Online consumer behavior modeling

How to attract and win over the consumer in the highly competitive Internet marketplace?

What would the best marketer of the world do?

What would Sun Tzu do? 

What would Alexander the Great do? 

Well, the answer is not so simple, since online consumer behavior has become an emerging research area as well as a key factor for many companies. Digital ecosystems are evolving so rapidly and it seems that even the greatest strategists might face difficulties in consumer behavior modeling and optimizing consumer interactivity with the brand.

But how do we, marketers, approach the construct "online consumer behavior" in the first place? As suggested by Douglas et al. (1994), strong theoretical and conceptual frameworks can be developed through an integration of constructs from different disciplines. It is not coincidental that academic literature on online consumer modeling may be found in Journals like the ones of Electronic Commerce, Marketing Management, Decision Support Systems, Economic Psychology, Interactive Marketing, Management Information Systems and many more. Theory of Reasoned Action (TRA) and its family theories including the Technology Acceptance Model (TAM) and the Theory of Planned Behavior (TPB) are dominant theories in process of explaining online consumer actions. Expectation-Confirmation Theory (ECT) and Innovation Diffusion Theory (IDT) have also been repeatedly tested in the study of online consumer behavior.

However, Cheung et al. (2003) tried to approach it by integrating the three key concepts of intention, adoption, and continuance, based on Fishbein’s attitudinal theoretical model (Fishbein 1967) and the expectation-confirmation model (Oliver 1980). Their Model of Intention, Adoption, and Continuance (MIAC) seems an interesting framework for developing the framework of their theory.

MIAC

Thus, the modeling framework proposed, as an extension to Hoffman and Novak (1996), it the following:

The five domains of consumer, product, medium, intermediary characteristics as well as enviromental influences were integrated in MIAC (intention, adoption, repurchase) in order to provide a cohesive view on online consumer behavior.

It seems that a wise marketer may segment the market, target the correct target segment with the correct product, both core and augmented, and position it with the correct marketing mix, in order to win competition. However, medium characteristics should not be ignored; Trust and perceived risk have been widely investigated in the study of consumer online purchase intention. Some recent studies (Lee and Turban 2001) focused primarily on the trust formation process in the context of Internet shopping. Furthermore, environmental influences including norms, cultures and cultural contextual sets should also not be underestimated; A consumer behavior pattern, e.g. adoption and success of a fragrance e-shop in Mexico could be totally different in Greece or India.

Electronic commerce is rapidly changing the way people do business all over the globe. In the B2C segment, Internet sales have been increasing dramatically over the last few years. Customers, not only those from well-developed countries but also those from developing countries, are getting used to the new shopping channel. Understanding the factors that affect intention, adoption and repurchase are important both for researchers and practitioners. MIAC is a nice model to begin with.

Classical consumer behavioral theories provide researchers with a good starting point in understanding online consumer behavior. However, we should take the IT component into serious consideration when doing research in online consumer behavior. Instead of blindly borrowing theories and models from other disciplines, researchers and practitioners should test their own behavioral models declaring what is unique and specific to the context of consumer-based electronic commerce. This may be the only way that marketing can survive in the Digital Age.

Read more »

New Brand Assets: Where are we going?

"We show that easily accessible digital records of behavior, Facebook Likes, can be used to automatically and accurately predict a range of highly sensitive personal attributes including: sexual orientation, ethnicity, religious and political views, personality traits, intelligence, happiness, use of addictive substances, parental separation, age, and gender. The analysis presented is based on a dataset of over 58,000 volunteers who provided their Facebook Likes, detailed demographic profiles, and the results of several psychometric tests."

Kosinski et al. 2013 (The Psychometrics Centre, University of Cambridge)

Like. Comment. Engage. Sponsored post again? Uhm.. Is your behavioral targeting ok? No?It's fine. Don't worry. Numbers are fine.
How have brands embraced digital ecosystems like Facebook as a key marketing channel to drive engagement and brand awareness? Let's use FB like waveforms as a metric of brand engagement and let's illustrate some examples of FB like pages:

Multinational retail corporation with 35M likes

Ivy League management magazine:

A worldwide branded commodity

A multinational sports company

Leading Telecommunications company in Greece-edited

Leading retail operator in Greece

It seems that these waveforms are similar to a manipulated oscillograph signal where the peak comes in "Sponsored Campaigns" periods and then to zero, till the next paid campaign. 

There are tons of examples like these. Off course, some of these effects could be explained by deeper analysis of econometric advertising models, mentioned here, but is this the best we can do with these new media? Where is organic traffic? Does paid traffic have a decent ROI in terms of brand engagement?

It depends.

I believe that the virtual digital economies which are developing as we speak will reshape the landscape of marketing as we know it. But do marketers do their best in order to engage with the customer, as brand ambassadors? The images above indicate that apart from the times that we press the button "Boost Post", things do not go very well in terms of brand engagement. Further KPI's, like numbers of comments and shares, retweets and favorites, per brand page, analyzed with statistical tools, should easily verify this hypothesis. I understand that perhaps, the higher the numbers, the higher is the brand perception and other key brand assets, but are the results discrete and measurable? 

In the long term, does the world have enough strategists to make optimum use of the new technologies and channels of communication?

For the time being we have:

• Google, Facebook, Twitter charging and gaining billions for "awareness" and "engagement".
• Overvalued IPOs, overnight millionaires, questionable business models, startups offering frivolous services, fake engagement tools and similar services.
• Marketers trying to deliver actions with ROI but frequently unable to achieve optimum brand engagement, let alone incremental sales or other corporate goals.
• 3B customers who are now online but have the same purchasing power that they would have if digital marketing wouldn't exist in a parallel universe.
• Academics trying to follow up and model consumer behavior versus both real-life and digital marketing stimuli as technology advances.

A nice bubble? I think, yes.

How would Aristotle structure his philosophical pillars if he was writing his "Athenian Constitution" in his free time while working as a Chief Marketing Officer for Facebook or Google?

21st century. Who will pay for the news?

Read more »

5+1 Latest Viral Commercials

The latest February commercials that went viral!

5. M&M'S Superbowl Ad

4. Jägermeister UK

3. Jerry Ricecake Superbowl Ad

2. U2 "Invisible" & Bank of America

1. Wonderful Pistachios Superbowl Ad

Lucky Star: SOS Children Villages

Read more »

Target markets and ANN

Once we, marketers, have determined our business products or services, we must accurately identify our target markets. The target market is the actual customer group, or audience, in which our business will attempt to sell its products and services. Target marketing tailors a marketing mix for one or more segments identified by market segmentation and usually contrasts with mass marketing, depending on the industry, the competition, the marketing budget of the business, NGO vs non-NGO, etcetera. Many techniques have been used in order to select the target markets, such as statistical regression (Bult&Wansbeek, 1995), neural computing (Zahavi&Levin, 1997) and fuzzy clustering (Setnes&Kaymak, 2001). Direct marketing is, off course, a well-known type of marketing, used by businesses, in order to target, call to action and increase profits. A well-executed direct advertising campaign can have a satisfying Return On Investment, since many potential customers responded to a clear call-to-action. The growing interest in direct marketing techniques has made it an important application field for marketers and data mining professionals.

CHAID vs ANN

Artificial neural networks (ANN) are "computational models inspired by animals' central nervous systems that are capable of machine learning and pattern recognition". They are presented as systems of interconnected neurons that can compute values from inputs by feeding information through the network. 

Commonly, a class of statistical models may be called "neural" if they

1. consist of sets of adaptive weights, i.e. numerical parameters that are tuned by a learning algorithm, and
2. are capable of approximating non-linear functions of their inputs.

The adaptive weights are conceptually connection strengths between neurons, which are activated during training and prediction.

As aforementioned, marketers should divide the market into segments. A product offer would then be made to those people who are considered to be in the market segment for which the product was meant. Automation segmentation techniques have been used to divide the customers into a number of segments, like CHi-squared Automatic Interaction Detection (CHAID). CHAID is a type of decision tree technique, that is used for prediction as well as classification and detection of interaction between marketing variables. In practice, CHAID is often used in the context of direct marketing to select groups of consumers and predict how their responses to some variables affect other variables.

When a segmentation method such as CHAID is used to determine target segments which are more likely to respond to marketing actions, the data set is split into disjoint groups and the groups that are considered to be most likely to respond are targeted through a marketing action, e.g. mailed preferentially. Neural networks, however, do not segment data into disjoint groups. That means, a different approach is needed to develop target selection models by using neural networks. Since target selection models try to differentiate responders from non-responders, marketers could consider formulating the problem as a classification problem and develop a network that classifies each input pattern as a responder or non-responder. However, the classes are not well separated in target selection problems.  

Furthermore, the miss-classification costs  are asymmetric. Miss-classifying a possible responder as a non-responder and not targeting that individual is more costly to the direct marketing company than miss-classifying a non-responder as a responder. Therefore, another method than pure classification should be used. One solution could be to define a measure for the "likelihood" of an individual to respond (call-to-action) via the marketing action, e.g. mailing. The individuals are then ordered according to this measure, and only the ones that score above a given threshold θj are targeted (target scoring). 

Artificial neural network models for target selection fall into this category. The output of the neural network is then a measure of likelihood of response (Potharst et al., 2001). Now, the goal of determining the neural network would be to determine a correct set of network parameters (weights) such that a good indication of the likelihood of response of the supporters is obtained, given the inputs to the network. Naturally, these inputs must convey information about the characteristics of the supporters, including demographic, geographic, or even psychographic and behavioral variables.

ANN configuration and analysis

In target selection problems we form the model of the likelihood of response at a single point in time, given various supporter characteristics and/or their past behavior. In real life, feed-forward neural networks are sufficient for target selection problems. The nonlinear activation function of the neurons also has an influence on the final neural network model for a given network structure and the number of neurons. The logistic sigmoid function:

is often used since the neural network output is expected to lie in the range [0,1] (likelihood of response). Thus, the neural network model can be analyzed as a generalized nonlinear regression model with the likelihood [0,1] of response to our marketing action as an output .

We may visualize our target selection model through a hit probability chart. 

Potharst et al. 2001

A hit probability chart exposes what percentage of a selected population may respond to our marketing action. The x-axis shows a percentage of the total group selected for targeting and the y-axis shows what percentage of that group is a responder to our call-to-action. The hit probability chart for a successful model starts with high values of hit probability. As the size of the selected group increases, the percentage of responders within the selected group will decrease until the hit percentage is equal to the percentage of responders within the total marketing action eg. mailing considered in the campaign.

In real life, after data preparation and data set selection, marketers may use IBM SPSS Neural Networks as a modern modelling tool.  

With SPSS Neural Networks, marketers may select either the Multilayer Perceptron (MLP) or Radial Basis Function (RBF) procedure. Both of these are map relationship techniques implied by the data. Both use feedforward architectures, meaning that data moves in only one direction, from the input nodes through the hidden layer of nodes to the output nodes. Prior research has indicated that neural network models may perform better than the CHAID model, while they perform on an equal level with the logistic regression models we aforementioned.

Conclusion

In general, ANN for Marketing can be viewed as a modelling tool that helps marketers to take decisions. ANN can recognize patterns, pick up key information that are not easily identifiable and develop relationships among them. Marketers are expected to fully take advantage of the ANN modeling in order to solve their target selection problem faster and more efficiently.

Read more »

Econometric models of advertising

Advertising is a powerful form of marketing communication used to "encourage, persuade, or even manipulate an audience to take or continue to take some action". The final desired result is to drive consumer behavior with respect to a commercial offering, although political and ideological advertising is also common. The marketing mix has been the key concept to advertising. The marketing mix refers to four variables (the 4 P's) that a marketing manager can control in order to influence a brand’s sales or market share. The basic question that managers face nowadays is, what level or combination of these variables maximizes sales, market share, or profit? Or, otherwise, "How do sales or market share respond to past levels of or expenditures on these variables"?

Researchers have developed a variety of econometric models of market response to the marketing mix, most of these which have focused on market response to advertising or pricing (Sethuraman, 1991). The reason may be that expenditures on these variables seem the most discretionary, so marketing managers are most concerned about how they manage these variables. The basic philosophy underlying the approach of response modeling is that past data on consumer and market response to the marketing mix contain valuable information that can enlighten our understanding of response. 

Seven response models to advertising are have been already patternalized: Current, shape, competitive, carryover, dynamic, content and media effects of advertising. The first four are common across price and the other marketing variables. The last three are unique to advertising. 

Current and Carryover Effect

The current effect of advertising is the change in sales caused by an exposure of advertising occurring at the same time period as the exposure. The x-axis here is time, while sales are on the y-axis and the baseline sales are the dashed line. The current effect of advertising is the spike in sales from the baseline given an exposure of advertising. Decades of research indicate that this effect of advertising is small relative to that of other marketing variables and quite fragile, as it can be easily drowned out by the noise in the data. It can be captured by the model:

Yt = a + b1 At + b2Pt + b3Rt + b4Qt +  et 

where Y represents the dependent variable (e.g. sales), while the other capital letters represent variables of the marketing mix, such as advertising (A), price (P), sales promotion (R), or quality (Q). The parameters a and bk are coefficients that the researcher wants to estimate. bk represents the effect of the independent variables on the dependent variable, where the subscript k is an index for the independent variables. et are errors that can be approached as "white noise".

The carryover effect of advertising is that portion of its effect that occurs in time periods following the pulse of advertising. The carryover effect may occur for several reasons, such as delayed exposure to the ad, delayed consumer response, delayed purchase due to consumers’ backup inventory, delayed purchase due to shortage of retail inventory, and purchases from consumers who have heard from those who first saw the ad (word of mouth). The carryover effect may be as large as or larger than the cur-rent effect. Typically, the carryover effect is of short duration, rather than of long duration. It can be captured by the multiplicative model:

Yt = Exp(a) At b1 Pt b2 Rt b3 Qt b4 et

with sales (Y), advertising (A), price (P), sales promotion (R), quality (Q) inter correlated. Here, the independent variables have a synergistic effect on the dependent variable. In many advertising situations, the variables could indeed interact to have such an impact. For example, higher advertising combined with a price decrease may enhance sales more than the sum of higher advertising or the price drop occurring alone.

Shape Effect

The shape of the effect refers to the change in sales in response to increasing intensity of advertising in the same time period. The intensity of advertising could be in the form of exposures per unit time and is also called frequency or weight. The x-axis now is the intensity of advertising in a period, while the y-axis is the response of sales.  It can be captured by the exponential attraction model:

Mi = Exp (Vi )/Σ Exp Vj 

where Mi is the market share of the ith brand (measured from 0 to 1), Vj is the marketing effort of the jth brand in the market, Σ stands for summation over the j brands in the market, Exp stands for exponent, and Vi is the marketing effort of the ith brand. We know that Vi = a + b1 Ai + b2Pi + b3Ri + b4Qi + ei, as aforementioned, thus:

Mi  = Exp (Vi )/Σ Exp Vj = Exp (Σbk  Xk  + e)/ Σj Exp(Σkbk Xik + ej),

where Xk (0 to m) are the m independent variables or elements of the marketing mix, a = b0 and Xi0 = 1. The use of the ratio of exponents in the above equations ensures that market share is an S-shaped function of share of a brand’s marketing effort.

The diagram above shows three typical shapes: linear, concave and S-shape. Of these three shapes, the S-shape seems the most plausible because it suggests that at some very low level, advertising might not be effective at all because it gets drowned out in the noise. At some high level, it might not increase sales (advertising elasticity-->0) because the market is saturated or consumers are insensitive from the repetitive advertising.

Competitive Effect

Advertising normally takes place in free markets. Whenever one brand advertises a successful innovation or successfully uses a new advertising form, other brands quickly imitate it. Competitive advertising tends to increase the noise in the market and thus reduce the effectiveness of any one brand’s advertising. The competitive effect of a target brand’s advertising is its effectiveness relative to that of the other brands in the market.  The simplest method of capturing advertising response in competition is to measure and model sales and advertising of the target brand relative to all other brands in the market with more complex models, mainly based on the aforementioned econometric model of the "Shape Effect". These first four models may also apply to "Price" instead of "Sales" dependent variable.

Dynamic, Contents and Media Effect

Dynamic effects are those effects of advertising that change with time. Included under this term are carryover effects discussed earlier and wearin, wearout effects. Wearin is the increase in the response of sales to advertising, from one week to the next of a campaign, even though advertising occurs at the same level each week. Wearout is the decline in sales response of sales to advertising from week to week of a campaign, even though advertising occurs at the same level each week. Wearout typically occurs at the end of a campaign because of consumer indifference. 

Content effects are the variation in response to advertising due to variation in the content or creative cues of the ad. This is the most important source of variation in advertising responsiveness and the focus of the creative talent in every agency. Thus, the challenge for marketers is to include measures of the content of advertising when modeling advertising response in real markets.

Media effects are the differences in advertising response due to various media, such as TV, or newspaper, and the programs within them, such as channel for TV or section or story for newspaper.

The last three effects (dynamic, content, media) may be captured by models similar to the S-shaped response with modeling interactions integrated due to the competitive environment or even competitive distributed lag Models, however in reality their mathematics can be quite complex (Tellis, 2006).   

Conclusion

Prudent marketing managers need to take into account how markets have responded to the marketing mix in the past. Past may predict future with relative certainty but it also contains valuable lessons that might enlighten the future. The aforementioned econometrics of response modeling describe how a researcher should model response to the marketing mix and especially to advertising in order to capture and control the most important effects validly. Which may be the "raison d'être" for marketing managers, in the first place.

Read more »