Best Uses of Big Data in Marketing
Of all of its applications, Big Data’s potential and actual benefits are perhaps most readily seen in marketing. Marketing, as defined by the American Marketing Association, is defined as:
“Marketing is the activity, set of institutions, and processes for creating, communicating, delivering, and exchanging offerings that have value for customers, clients, partners, and society at large.”
With this definition, it’s easy to see the linkages between marketing and Big Data. Information can yield product insights allowing you to create products people want. It helps you understand how to effectively communicate the value of those products. You can optimize your distribution (and production) strategies to deliver your product to your consumer, and you can determine the appropriate rate of exchange (price) to ensure a healthy profit. In sum, the more information you have the more informed marketing decisions you can make.
Clearly, there is great potential between the concept and the process. But in practice, how is Big Data used? And how should it be best be used to create value and achieve a firm’s strategic marketing objectives?
In this article, we will explore 1) the benefits of using Big Data in marketing; 2) marketing planning using Big Data; 3) Big Data and its impact on the four Ps; 4) Big Data and digital marketing; 5) ROI and assessment; 6) using Big Data to build and strengthen brand loyalty; 7) the future of Big Data in marketing; and 8) a marketing case study of a firm using Big Data.
BENEFITS OF USING BIG DATA IN MARKETING
Big Data’s benefits for marketers are numerous. Harnessing the information available from a far greater number of sources than ever before, as a marketer you can:
- Create a more accurate profile of your target consumer(s);
- Predict consumer reaction to marketing messages and product offerings;
- Personalize those marketing messages and product offerings;
- Optimize your production and distribution strategy;
- Create and use more accurate assessment measures;
- Perfect digital marketing and campaign-based strategies;
- Retain more customers less expensively; and
- Obtain product insights, among other tactics.
This is not meant to be an exhaustive list, but the marriage of Big Data and marketing certainly does enhance long-held marketing capabilities and give rise to an impressive set of new ones.
Big Data, Big Opportunities for Marketing
MARKETING PLANNING USING BIG DATA
Assuming a marketer has the appropriate IT hardware/software infrastructure in place and personnel to manage it (whether in-house or outsourced), they will want to begin to work with the data analysts to explore the data itself. They will need to know what they should be gathering, and which tools they are going to use. These are critical questions to begin to handle the avalanche of data available. They also need to determine in which areas they will be gathering and linking datasets in order to determine what questions they should be asking. They should start in areas that will give them a distinct competitive advantage, and actionable insights.
The more insights incorporated in marketing plans, the more effective those plans are likely to be. Further, marketers can harness predictive analytics – a series of statistical techniques and modeling methods to forecast future occurrences using historical data. This can yield powerful insights about consumer preferences and probability of purchase. Predictive analytics is bug business. According to technology research firm IDC, the market for predictive analytics technology is projected to grow from $2.2 billion to $3.4 billion by 2018.
They can also harness machine learning in the analyses of problems. Using smart computer software to handle some types of analyses saves time and money as computers can compute and model faster than humans can. Many firms, using proprietary algorithms, offer their services to marketers/market research department.
BIG DATA AND THE FOUR Ps
The use of Big Data has implications for every aspect of marketing. Marketing is often described in terms of the four Ps: promotion, product, place, and price. Some marketers /marketing professors add a fifth P: packaging. Big Data can help hone marketers’ understanding of consumer preferences to design the kind of packaging that would attract consumers and more readily lead to a sale. But in terms of the more conventional definition, first, let’s start with pricing.
By incorporating various real-time datasets, including supplier and inventory data, models of consumer likelihood to purchase, and financial forecasts, firms can employ dynamic pricing – allowing it to offer different prices at different times in different places to different consumers in order to optimize revenue. For example, hotel chains may offer one standard rate on their website on a particular day. They might also offer it as a part of package deals with strategic partners such as rental car companies or airlines. The hotel component of the price may be at a markup or a discount depending on the forecasted price. As the hotel begins to fill up, the hotel’s management can increase the price of the room, both as a standalone room, and as part of the package deal(s), as the room is both in demand and in short supply. If the demand plateaus shortly before the check-in time, hotel management can deeply discount it and offer it on discount hotel websites.
Further, the firm can vary the pricing based on the consumer, or characteristics of the consumer. For example, an examination of data might yield that owners of a certain type for smartphone or those using a certain type of browser, are more likely to make hotel reservations independent of prices. The hotel can then increase the prices for those who access their site with said smartphone or browser. This is not a purely hypothetical example. Many hotels and hotel chains, such as Fairmont Raffles Hotels International, currently use dynamic pricing to optimize revenue. And hotels aren’t the only industry to adapt this strategy. Retailers, such as Amazon and Walmart, among others are well-known for doing so. Many well-known technology firms have incorporated dynamic pricing as well.
One of Big Data’s most common uses is to obtain product insights. Firms can easily conduct qualitative and quantitative market research online at a much lower cost than two decades ago. Online survey tools and videoconferencing tools make focus groups and surveys with large sample sizes much easier to conduct as well. Firms can monitor the web and social media for mentions of their brand by consumers. They can review the analytics for their own digital assets (website, microsite(s), blog, social media, and plenty of third-party signals to gain actionable insights. And they can also interpret that information to figure out ideal product extensions.
Marketers can use Big Data to determine the optimal channels to place their products. They can set up supply chains accordingly. In some cases, changes in placement are borne of necessity. Many newspapers, facing years of declining advertising and subscription revenue, have chosen to either go completely online and erect a paywall, or downscale print circulation and incorporate a free or paid online subscription model. Determining which option, and in the case of the latter, how to downscale (i.e. reduced subscription delivery dates, reduced retail distribution) can be determined in part by a careful analysis of Big Data. Possible variables might include areas in which subscriptions are declining; retail outlets where sales have either stagnated or declined any test of the effect of a paywall on existing subscribers, and more.
Big Data can help consumers pinpoint the most likely targets to buy product, by allowing marketers to craft a more complete profile of their average customer. They can easily test the marketing messages that work and adjust quickly midstream if they don’t. In the hands of a capable marketing research firm or in-house marketing department, Big Data can be harnessed to test and predict likely consumer reaction to various marketing messages. For example, a firm sending out a bulk email to 250,000 consumers can use marketing data to create a psychographic profile of the average consumer, extrapolate their motivations, and write copy that speaks to them. They can test possible headlines with small sample sizes to see how many people opened each, and send out only those with high open rates. They can also use open rates and click through rates to qualify leads and move them further along the sales funnel. They can then compare this to actual sales from these consumers, as well as social media mentions, in-person visits, and many other pieces of data.
BIG DATA AND DIGITAL MARKETING
The Internet is a major source of Big Data. Everything from website analytics to social media mentions to click-through ad rates can be easily aggregated, analyzed, and interpreted. It has also led to the creation of many new forms of online or digital marketing. A complete exploration of them is beyond the scope of this post, but they include (and are not limited to):
- Banner advertising: acquiring digital ad space on third-party websites that drives traffic to one of your digital assets
- Search engine marketing: using a combination of digital advertising, content marketing, SEO, and other strategies to increase a particular brand’s visibility in search engine rankings
- Content marketing: brand, product, or search engine marketing through text, audio, photo, and video
- Brand storytelling: a form of digital marketing which emphasizes the brand promise through text, audio, photo, and video (e.g. photos of customers happily using the product
- Search engine optimization: ensuring that your digital assets are properly coded to rank high in search engine results
- Retargeting advertising: cookie-based advertising strategy that displays a firm’s ads to its site visitors on third-party websites
- Social media marketing: brand or product marketing using social networks and social media tools
- Mobile advertising: advertising on mobile devices such as smartphones and tablets
- Native advertising: advertising that appears similar to organic content on the website on which it appears (e.g. Facebook ads that appear like status updates)
Digital marketing various sub-disciplines all collect and aggregate data which can be analyzed for patterns and insights. Further, these insights can be rapidly integrated into any digital marketing strategy. For example, an insight that the majority of online visitors to a firm’s digital assets are looking for a small set of content with common keywords, can lead digital marketers to develop new content with those keywords on all platforms. They can also look at that content in light of traditional marketing messages, product lines, and other existing strategies, and refine them accordingly.
ROI AND ASSESSMENT
Digital marketing has allowed marketers to quantify their efforts and quickly determine their success or failure. Big Data offers marketers more dimensions along which to assess that success of failure. For example, marketers looking to launch a brand awareness firm had little in the way of measurement two decades ago. Today, they can assess everything from social media mentions and likes to website visits and even sales. This is heartening for CEOs and executives who, pre-Internet often waded through murky metrics of success. Even MBA and undergraduate business programs are increasingly offering quantitative marketing programs which teach students to leverage insights from Big Data to refine and optimize marketing strategies.
BUILDING BRAND LOYALTY
By collecting and aggregating so much information about consumers, marketers are now able to respond to individual consumers in a very personal fashion. By employing tactics that appeal to one’s fundamental motivations, preferences, experiences, and emotions, marketers can enable them to create a strong and lasting connection between customer and brand. For example, firms that aggregate customer service calls might take note that a customer who contacted a call enter has a kindergarten-age daughter, and might incorporate that in their marketing strategy directed at her. A follow up email from the associate with whom she spoke might acknowledge the conversation and suggest products that might suite her daughter; this could also be incorporated into the firm’s recommendation engine for her user profile.
In order to deliver on the promise of personalized services, marketers need as much information as possible about consumers and customers. This is where Big Data comes in – providing targeted customers with information important to them.
FUTURE OF BIG DATA IN MARKETING
Industry pundits predict that Big Data will assume an ever-more central role in marketing as machine learning evolves and allows data scientists to analyze disparate data types ever quicker. Others predict deeper customization of product and personalization of services; others see the increasing rapidity with which marketing messages are developed becoming increasingly important revenue drivers. Ethics are, and will increase in, importance the more data is gathered. The more firms gather large datasets, the more government regulation will likely grow, especially in certain areas like privacy and security. However, many believe that while the tools are increasing in refinement and sophistication, the basic definition remains the same: “creating, communicating, delivering, and exchanging offerings that have value for customers, clients, partners, and society at large.” As one marketing expert, Bob Borchers, the chief marketing officer of Dolby Laboratories, said in a recent interview with Fortune:
“Big data really isn’t the end unto itself,” he said. “It’s actually big insights from big data. It’s throwing away 99.999% of that data to find things that are actionable.”
Marketers have been chasing those insights since the pre-computer days. With Big Data, they have reams and reams of data with which to do it.
Big Business: Unlocking Value from Big Data with Analytics
One of the most well-known examples of Big Data in marketing is Netflix. An upstart video rental company competing against the likes of then-market leader Blockbuster video, Netflix began as a mail-order DVD service in 1997. Its website featured a sophisticated recommendation engine, an algorithm-based program that predicts consumer video preferences based on their past choices and many other bits of customer data available from its subscribers. For Netflix, which also employed a flat price for unlimited DVDs, this meant creating a tremendous value for their consumers. Beyond being able to undercut Blockbuster for multiple movies, they were able to develop a deeply engaged and loyal customer base by analyzing their customer’s signals, continuously refining their algorithms and coming up with increasingly accurate predictions of what their customers would want to view next. Further, Netflix’s forays into original programming, starting in 2012, were done using considerable analysis of the data they had captured during their ten years in business. This included everything from viewer records to ratings to comments to meta tags. Today, Netflix counts more than 44 million paying members, and Blockbuster, which filed for bankruptcy in 2010, is a faint memory, with most retail locations closed and a brand name now only used for an on-demand cable channel.