Showing posts with label Internet Marketing. Show all posts
Showing posts with label Internet Marketing. Show all posts

Wednesday, August 21, 2013

Facebook Likes and Human Behavior

Earlier in 2013, the Psychometrics Centre of the University of Cambridge conducted an impressive research about how private attributes are predictable from digital records of human behavior, like Facebook likes. Digitally mediated behaviors like Facebook likes can easily be recorded and analyzed, fueling the emergence of computational social science and new services such as personalized search engines and targeted online marketing. However, the widespread availability of extensive records of individual behavior, together with the desire to learn more about customers and citizens, presents serious challenges related to privacy and data ownership. 

The researchers (Kosinski, Stinwell & Graepel) showed 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. The proposed model uses dimensionality reduction for preprocessing the Likes data, which are then entered into linear regression to predict individual psychodemographic profiles from Likes. The model correctly discriminates between homosexual and heterosexual men in 88% of cases, African Americans and Caucasian Americans in 95% of cases, and between Democrat and Republican in 85% of cases.

Predicting individual traits and attributes based on various cues, such as samples of written text, answers to a psychometric test, or the appearance of spaces people inhabit, has a long history. Human migration to digital environment renders it possible to base such predictions on digital records of human behavior. It has been shown that age, gender, occupation, education level and even personality can be predicted from people’s Web site browsing logs. Similarly, it has been shown that personality can be predicted based on the contents of personal web sites, music collections, properties of Facebook or Twitter profiles such as the number of friends or the density of friendship networks, or language used by their users. Furthermore, location within a friendship network at Facebook was shown to be predictive of sexual orientation.

The research 

The Psychometrics Centre, University of Cambridge

The study was based on a sample of 58,466 volunteers from the United States, obtained through the my Personality Facebook application, which included their Facebook profile information, a list of their Likes (n = 170 Likes per person on average), psychometric test scores, and survey information. Users and their Likes were represented as a sparse user–Like matrix, the entries of which were set to 1 if there existed an association between a user and a Like and 0 otherwise. The dimensionility of the user–Like matrix was reduced using singular-value decomposition (SVD). 

Numeric variables such as age or intelligence were predicted using a linear regression model, whereas dichotomous variables such as gender or sexual orientation were predicted using logistic regression. In both cases,the researchers applied 10-fold cross-validation and used the k =100 top SVD components. For sexual orientation, parents’ relationship status, and drug consumption only k = 30 top SVD components were used because of the smaller number of users for which this information was available.

Prediction of Dichotomous Variables

The Psychometrics Centre, University of Cambridge
The aforementioned figure shows the prediction accuracy of dichotomous variables expressed in terms of the area under the receiver-operating characteristic curve (AUC), which is equivalent to the probability of correctly classifying two randomly selected users one from each class (e.g., male and female). The highest accuracy was achieved for ethnic origin and gender. African Americans and Caucasian Americans were correctly classified in 95% of cases, and males and females were correctly classified in 93% of cases, suggesting that patterns of online behavior as expressed by Likes significantly differ between those groups allowing for nearly perfect classification. Christians and Muslims were correctly classified in 82% of cases, and similar results were achieved for Democrats and Republicans (85%). Sexual orientation was easier to distinguish among males (88%) than females (75%), which may suggest a wider behavioral divide (as observed from online behavior) between hetero and homosexual males.
Good prediction accuracy was achieved for relationship status and substance use (between 65% and 73%). The relatively lower accuracy for relationship status may be explained by its temporal variability compared with other dichotomous variables (e.g., gender or sexual orientation).

Predictive Power of Likes

Individual traits and attributes can be predicted to a high degree of accuracy based on records of users’ Likes. The best predictors of high intelligence include “Thunderstorms,” “The Colbert Report,” “Science,” and “Curly Fries,” whereas low intelligence was indicated by “Sephora,” “I Love Being A Mom,” “Harley Davidson,” and “Lady Antebellum.” Good predictors of male homosexuality included “No H8 Campaign,” “Mac Cosmetics,” and “Wicked The Musical,” whereas strong predictors of male heterosexuality included “Wu-Tang Clan,” “Shaq,” and “Being Confused After Waking Up From Naps.” 

Accuracy of selected predictions as a function of the number of available Likes. Accuracy is expressed as AUC (gender) and Pearson’s correlation coefficient (age and openness). About 50% of users in this sample had at least 100 Likes and about 20% had at least 250 Likes. Note, that for gender (dichotomous variable) the random guessing baseline corresponds to an AUC = 0.50. The Psychometrics Centre, University of Cambridge.

Moreover, note that few users were associated with Likes explicitly revealing their attributes. For example, less than 5% of users labeled as gay were connected with explicitly gay groups, such as No H8 Campaign, “Being Gay,” “Gay Marriage,” “I love Being Gay,” “We Didn’t Choose To Be Gay We Were Chosen.” Consequently, predictions rely on less informative but more popular Likes, such as “Britney Spears” or “Desperate Housewives” (both moderately indicative of being gay).


Similarity between Facebook Likes and other widespread kinds of digital records, such as browsing histories, search queries, or purchase histories suggests that the potential to reveal users’ attributes is unlikely to be limited to Likes. Moreover, the wide variety of attributes predicted in this study indicates that, given appropriate training data, it may be possible to reveal other attributes as well.

Predicting users’ individual attributes and preferences can be used to improve numerous products and services. For instance, digital systems and devices (such as online stores or cars) could be designed to adjust their behavior to best fit each user’s inferred profile. Also, the relevance of marketing and product recommendations could be improved by adding psychological dimensions to current user models. For example, online insurance advertisements might emphasize security when facing emotionally unstable (neurotic) users but stress potential threats when dealing with emotionally stable ones. 

Moreover, digital records of behavior may provide a convenient and reliable way to measure psychological traits. Automated assessment based on large samples of behavior may not only be more accurate and less prone to cheating and misrepresentation but may also permit assessment across time to detect trends. Moreover, inference based on observations of digitally recorded behavior may open new doors for research in human psychology.

On the other hand, the predictability of individual attributes from digital records of behavior may have considerable negative implications, because it can easily be applied to large numbers of people without obtaining their individual consent and without them noticing. Commercial companies, governmental institutions, or even one’s Facebook friends could use software to infer attributes such as intelligence, sexual orientation, or political views that an individual may not have intended to share. One can imagine situations in which such predictions, even if incorrect, could pose a threat to an individual’s well-being, freedom, or even life. Importantly, given the ever-increasing amount of digital traces people leave behind, it becomes difficult for individuals to control which of their attributes are being revealed.

There is a risk that the growing awareness of digital exposure may negatively affect people’s experience of digital technologies, decrease their trust in online services, or even completely deter them from using digital technology. It is our hope, however, that the trust and goodwill among parties interacting in the digital environment can be maintained by providing users with transparency and control over their information, leading to an individually controlled balance between the promises and perils of the Digital Age.

Monday, July 08, 2013

Behavioral Targeting: The Holy Grail of Online Marketing

Behavioral targeting, under which users are presented with advertisements based on their past browsing and search behavior and other available information (e.g., hobbies registered on a website), has been hailed as the new Holy Grail in online advertising.We will refer to the economic implications when an online publisher engages in behavioral targeting. Revenue for the online publisher in some circumstances can double when using behavioral targeting. On the other hand, increased revenue for the publisher is not guaranteed: in some cases the prices of advertising and hence the publisher's revenue can be lower, depending on the degree of competition and the advertisers' valuations. Although social welfare is increased and small advertisers are better off under behavioral targeting, the dominant advertiser might be worse off and reluctant to switch from traditional advertising.

A simple question

Who benefits (and what are the conditions required) from behavioral targeting as compared to traditional advertising? Would the online publisher benefit from the targeting of advertisements? Because of the increased effectiveness of behaviorally targeted advertisements, conventional wisdom would suggest that the answers to these questions are easily predicted, as summed up in an article in the Economist about behavioral targeting:  [...], Advertisers will be prepared to pay more to place ads, since they are more likely to be clicked on. That in turn means that websites will be able to charge more for their advertising slots.  (Economist, 2008)

However, this expected relationship between charges and clicks does not necessarily emerge when the advertisement slot is auctioned off . Instead, using targeted advertisements turns out to be similar to product differentiation: it causes relaxed competition between the advertisers, and hence it is possible that advertisers need not pay as much as they do under traditional advertising. That is, by focusing on a specific user segment, an advertiser's advertisement may be selected with a relatively low price on this segment, whereas under traditional advertising his advertisement would never have been selected or would have been selected only at a higher price. This competitive effect can depress the online publisher's income by realizing a lower revenue per click-through.

Competitive and Propensity effect

On the other hand, the negative effect of relaxed competition for online publishers might be off set by a positive propensity effect. Through targeting advertisements,the probability of a click-through is increased resulting in a higher volume of click-throughs, which positively contributes to the publisher's revenue. Whether the publisher can benefit from behavioral targeting depends on the trade-off between the competitive effect and the propensity effect. Behavioral targeting outperforms traditional advertising only if the competitive effect is dominated by the propensity effect. In particular,when the advertisers competing for the advertising space are comparable and the number of advertisers is large, behavioral targeting generates more revenue for the publisher. This gain under behavioral targeting is increasing in user heterogeneity and the number of advertisers, and the expected revenue for the publisher can double compared to traditional advertising.

 Online consumer heterogeneity: An advertiser for each face.


The whole research,conducted by Jianqing Chen and Jan Stallaert,University of Texas and Connecticut respectively, proved that that the effect of behavioral targeting on different advertisers' payoffs is asymmetric. While small advertisers are generally better off under behavioral targeting by winning their favorable users, the dominant advertiser may or may not be better off .The dominant advertiser is worse off under behavioral targeting when it has a significant competitive advantage over its competitors because under traditional advertising, he would otherwise grab a larger group of users and still realize a decent payoff . The real benefit brought by the increased effectiveness of behavioral targeting is realized in social welfare. In the end,the social welfare of both publisher and advertisers can be maximized under behavioral targeting.

Wednesday, June 12, 2013

Viral Marketing-A myth?

That ideas can go viral is now a given in corporate marketing and the culture – it’s even part of the plot of the 2011 Pulitzer Prize-winning novel “A Visit from the Goon Squad.” But new research suggests the term “viral” marketing does not describe accurately what happens in the market.

Sunday, April 14, 2013

5+1 tips για online business marketing

   Το οnline marketing είναι ένα θέμα με μεγάλη έκταση. Τι σημαίνει όμως online marketing; Τα κλασικά εργαλεία όπως banners, popups και newsletters είναι σε όλους γνωστά. Υπάρχουν όμως πολλές επιπλέον μέθοδοι για να εκτεθεί η επιχείρηση και τα προϊόντα της με τη χρήση ψηφιακών επικοινωνιακών εργαλείων στην αγορά.Πως βελτιστοποιούμε ομώς το online marketing για την επιχείρησή μας;

1. Καθορίζουμε το κοινό στο οποίο απευθυνόμαστε.

To online marketing δεν πρόκειται ποτέ να είναι επιτυχές εάν δεν απευθυνθούμε σε  ένα target market και ένα target audience. Ποιοι ενδιαφέρονται για το προϊόν μας; Άνδρες ή γυναίκες; Ποια είναι η ηλικία τους; Ποια είναι τα ενδιαφέροντα των υποψήφιων πελατών μας; Η πιθανή τους εκπαίδευση; Ποια η πιθανή οικονομική τους κατάσταση; Στοχεύουμε σε επιχειρήσεις ή καταναλωτές(b2b vs b2c);Η τμηματοποίηση αυτή της αγοράς είναι η βάση του online marketing.Εξάλλου, ένα επιχειρησιακό blog ή site από μόνο του δεν προσφέρει τίποτα, πρέπει να είναι δομημένο ώστε να αγγίζει τα ενδιαφέροντα του target group.Αφού διαμορφώσουμε ένα λεπτομερές προφίλ του πελάτη ,θα είναι πιο εύκολο να βελτιστοποιήσουμε τη στρατηγική marketing που θα ακολουθήσουμε.

Wednesday, April 10, 2013

Tips για SEO (Search Engine Optimization)

 Όταν καθημερινά πραγματοποιούμε αναζητήσεις στις μηχανές αναζήτησης όπως το Google,το Yahoo ή το Bing,για κάτι που ενδεχομένως ψάχνουμε, προϊόν ή υπηρεσία, η μηχανή αναζήτησης απαντά με αποτελέσματα σύμφωνα με τον   επιτυχημένο αλγόριθμό της. Πρέπει δε να έχουμε υπόψη μας ότι η κυριότερη μηχανή αναζήτησης (με ποσοστό τουλάχιστον 2/3 της συνολικής αγοράς) είναι η Google.
 Σε μία υποθετική λοιπόν αναζήτηση στο τα τρία πρώτα αποτελέσματα που βρίσκονται σε κίτρινο φόντο είναι διαφημίσεις. Τα υπόλοιπα είναι τα οργανικά αποτελέσματα αναζήτησης και αποτελούν ταξινομημένες ιστοσελίδες.

Τι είναι όμως το SEO?