The Digital Era has brought the world of industry many challenges and opportunities. The irrational exuberance of Nineties technologists and tech investors led to Internet businesses that were unwieldy, unmanageable, and unprofitable.

In the early 2000s after the dotcom bubble burst, many spectators wondered whether the marriage of Internet and business was merely a fad, while many speculators, wondered whether the markets would or could regain confidence in digital businesses.

However, in the decade and a half since the Tech bubble, the world has since seen the creation, growth, and maturation of digital industries, as well as innovative new digital approaches to traditional brick and mortar industries.

Few could have predicted the rise of the online-only business or the virtual company, or that the most successful Internet firms could be quite so successful – with market caps upwards of $400 billion. The Web 2.0. Era has even seen the creation of profitable offline businesses housed in online video games!

One major driver of competitive advantage is predictive analytics – the collection of statistical and computing techniques that allow firms to use historic and dynamic data, aggregated digitally, to create probabilistic models of the likelihood of future events. Predictive analytics has a range of applications, though it is used most commonly to optimize decision-making and to determine consumer preferences. This powerful field has become a cornerstone of the business strategies of many a household name. Indeed, it has even led to the creation of many innovative – and profitable, new business models.

Innovative Business Models Using Predictive Analytics

© | ImageFlow

Here, we will 1) briefly cover traditional business models; and then address 2) how brick-and-mortar firms are enhancing their businesses with predictive analytics; as well as illustrate 3) examples of new business models that incorporate predictive analytics.


Fundamentally, there are several basic business models: owners/landlords, manufacturers, distributors, and sellers. Within these broad categories are a number of more specific models – for example, sellers include wholesalers, brokers, traders, retailers, and multi-level marketers. There is also a variety of conventional pricing, supply chain, marketing, and other business strategies, which are often confused with business models and help differentiate firms from each other. Predictive analytics has upended many of these traditional models and strategies of doing business in myriad ways.

In short, predictive analytics allows firms to create models of consumer behavior that are correlated positively with historical data, and use these models to forecast future results. Because, in our Digital Era, data comes in in real-time and because we have developed highly sophisticated and robust hardware and software systems for processing this “Big Data,” firms can feed data into these models in real-time and adjust their business decisions automatically.

For example, firms can use predictive analytics to drive dynamic pricing, a strategy that predates predictive analytics, but has been enhanced by it significantly since. A hypothetical hotel might determine pricing during a holiday season by building a model involving prior year data, prior month data, and even prior hour data, by automatically feeding all purchase transaction information into a model designed to optimize revenue and changing pricing based on real-time performance. To the consumer, this might look like a pricing schedule that varies as vacancy rates rise or fall. Predictive analytics has greatly enhanced the performance of many brick and mortar establishments.


As digital enterprise began to flourish in the early 2000s, many brick and mortar retailers lost market share and went out of business. Those that did not, largely, developed successful online strategies to strengthen their existing competitive advantage, or create one. Some used digital technologies and predictive analytics to enhance their marketing; others used it to enhance their supply-chain management, among other strategies.

Innovations in marketing

Predictive analytics has allowed marketers to increase their estimations of consumer likelihood to purchase. This has allowed firms a to optimize their marketing mix, allowing them a greater ability to target consumers. It has been a boon to traditional retailers such as Macy’s. In 2014, the retailer partners with SAP, a leading provider of software and services, to improve its existing predictive analytics software. The new solution allows it to build multiple predictive models that aid it in targeted email marketing and digital marketing campaigns. In its first three months using the solution, Macy’s saw, on average, a ten percent increase in sales.

Other retailers use the data they receive to build a personalized shopping experience for the consumer online, and/or optimize the overall shopping experience in a retail location. By parsing consumer data and paring it with predictive analytics models, firms can create targeted online messages, unique promotional opportunities, and other incentives to drive them to a brick and mortar retail location.

Beyond driving traffic, predictive analytics’ real promise for marketers involves increasing customer lifetime value. Cost-per acquisition can be exceedingly costly, especially for non-essential items and/or in crowded markets. Personalized experiences can deepen customer engagement and brand loyalty, increasing the value of their lifetime purchases and decreasing your retention cost. When you determine the return on investment for particular customer segments, you can more effectively determine the optimal ones to target.

Analysis of sales data and sales analytics can also yield significant, positive implications for supply chain management.

Supply chain management

Supply chain management is another strategic business area that has been transformed by predictive analytics. Prior to the Digital Era, sourcing decisions were made based on annual evaluations of sales data, personal relationships with suppliers, regional distribution chains, and past practices. Now, predictive analytics provides brick and mortar chains the insights to be able to shift sourcing based on real-time data, determine whether new suppliers of a particular product or skew will increase or decrease revenue, and determine ideal wholesale and retail pricing.

One such firm is brick-and-mortar retailer Walmart, whose online storefront is a viable competitor to other market leading online retailers. Walmart, whose suppliers are located in more than 70 different countries, and whose stores stock an average of over 175,000 products, is immensely profitable because it aggregates data on every aspect of its retail operation and analyzes them to forecast demand and consumer purchase behavior. Each day, Walmart feeds the reams of data it receives from in-store and online sales-tracking and inventory management systems and feeds that information back into its supply and distribution systems. By coupling this with sales forecasting data from local demand forecasting models, Walmart stores can minimize all product shortages significantly.

Predictive analytics is also used to optimize sourcing and shelving decisions. Through simulation and analysis of historical data, Walmart is able to use predictive analytics to determine the product mix that will allow it to achieve the highest sales revenue at the lowest wholesale cost in the least amount of time. Part of this equation is determining where products should be located in the store to stimulate the most sales growth. All store variables are tracked vigorously and assessed in real-time to ensure the firm’s success.


Whether a firm is choosing what products to manufacture or distribute, a retailer is figuring out what to source, or a franchise is seeking the ideal owner-operators, predictive analytics can yield tremendous results for every business type. However, for some firms, predictive analytics has created such a significant competitive advantage that it has yielded entirely new ways of doing business.

Online-only distributors and retailers

During the early days of the Digital Era, the thought of an online-only distributors and retailers was hard to grasp. True, the reduced operating expenses could serve as a source of competitive advantage, but established brick and mortar businesses had the benefits of human capital, strong brand recognition, advertising dollars, facilities, and relationships.

However, as time has passed, digital technologies have emphasized data  – bringing it on par with, if not making it more important than, relationships as the basis for B2B sales and strategic partnerships in many industries. It has allowed us to have a distributed workforce, eliminating the need for central facilities. It has allowed us to build brand recognition in new ways, some for a fraction of traditional advertising costs (indeed, it has eroded them). Finally, it has allowed the online only distributor or retailer a significant source of competitive advantage in the form of Big Data and predictive analytics.

Take human capital – particularly in sales. Firms can use predictive analytics to predict consumer preferences through recommendation systems. They can then immediately upsell. A human salesperson requires training and a professional demeanor at all times to upsell, whereas a few lines of code are all that is needed for a consumer to receive a personalized suggestion. Netflix is a strong example of an online-only firm that use recommendation engines to drive sales. Netflix’s competitive advantage placed significant pressure on its early rival Blockbuster, which pre-2010, was a household name and market leader in video sales. But with 60,000 employees at its peak, compared to Netflix’s 2,000, as well as an inability to efficiently forecast demand and supply of its products, Blockbuster lost significant share to Netflix and eventually filed for bankruptcy. And while the recommendation engine reduces labor costs, it also drives sales (or in Netflix’s case brand loyalty, as they charge a flat fee) automatically and in real-time.

Human labor reductions are not the only use of predictive analytics. Indeed, many firms have used predictive analytics to forecast talent shortages, predict the likelihood of employee retention, and forecast an employment candidate’s probable performance with their firm. Google is well-known for using what is known as people analytics – the application of predictive analytics to human resources to forecast an employee’s future career trajectory inside the firm. It also uses people analytics to design workspaces and collaborative opportunities that optimize employee innovation. And Google has long been known as a firm whose talent has been a source of significant competitive advantage.

Online Auction-Based Businesses

Predictive analytics also can be used to drive auction-based businesses. For firms where price optimization is critical, like eBay, which derives revenue from fees placed on user auctions, predictive analytics can be used to determine the price. For firms that auction their inventory, and whose fees vary based on the total transaction amount, predictive analytics can help forecast demand and scarcity, and allow these firms to set initial bids efficiently.

Firms can use predictive analytics coupled with software purchasing applications to purchase goods or services automatically when those goods or services hit certain price targets or other thresholds. This is seen in high-frequency trading, where speculators bid on certain equity or debt instruments meeting certain conditions. Some programs use models and inputs developed entirely by humans. Others have paired predictive analytics with machine learning – the process by which computers process increasingly complex information by constructing learning algorithms. This paring allows programs to make decisions without human input concerning bid behavior based not just on the results of the model (that may include thousands of variables), but their own past purchase behavior.

eBay takes predictive analytics a step further than its online auctions. Given its scope (an e-commerce portfolio with approximately $300 billion in transactions), eBay uses predictive analytics to optimize decision-making. Their SAP-designed  system is used to predict problems through daily forecasting, and run simulations on the effect of decisions on the entire portfolio. Other firms, particularly those in financial and legal services use predictive analytics to forecast potential issues as such issues can be extremely costly for their clients and themselves.

Online /Mobile Advertising and Ad Exchanges

Digital marketing, which accounts not just for business strategy, but also comprises a healthy industry of digital marketing agencies, is driven by data and analysis. The fields of search engine optimization, for example, involves leveraging a firm’s digital and non-digital assets to drive traffic to a landing page. To do so, a firm must build models of what has worked to inform its strategies going forward. Since digital data can be obtained in real-time, and most web analytics programs now have considerable analytic and testing tools, digital marketers can quickly test and implement some strategies based on their models. Some firms implement programmatic buying – buying using the results of predictive analytic models and machine learning, to make and serve ads and marketing content to users.

The field of search marketing is also dependent on predictive analytics as it requires firms to forecast the search terms consumers will use to find their firms as well as firms in their industries, as well as related industries. Using web analytics programs allows digital marketers to determine the ideal keywords and key phrases on which to bid. Predictive analytics models can be used by firms to aid in the accuracy of the bidding process.

Search itself is driven by predictive analytics. Though the algorithms are proprietary, firms like Google and Microsoft create lists of relevant website links based on user inputs that are not just limited to what one types into the search bar. These firms spend hundreds of millions of dollars working to refine these lists, “predicting” that they will be exactly what you are looking for. And by transforming the way firms do business, innovative search engine firms, such as Google, Baidu, and Yahoo have revolutionized industry.

Digital advertising is driven by predictive analytics. Many online and mobile search ad-serving platforms create models of target customers using a client firm’s data. They then serve ads to online users who fit that profile, using cookies to track their movements. A user could visit a partisan political website and a food website, for example, and see the exact same ad on each. This is known as retargeting and is used by many firms, such as apparel retailer Levis. A properly implemented retargeting campaign can increase click-through rates and conversions.

Comments are closed.