Big Data: The Next Big Leap for Marketing Revolution

It’s the year 2018, May 5th—I am now in Shoprite, one of Africa’s foremost hypermarkets. I am here for my shopping and I thought that because I was a returning customer, the human-sized welcoming robot at the front door might greet me by my first name. It is very impressive! I had learned that robots have special sensors that can swiftly detect a person’s fingerprint and match it against a database of well over 14 million other customers in a split second for precision identification. Using a feminine voice, the robot has become my in-store salesperson, offering me my personalised offer of the week. I am trying to get into the groove of what seems to me to be “robotic selling”. The truth be told, the robot’s special bouquet included a package of ten different attractive offers that reflected my very personal preferences and up until then unexpressed need. For me, this is a surprise bundle as it is 100% true to my taste and liking. My shock is heightened when the robot alerts me to the fact that it is only four days to my son’s birthday and that the weather where I live has fluctuated from 32 to 27 degrees during the last ten days. To my astonishment, the robot adds that my car insurance renewal is due in nine days and that around 67% of my 121 friends connected on Facebook seem to have a similar liking for dogs. The robot does not stop there either as it continues to reveal more information about my shopping habits by saying “Looking back on your past weekend shopping experience and the way you move around the aisles in the supermarket, I have managed to find this wonderful deal for you.’’

I stand there frozen and gobsmacked; so you basically mean that it is all measured and recorded? The way I walk between the aisles and the average time that I spend every time I come shopping? What kind of technology is that? A nano-segmentation or some sort of open-world marketing? I wonder, because I cannot seem to understand how a robot can present me with a fully customised package of offers based on previously collected information. This is more like being under surveillance with cameras and computers tracking my preferences and attitudes to products.

My train of thought is interrupted by the robot advising me to go to Special Offer Table 20 to pick up my own personal special offer. I give in quickly and make my way to the table, trying to figure out how I could pay for this unplanned for and surprising offer which had come from out of the blue. As much as this wonderful offer seems to perfectly match my shopping needs and intimate preferences, I still think that it may cost me an arm and a leg! And again my train of thought is interrupted when the customer service agent at Table 20 welcomes me warmly with another piece of news. He starts “Good afternoon, Sir. It’s my pleasure to inform you that your bank has just approved an interest-free overdraft over a six-month period should you decide to accept our special offer today.’’ What? How come my bank got involved here? And what is more alarming is how they got to make such a deal with my bank that quickly?

I am positive that this is not some sci-fi flick; in fact this is the world we live in today—this is where marketing propositions become a collection of benefits configured as a customised offer with the most memorable branded experience that extends way beyond what television or radio advertising can deliver. I figured that we are witnessing an astounding revolution in marketing. At this moment, rest assured that every single penny spent on marketing, per customer, can be well justified by a measurable return based on lifetime value.

Welcome to the next big thing on how companies create, communicate and deliver superior value. This is the type of business model where unusual, unrelated, diverse and multi-context information acquired from different data sources is gathered to find perfect customized solutions, one customer at a time. Today, the best of Mathematics, Financial modelling, Neuroscience, Machine learning, Atmospheric customer mind-mapping, Geo-analytics, Brand management, etc. are now being carefully integrated into a fully designed intelligence system aimed at achieving optimal customer engagement.

Go back ten years in time and analyse the exuberance of the practice of marketing; budgets were soaring and advertising experts were competing for prime time on TV. Your daily newspaper did not have one single vacant spot without an ad on it. Visual stimulants and catchy tag lines were used to force every message down our throat with a monumental media budget used for creating eye-catching ads that would tap into the customer’s emotions, rather than logic and reasoning. As for those potential clients who did not respond to the usual media bombardment, instead they became ensnared by the promotional baits of buy-one-get-one-free, or BOGOF as it has become known.

At those times, there was a vague application of the famous 7Ps rules of marketing, particularly with the fifth ‘P’ of Performance/Profit being clearly overlooked. In the past there was absolutely no guarantee when it came to ‘Return on Investment’. There were several marketing interventions applied simultaneously and the results were hard to be regressed into a predictable equation showing the weighted impact of each marketing effort. At least if the most influential variable(s) was isolated month-on-month, marketers could focus more on applying what works best in more instances to ensure the consistent performance level needed, rather than operate in haphazard and opportunistic wins. It is therefore not a surprise that John Wanamaker lamented that “half of my marketing spend is wasted and I really don’t know which one it is!”

Now the game has been altered very smartly. The reality of increasing economic pressures and declining margins means that the marketing budgets had better take a big cut. The fierce pricing wars and the quest for cost leadership mean that businesses are now more willing to attract customers with penetration pricing for short-term gains, even when that would mean a subtle de-emphasis on brand building.

The current fragmentation of media has also heightened and worsened the customer targeting quagmire. It is now more difficult to pin down customers through multiple media apertures. There are too many screens and content struggling for their limited share of attention every hour of the day. The digitalization of media has also led to the production of a colossal volume of data. These data are not just numbers, many are unstructured (i.e. voice, text, picture, video) with opportunities for richer and contextual understanding of the underlying drivers of behaviour. It may interest you to know that there are online crawlers that decode the sentiment of every video and sound you view on YouTube, and how long you watched each video for, taking cognisance of actual running time and buffering delays.


But isn’t this the same as marketing research? Actually, no—it isn’t. Research is a report of what I think I know, while Big Data is an unobstructed diagnostic of what I don’t know and I may never know about myself, my buying behaviour, or the influence of my multi-layered social ecosystem. What guarantees can we possibly have that an artisan living on less than N50 per day will speak the truth when engaged in an air-conditioned focus group atmosphere that is miles away from their natural environment? With an expected focus group participation fee, their lying coefficient will take a good ride to give whatever answers that the interviewers would like to hear.

Recent studies have indicated that there is no research that can possibly tell us so much about the customer’s interests than the Facebook pages that they like and the groups/people they are following. There is no smarter and more accurate way to have a better grasp of the customer’s personality or idea of the person they aspire to be like than knowing which individuals they follow on Twitter. Channels that a customer subscribes to on YouTube give greater knowledge and a better understanding of their media preference than thousands of media habit studies. An integrated, real-time customer transactional history, frequency of calls to customer service, query type, services portfolio, anecdotal insight on customer’s locations and emerging contextual knowledge (sensor data, temperature, traffic pattern, retail outlets per square kilometre etc.) have more relevant insight for profitable customer engagement.

Sean Rad’s conceptual truth that “Data beats emotions” makes a lot of sense in this regard, because customer irrational rationality, unusual reasoning, and unexpressed behaviour do provide a richer contextual understanding of attitude when overlaid on a dynamic algorithm to build predictable pattern of profitability. We only need to torture the data long enough and it will confess (Ronald Coase, Economist).

The ability to collect these multiple and diverse data sources and put them all together in an organized, practical and relevant knowledge ecosystem is simply how Big Data works.

Big Data, or Computational Marketing, thus provides a distinct competitive advantage to businesses to leverage both structured and unstructured data in gaining more insight for real-time marketing efficiency, process optimization, customer management and value preservation. The return on investment is delivering quantifiable results, which leads to a mutually profitable scenario for all possible parties involved: the company, the customer and ultimately the shareholders.

The concept of Big Data does not just focus on marketing precision. The principles of accurate targeting can be efficiently applied to several industries. Kudos to great players such as IBM—it leads the application of Big Data in Africa through its recently opened laboratory in Nairobi, Kenya. The mandate is how to use the multi-context intelligence that Big Data provides, to lead the creation of world class cities while addressing social empowerment, healthcare, agricultural innovations, priority education and providing effective public safety.

For instance, with a Community Knowledge Ecosystem, the data collected at hospitals—in regard to the number of out-patients, the treated cases, common diseases and frequent visits—is shared for healthcare planning on a real-time basis. Additionally, this information is shared with insurance companies to provide critical information necessary for accurate premium determination and risk efficiency management. Exercise and diet-related influence can be auto-integrated to calculate the life insurance premium payable. For example, those customers with more exercise milestones get lower premium rates when using Healthcare Influence Analytics. Nike and United Healthcare are rumoured to be collaborating to launch a healthcare solution by 2015, targeted at the $60 billion weight-loss industry and based on Big Data applications. Effective road mapping and city management can be done with the real-time application of images captured by CCTV cameras at various crossings. This will help in identifying the pattern of vehicular movement at different hours and the major causes of traffic jams, thus throwing a spotlight on accident-prone areas and aiding the provision of ready government intervention.

Furthermore, a UN FAO program on Precision Agriculture uses farm sensors for real-time data gathered on weather prediction, soil conditions, crop features and many other data sets. An example of information is presented in the platform, as well as on the iPad and iPhone app Mobile Farm Manager. These assist farmers in figuring out which crops to plant and offer advice on the best farming timing as well. Also, this technique can help the farmer learn about the best return and gains by showing and guiding them along the most suitable path when starting the farming process. The principle here is that if we can aggregate the annual farming experience and perspective of many farmers over a period of time, we can build a crowdsourced knowledge system that can be of benefit to all. For instance, last year, Monsanto paid $930M for Khosla Ventures-backed startup Climate Corporation, which sold automated weather insurance to farmers, after doing the actuarial calculations derived from Big Data weather predictions.

Big Data is not all good news however. We must be aware of the caveat. There is no new innovation that has climbed the stairs to success without hitches and glitches. Data combing is still the biggest challenge out there. So, who covers the cost of this innovation after all? For us, it is like living in a glass house where everyone can have a peek into our lives. It may become so invasive, like phone tapping or paparazzi hacking into personal space. Every customer will become a celebrity and being famous has never been so arduous. Every click you make on the Internet is kept for future referencing and there will be a very fine line between privacy and commercial ethics.

Today, as we share the excitement of the Big Data evolution that is making a huge transformation in our lives, I will probably ask myself a few flippant questions a few years from now, like why did the robot at Shoprite give me that personalized package offer? I have come shopping to explore new things today—I am with my children and they are far from excited about hearing that school reopens in just four days, and why on earth did you have to disclose my son’s birthday? Now my plan for a surprise birthday party just got ruined by that robot. For heaven’s sake, do not embarrass me in front of my family and neighbours by mentioning that I cannot afford to ask for an interest free 6-month overdraft from my bank either! Now I have to apply for a business loan next week. God, when has my life become a stock of public scrutiny? I was really happy struggling with my decisions, rather than standing there watching someone else making them for me – those people who will always have enough to sell and I will never have enough to buy.


It is also worth mentioning that no matter how accurate the analysis is, there will always be lies, damned lies, and statistics—a phrase attributed to Benjamin Disraeli in Mark Twain’s autobiography. As much as data mining will be so important, the ultimate will also be experience and intuition. After all, “what is intuition at its best, but large amounts of data of all kinds, filtered through a human brain, rather than a mathematical model?” (Steve Lohr, New York Times). We must always avoid the risk of drowning in irrelevant information and starving for knowledge.

As much as Big Data comes with its intrigues, we can all comfortably agree with Arthur C Nielsen, the founder of AC. Nielsen that “The price of little light is less than the cost of darkness.”