I am not sure if I disagree with the central thesis of the article, but the argumentation seems rather weak in many places, for example he says:
Let's take the example of the number of people who walk into a store and buy a certain product. By studying 10 customers you can get an estimate of the probability of the purchase and predict sales; by studying 100 customers you get a better estimate. By analyzing 1000 customers you have a very good estimate. You could study a million customers, but the results are unlikely to be vastly different from the results of analyzing 1000 customers and probably not worth the expense.
This makes sense only for very low-dimensional data sets, but most real-life problems are very high-dimensional and in high-dimensional spaces a million customers (or any other number of samples you might reasonably expect to gather) might still very well be only a small sample. So, there are situations where because of various constraints you will never be able to get to this level where more data stops yielding better results. You can watch a nice presentation from Peter Norvig about the unreasonable effectiveness of data:
That specific point stuck out as incorrect to me as well. And given that it's one of the only technical points in the article, doesn't really inspire confidence in the author. That said, however, I do tend to agree with the overall hypothesis that the value add of "big data" is very marginal for most established industries, for exactly the reason he mentions: they've all been doing this for 20 years now. It's just become cheaper and easier recently. There are certain domains where modern data techniques can have a transformative effect, he mentions one (law). I don't think the data analytics solutions and companies that are being talked about are capable of generating these kind of transformative innovations though; so, I think the author is absolutely correct to say that the big data hype is overblown. There _will_ be big opportunities for big data, but they will come from disruptive startups finding new ways of doing old jobs. Not from crunching a few more GB of your data for a few less bucks.
Good point. Also in most situations data collection goes far beyond 1,000 customers at little to no additional cost. This doesn't necessarily break the author's point though since if there was tremendous value to this data, programs like Google Analytics wouldn't be free. The fact is that most of this data is worthless. You get most of your value from one or two data points that you chose to quantify a qualitative goal.
My perspective comes from my first market research job where I was hired to "make sense" of Nielsen's data for a TV series. It was a cop-out on the company's part - a trained high schooler could have done what I did, organizing thousands of numbers to recognize the 4-5 key takeaways to present back to them. I think the intimidation of large sets of data often tricks people into thinking they're more valuable then they are, and as data collection intensifies, so does this fallacy.
This article seems to be discussing a very different 'Big Data' to that I see. I'm sure the author's points are true in their world, but largely they aren't in mine (mostly online media stuff). Here's a rebuttal of a sort:
This data collection is a recent phenomenon: only since media moved online has it been cheap enough to collect sufficient data to do interesting things. Furthermore, you can actually use that data is more meaningful ways (e.g. recommendation systems) that weren't possible before.
That value is not being extracted by the companies that do currently collect data: surprisingly true from my discussions with media companies. The largest companies (Yahoo!, Netflix, etc.) sure do, but the second tier don't.
That companies will need outside help to extract insights: This perhaps indicative of the biggest difference between our worlds. It seems the author is talking about batch jobs, where you do some rudimentary analysis on a data, makes some slides, present to management, and go home. The people I know are talking about deploying automated systems into production (think recommendation systems again). This requires a large array of skills (machine learning, software engineering, distributed systems, etc.) that many companies don't have. "Minitab, SAS, SPSS, Systat" are not sufficient tools in this world.
The insights gathered from ever larger data sets have more value and are more accurate than insights gathered from smaller data sets: The conventional wisdom in machine learning is that more data trumps more sophisticated algorithms. E.g. http://anand.typepad.com/datawocky/2008/03/more-data-usual.h...
Unstructured and cross functional data have huge value waiting to be extracted: Here we agree!
Perhaps, but many of the basic warnings still apply.
In industry and web businesses alike I see the biggest issue is not capturing or storing the data; it's one of picking up the tools and data already at their disposal and using them to help customers. That's not a gap outside consultants are going to help a lot with as it is mainly a cultural thing, and smart web businesses have that from the start.
Glad to see somebody paint the counter-point to all the hype. I don't believe that there is no substance to the 'big data story' but these points hold up well.
big corp + generic big data + corporate management = ???
Established corporations may be too sluggish to incorporate the insights from the data they are sitting on. Smaller companies will be able to do magical things with machine learning techniques, and will have the agility to act on their insights.
Thus, it is not so much that big data doesn't have value, but that you need a whole management-fat-head-free organizational structure to take advantage of the value.
A big reason companies need outside help is because they don't share data and results - if something works at company A and saves them a lot of money, that information isn't passed to company B and thus you have everyone trying to re-invent the wheel.
A third party that works with many companies builds up a vocabulary of techniques, data sets, and experience that allow them to enter a new environment and be 10 times as effective as any internal team of analysts.
Don't know if anyone has heard of Lattice Engines (http://www.lattice-engines.com/) but they are today integrating external data with internal cross-functional data to extract insights that are saving large companies (that I'm sure already have teams of very talented analysts) millions of dollars.
10 comments
[ 3.4 ms ] story [ 36.5 ms ] threadLet's take the example of the number of people who walk into a store and buy a certain product. By studying 10 customers you can get an estimate of the probability of the purchase and predict sales; by studying 100 customers you get a better estimate. By analyzing 1000 customers you have a very good estimate. You could study a million customers, but the results are unlikely to be vastly different from the results of analyzing 1000 customers and probably not worth the expense.
This makes sense only for very low-dimensional data sets, but most real-life problems are very high-dimensional and in high-dimensional spaces a million customers (or any other number of samples you might reasonably expect to gather) might still very well be only a small sample. So, there are situations where because of various constraints you will never be able to get to this level where more data stops yielding better results. You can watch a nice presentation from Peter Norvig about the unreasonable effectiveness of data:
http://www.youtube.com/watch?v=yvDCzhbjYWs
Or read about the well-known curse of dimensionality:
http://en.wikipedia.org/wiki/Curse_of_dimensionality
My perspective comes from my first market research job where I was hired to "make sense" of Nielsen's data for a TV series. It was a cop-out on the company's part - a trained high schooler could have done what I did, organizing thousands of numbers to recognize the 4-5 key takeaways to present back to them. I think the intimidation of large sets of data often tricks people into thinking they're more valuable then they are, and as data collection intensifies, so does this fallacy.
This data collection is a recent phenomenon: only since media moved online has it been cheap enough to collect sufficient data to do interesting things. Furthermore, you can actually use that data is more meaningful ways (e.g. recommendation systems) that weren't possible before.
That value is not being extracted by the companies that do currently collect data: surprisingly true from my discussions with media companies. The largest companies (Yahoo!, Netflix, etc.) sure do, but the second tier don't.
That companies will need outside help to extract insights: This perhaps indicative of the biggest difference between our worlds. It seems the author is talking about batch jobs, where you do some rudimentary analysis on a data, makes some slides, present to management, and go home. The people I know are talking about deploying automated systems into production (think recommendation systems again). This requires a large array of skills (machine learning, software engineering, distributed systems, etc.) that many companies don't have. "Minitab, SAS, SPSS, Systat" are not sufficient tools in this world.
The insights gathered from ever larger data sets have more value and are more accurate than insights gathered from smaller data sets: The conventional wisdom in machine learning is that more data trumps more sophisticated algorithms. E.g. http://anand.typepad.com/datawocky/2008/03/more-data-usual.h...
Unstructured and cross functional data have huge value waiting to be extracted: Here we agree!
big corp + generic big data + corporate management = ???
Established corporations may be too sluggish to incorporate the insights from the data they are sitting on. Smaller companies will be able to do magical things with machine learning techniques, and will have the agility to act on their insights.
Thus, it is not so much that big data doesn't have value, but that you need a whole management-fat-head-free organizational structure to take advantage of the value.
- big data is overhyped
- potential benefits are likely smaller than you think (this is true for most new technologies)
BUT:
- he seems unaware about modern analysis tools and their power
- state of the art in machine learning has actually advanced a lot in the last 20 years
- as a result a lot of problems that were intractable back then are being solved (Jeopardy, Go)
- there are completely new problems (all the social stuff, huge recommender systems)
- big companies are not good at innovating
- outsourcing is MUCH harder than it seems, especially to Bangalore
A third party that works with many companies builds up a vocabulary of techniques, data sets, and experience that allow them to enter a new environment and be 10 times as effective as any internal team of analysts.
Don't know if anyone has heard of Lattice Engines (http://www.lattice-engines.com/) but they are today integrating external data with internal cross-functional data to extract insights that are saving large companies (that I'm sure already have teams of very talented analysts) millions of dollars.