Ever look out at the ocean on a cloudy day? The huge gray mass above stretches out to meet the darker gray mass below at a black line on the horizon?
Standing on that beach, some people feel the ocean’s irresistible allure and comforting power. Others feel like they’re being sucked between two insatiable plates that will crush them at that line in the darkness.
An ocean on a cloudy day is an apt comparison for Big Data and metadata. Big Data stretches its expanding, roiling clouds of content over an equally roiling sea of metadata. Both are massive and powerful. They can both be threatening.
The desire to mine Big Data is making billionaires out of “mining equipment companies,” and references to their algorithms, claims of superior computing speed and boasts of expansive storage capacity are everywhere. Big Data is big content, and that content is getting bigger exponentially. How do we find what we need and want? The answer to that question is to be found in marketing research. A company’s marketing research team will develop expertise in web analytics in addition to what they already know about market analytics. They will need to incorporate more and more disciplines to turn data into information, information into knowledge and knowledge into wisdom.
Once one begins to get a handle on Big Data—or at least has a plan on how to handle it—he or she faces that almost surreal world of metadata. From the murky world of spying, the world learned there is useful information that is with the content but is not the content. “Metadata is the ‘data about data’, or the data that can be taken from an individual piece of content,” says Emma Battle in a blog for Success 360.[1]
In 2010, Raffi Kirkovian, a Twitter employee, published a “Map of a Twitter Status Object” that identifies 37 discrete pieces of information contained in a Tweet other than the actual content of the tweet.[2]
Four years later that seems to have grown, “At 140 characters a tweet seems tiny, but it can yield a wealth of information. According to Elasticsearch, a startup that builds software to help companies mine data from social media, there are 150 separate points of so-called metadata in an individual tweet,” says Elizabeth Dwoskin in a Wall Street Journal blog.
For marketing researchers this can be a bonanza, “A marketer can look at tweets sent by their target audience and see that the majority of the tweets have times stamped after 5:00 p.m. The marketer can then conclude that the best time to reach their target audience on Twitter may be after 5:00 p.m.,” says Battle.
How do marketing professionals go from data to decisions? Through interpretation. The data that is collected and analyzed “is used to enable the team to identify patterns, draw conclusions, solve the research problem, and achieve the research objectives,” according to SMstudy® Guide – Marketing Research, a book in the SMstudy® Guide series on sales and marketing.[3]
The Guide recommends that data interpretation start with three important inputs: the analyzed data, the research problem and objectives. During the interpretation process, “findings from the research analysis are compiled and reported to the marketing team and senior management and are ultimately used to inform marketing and business decisions.” In deciding what to compile and what to report, the researcher will rely on the research problem and objectives because they “provide a focused and definite direction to the data interpretation process,” according to the SMstudy® Guide.
With focus and direction, the marketing researcher uses three categories of tools to identify patterns and draw conclusions that will meet their company’s or client’s needs: tables, charts and expert judgment. Tables such as spreadsheets by Microsoft and Google help researchers organize large amounts of data. Some, like Microsoft’s Excel, provide a variety of filters and grouping tools for this purpose.
There are thousands of charts available to the market researcher. When one uses the term “chart” to be a category name that includes diagrams and graphs, the number of methods for visually displaying often complex relationships explodes. The SMstudy® Guide highlights bar charts, stratum charts, pictograms and cartograms for their usefulness and broad-based familiarity.
Once one has an excellent collection of tables and charts, something is still needed to make complete sense of them all: expert judgment. “The ability to appropriately interpret the data develops with experience. Inexperienced researchers can sometimes interpret data in a preferred way because of their comfort level with a given method. A researcher should try to seek the opinions of industry experts and research experts, who can provide valuable inputs in choosing the best way to interpret data within the given constraints,” says SMstudy® Guide’s Marketing Research book.
When relevant inputs are processed with appropriate tools, the researcher draws conclusions that are used to solve the research problem and inform marketing decisions. In short, accurately interpreted research means you know the problem AND the best solution options. And knowing is a great feeling between the clouds and the ocean.
[1] Battle, Emma. (7/23/14) “Metadata, Mega Data or Big Data What’s in It for Marketers” Success 360. Retrieved on 4/21/16 from www.success360i.com/metadata-mega-data-or-big-data-whats-in-it-for-marketers/
[2] April 18, 2010 Raffi Kirkorian published a “Map of a Twitter Status Object” http://online.wsj.com/public/resources/documents/TweetMetadata.pdf
[3] For more information about the SMstudy® Guide please, visit http://www.smstudy.com/SMBOKGuide/overview-of-SMstudy-guide
