-jacking of the term "self-service". I would (strongly) argue that getting a weather forecast from your favourite web site is self-service information delivery, but there is no hint of a self-service BI tool in sight.

In fact, if you think about it, self-service in everyday life means the opposite to what it does in BI. After all if you go to a self-service restaurant, you are not shown into the kitchen to cook your food, instead you choose from a convenient selection of pre-prepared offerings.

BI teams have a lot to learn about end-user BI from the fast food industry :-)

 

 
jries921
50%
50%
jries921,
User Rank: Ninja
2/7/2015 | 3:23:20 PM
No, but...
Not everyone needs to be a mathematician, computer programmer, scientist, auto mechanic, musician, seamster/seamstress, or any number of other professional specialties either, but all of the above and many more are useful skills to have, even if one has no hope of attaining professional-level competence in any of them.

 
Joe Stanganelli
100%
0%
Joe Stanganelli,
User Rank: Ninja
2/3/2015 | 2:53:39 AM
Re: Simple answer
@Susan: Reminds me of Moneyball (one of my favorite books ever).  An entire culture infused with "old thinking" by the same people shuffled around from organization to organization regardless of actual performance, threatened and disrupted by people who actually thought to look at the numbers.

One of my favorite anecdotes from that book involves a telling of a sabremetrician's analysis of how a stadium remodel will impact a particular team (I think it was the Houston Astros, but I forget because it's been so long since I've read the book).  His findings: "Sorry, if you make these changes, you lose more games."  His study was highly controversial in the organization and was buried, instead of acted upon -- because upper management had already decided that they wanted to make the proposed stadium changes so that there would be more homeruns and thus cause more excitement for fans.
Li Tan
50%
50%
Li Tan,
User Rank: Ninja
2/3/2015 | 2:43:54 AM
Re: Priorities
The basic requirement for data analytics is that it can reveal valuable result. The tool should be accessible and easy to use but all these would be vain if there is no valuable result available.
Susan_Nunziata
50%
50%
Susan_Nunziata,
User Rank: Strategist
1/31/2015 | 10:41:05 PM
Re: Simple answer
@pfretty949: Developing the kind of analytical culture you're talking about is easier said than done. I've seen many bad business decisions made by people who felt they had the "right" data to support their preconceived notions without recognizing that they were injecting their own biases. Or, perhaps they DID recognize this and it was precisely the point. Nobody likes to hear that the data doesn't support their goals or strategies. A lot of education will be needed to create a truly sustainable analytical culture in any organization.
shamika
50%
50%
shamika,
User Rank: Ninja
1/31/2015 | 9:22:53 AM
Re: Simple answer
@Henschen, I agree with you. It is a good idea to have their analysis based on the individual business units. It will add more value than centralizing everything.
shamika
50%
50%
shamika,
User Rank: Ninja
1/31/2015 | 9:08:10 AM
Re: Priorities
"Making the analysis intelligent is just as important as making analytics accessible and easy to use". I like this statement. Most analysts pay less attention this aspect.
Joe Stanganelli
IW Pick
100%
0%
Joe Stanganelli,
User Rank: Ninja
1/29/2015 | 9:50:31 PM
Priorities
Mamet put it best: "The boat needs to look like a boat.  The sail doesn't need to look like a boat."
D. Henschen
50%
50%
D. Henschen,
User Rank: Author
1/29/2015 | 6:04:49 PM
Re: Simple answer
That "analytical culture" is the really hard part. In my view it can't be purely bottom up or top down. Collaboration is key. Facebook has talked about a blended structure, with analytics leaders embedded within lines of business but also reporting up to a centralized analytics leader, so the organization avoids duplicative projects and everyone can learn from successes and failures pursued by inidividual business units. Analytics leaders in business units could help teams and individuals with best practices for dashboards and reports and making available trusted data.
asksqn
50%
50%
asksqn,
User Rank: Ninja
1/29/2015 | 5:34:25 PM
Not a requirement but valuable, nonetheless
Not necessary to be a data analyst, but certainly very helpful in the ability to parse/interpretet the data. 
Page 1 / 2   >   >>
8 Steps to Modern Service Management
8 Steps to Modern Service Management
ITSM as we know it is dead. SaaS helped kill it, and CIOs should be thankful. Hereís what comes next.
Register for InformationWeek Newsletters
White Papers
Current Issue
Top IT Trends to Watch in Financial Services
IT pros at banks, investment houses, insurance companies, and other financial services organizations are focused on a range of issues, from peer-to-peer lending to cybersecurity to performance, agility, and compliance. It all matters.
Video
Slideshows
Twitter Feed
InformationWeek Radio
Archived InformationWeek Radio
Join us for a roundup of the top stories on InformationWeek.com for the week of August 14, 2016. We'll be talking with the InformationWeek.com editors and correspondents who brought you the top stories of the week to get the "story behind the story."
Sponsored Live Streaming Video
Everything You've Been Told About Mobility Is Wrong
Attend this video symposium with Sean Wisdom, Global Director of Mobility Solutions, and learn about how you can harness powerful new products to mobilize your business potential.