Behaviour in Big Data: the smarter way to target your customers

In Blog article by Freethinking Business Consultants

By Rob Smith, Data Analyst at Freethinking

It’s barely 6 am, and I’ve already received three SMSes letting me know I ‘qualify’. If the messages themselves weren’t so blatantly spammy, I’d probably let this good news permeate my almost-conscious state, and carry it forward with me throughout the day.

Why? Because I qualify. The downside is what I qualify for, though. At present, I don’t need a cellphone contract, or a loan, or an extended credit term on a couch. In fact, I don’t know when last I would’ve expressed interest in any of the above – promptly turning these texts into a dull irritation.

On the other side of these urgent messages is a person – sadly, an inexorable one wielding a budget, and one so lucky we should never meet. The truth is that if I actually needed any of those things I’m on the spam-list for, I probably wouldn’t mind; if these things mattered to me, by all means, let me know I qualify… but I don’t, and they don’t.

Marketers continue to pester ordinary citizens with endless streams of offers, enduring low conversion rates, due to a lack of insightful data on potential customers. Instead, we cling to outdated methods of targeting and segmenting our audiences – which is largely based on demographics.

Agencies and marketing departments continue to operate in these conditions.  They know various things about me – approximate age, race or ethnicity, the city or suburb I live in, and they could maybe even make a few educated guesses – and use these tiny glimpses into my world as reason enough to reach out to me about their products and services.

Sure, because of my age there may be certain things I’m more likely to be interested in or make use of – but that’s quite the stretch, and frankly, a giant waste of marketing spend. So, what can be done to spend better – for the sake of my sanity (and the careers of reckless advertisers)?

The answer lies in this simple fact: People are what they do; not what they look like.

“People are what they do; not what they look like.”

Finding the treasures within your data

In reality, few direct marketers and third-party agencies have access to the kind of behavioural data that would enable reaching out based on activity – that is, outside of what Facebook happens to track about its users.

But for many organisations looking to reach out to existing customers (to upsell, convert or cross-sell), a wealth of information already exists in the form of transactional logs, CRM databases, and everywhere else user interactions are stored.  It merely requires some connecting of the dots, so-to-speak, by using a big data tool.

Now “Big Data” has received a lot of attention lately, partly due to its immediately evident stature, but is, in many circles, in danger of falling into the “buzz word we never quite understood so completely missed out on” category.

What makes it so powerful from a business perspective is that it can be used to mesh together many varied data sources – such as transaction logs and CRM data – to build extremely comprehensive views and understanding of customers and their behaviours, uncovering previously unknown correlations and trends.

All of these different data sources are valuable on their own but tied together using a unique identifier for each customer, and they become precision tools for driving growth, retention, and overall customer happiness – all the warm and fuzzy things one wouldn’t expect machines to offer.

This allows marketers to refine customer targeting: instead of placing bets with marketing spend on demographics alone, companies can shift that focus and budget to finding customers who exhibit behaviours likely to indicate interest or intent to purchase and match additional demographic criteria.

This kind of targeting, beyond having the potential to reach and engage more qualified customers, has the pleasant side-effect of appearing less like spam than the traditional approach with a much larger audience, but a more tenuous connection to a user’s needs and behaviour.

Building your Big Data practice

But having the data is just the starting point. To start truly deriving insights and value from your information, there are some foundations we must address:

Firstly, it’s critical to construct an Analysis Framework. Understanding objectives, what questions need to be asked, and how to extract the answers (and from where) in a repeatable and ultimately reliable way should become the foundation for how data is stored, processed, and made accessible.

The Framework should highlight the need for unique keys that tie individual customers to their data. Without the keys, there is no unlocking the bulk of the value embedded in the data.

From there on out, it’s important to continually identify business goals and related behaviours that lead to conversions – before combining them with demographics to identify high-quality, well-suited customers for specific products/services.

This ‘continuous improvement’ philosophy helps to ensure your targeting is always current, relevant, and value-adding to the business – driving sustained ROI improvement over time.