What Businesses Need to Understand About Big (and Small) Data

  by    0   0

What is “big data” and why should it matter to marketers?

I recently posited the question, “Do marketers know what ‘big data’ is?” at a networking event for marketers, and the response I got was unexpected: “Do they really need to?”

I was taken aback, likening the response to questioning whether or not we should include social media strategies in marketing campaigns today.  Marketers today, whether they are aware of it or not, constantly work with data so of course they should know what “big data” is and how it differs from “small data.”

Then I realized that, until recently, there was no WAY for marketers to understand big data. The shift of data availability beyond statisticians and IT managers is a recent development.

The surge in conversation about “big data” is the result of a sequence of events so let’s talk about how “big data” came about.

Up until the invention of the PC, data was a cruel mistress, demanding hand-written calculations, myriad sheets of paper, and a lot of patience to analyze.

The rise of the PC was a critical first step in giving data a home and a way to sort through thousands of items. Users can now access vast amounts of data easily (via cloud computing, computational capabilities, etc.).

As a result of the ability to compute large quantities of information and display the output graphically, the number of channels we use to market our products and services has grown causing an increase in the amount of data we can collect.

Finally, based on availability and technological capability to manipulate and illustrate our data, the business intelligence community has risen along with products to make it easier.

So why does any of this matter to marketers? Data today refers to a set of three characteristics: condition, location, and population.


This refers to the data’s readiness for use. For instance, if you have a list of email addresses that have been confirmed through a Captcha validation system and have opted in to receiving communication from you, then that data is likely ready to use (often called “clean”).

Clean data is small data. Contrast that with a list of purchased email addresses that must be validated as correct, relevant to your organization, and willing to receive messages from you. This data is not well conditioned and requires time and cost to clean, making it “big data.”


This refers to where the data originates and its compatibility with a usable format.

Using our email example, data that lives in an email distribution client such as Marketo or MailChimp has a single location and is compatible with the format that it needs to be sent from, making it “small data.”

Data that requires merging from multiple sources in a variety of formats or with differing variables is “big data.”


This refers to the individuals that have qualities in common to the need in consideration – in this instance, your email list.

A “small data” set includes a known population that is not expected to have changes to its composition in the short term, which allows marketers to use this data to answer a specific question or need.

For example, when looking to market a product, consider a list of users who recently purchased a similar item from you.

Since they purchased a similar item it is safe to say that emailing them is fine. Conversely, “big data” would represent your large purchased email list with unknowns, possible duplications, and unsubscribes. This list cannot be used for targeted email marketing sends in its current form (at least not by a good marketer).

So why should marketers be concerned with data size? Here are a few reasons:

Bad Data = Bad Marketing

You need to know when someone is giving you bad data. Data can be (and is) manipulated to say what people need it to say.

This can be a powerful tool if you are the one controlling the data, but receiving large data sets or outcomes from data can be daunting if you trust others to provide you with accurate information to do your job.

This is not to say that people always willingly provide incorrect data or outcomes – often times they may not know themselves that it is inaccurate. This leads to number two.

Maslow’s Hammer

If you have a hammer in hand, you eventually start to see a nail.

This means that if you are looking for something, you naturally increase your odds in finding it.

Data will always have structures or characteristics that naturally group items together. When working with “big data” it’s important to understand that these groupings may not be the result of anything other than chance, and trends should be determined by testing.

For example, an email campaign may appear to be opened more frequently in one geographic location than another. This could cause a decision to market more heavily in that area, when the grouping of email opens could have, in fact, been random.

Aggregation is Power

Having access to the data your organization gathers means you can unearth powerful information about your users, potential customers, and own capabilities.

What you may discover is a ton of data that needs a better way to be analyzed for your needs, or that you need to be capturing more data from particular sources to round out your set. Either way, knowledge is power so get your data and get going!

Beyond the reasons mentioned above, I know that investing in big data is expensive, and analyzing any data has a cost – whether it’s via purchased tools or expended man-hours.

The pros of big data are many, but one of the most important for marketing professionals is big data’s ability to enable you to discover new, relevant patterns that may not be noticed when using a small data sample.

Much of the data collected by marketers does involve an understanding of evolving behavioral patterns, making small data outdated quickly. Investing in big data and a way to manage it can have a great impact on your organization.

The most important thing to remember is that whether your data is big or small, it’s the correct analysis that counts.

0 thoughts on “What Businesses Need to Understand About Big (and Small) Data

  1. David Boozer says:

    Great info Danny…once again….=)

  2. Danny Brown says:

    Cheers, sir, appreciated!

  3. dbvickery says:

    Actually, you had me going for the first portion of this post. I was wondering where you were going with it…and I was wondering if I would agree. I do not necessarily equate “big data” to “unconditioned data”…or “clean data” to “small data”.
    But I absolutely agree with the latter portion of the post. One of my favorite quotes from a conference last year was “big data is the new algorithm”. Put a data scientist in the room with big data, and that scientist can come up with consistent correlations between attributes of that data – that may have been hidden in plain sight. It also should help make any forecast models more accurate (especially if you have seasonal or year-over-year data). I also agree that the power of aggregation is absolutely a bonus of big data. You now have much better representation of your consumers, or organization, and can get ahead of the curve with product development, customer service response, investment choices (time/talent/money), etc.
    Thought provoking post, Danny.

  4. Danny Brown says:

    dbvickery Damn, if I’ve made you think, mate, then I’ll take that as a huge compliment. 🙂
    Completely agree – I can see a future trend for brands to employ a mix of data scientist and behavioural analyst to make sense of the data, and what it really means from a human angle. Get that info right….

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.