Review of Web Advertising
Date: Mon, 23 May 2005 17:17:05 +0100
From: Marina Santini <Marina.Santini@itri.brighton.ac.uk>
Subject: Web Advertising: New forms of communication on the Internet
AUTHOR: Janoschka, Anja
TITLE: Web Advertising
SUBTITLE: New forms of communication on the Internet
SERIES: Pragmatics & Beyond New Series 131
PUBLISHER: John Benjamins
Marina Santini, PhD student, University of Brighton, UK.
The book "Web advertising. New forms of communication on the Internet" by
Anja Janoschka includes Acknowledgements, a Table of contents, a Table of
figures, seven chapters, endnotes, References, and an Index. It contains a
useful analysis of web advertising from three points of view --
communication, language and textuality -- which can help scholars working
in language variation and evolution, linguists dealing with textuality and
hypertexts, genre analysts, communication and media experts and also non-
specialized readers. The author collects in a single book some important
observations from the literature, and additionally suggests new models in
order to explain the specificity of the form of communication she analyzes
in depth, namely web advertising.
Chapter 1 contains the "Introduction" of the book. It includes the purpose
of the book, a short description of the data used in the study, and the
outline of the chapters. The book illustrates a study carried out on web
advertising and has three main aims. The first aim is to describe and
explore the new dimensions and forms of online communication. The second
aim is the analysis of written language used in online advertising. The
third focus is on the new structure and functions of online information
enacted in hypertexts. The data she analyzes includes different types of
web advertisements. The analytical approach is qualitative, with
decontextualized examples (from year 2000 to year 2002, plus a few later
examples) exemplifying the main points. The analysis focuses only on
advertising messages in the written form, and not on the web page as a
whole or on multimedia aspects.
Chapter 2 ("Traditional advertising") describes the main features and
types of traditional advertising, which includes print ads, TV
commercials, direct mailing and coupon ads. More specifically, traditional
advertising subsumes conventional advertising (including print ads, TV
commercials, radio commercials, billboards, etc.) and direct advertising
(including direct mailings, coupon ads, etc.). While conventional
advertising is characterized by a uni-directional message transfer to mass
audience, direct advertising addresses individuals in order to get
feedback by integrated response elements in the advertising message. From
a functional point of view, advertising is not only informative (p. 15),
but it is mainly persuasive (p. 18). One way in which advertising tries to
influence its audience is explained by the AIDA (Attention, Interest,
Desire, Action) concept. The AIDA concept describes a consecutive mental
process in which the successful achievement of one mental process
initiates the next. That means that the first step is to attract
attention. In a second step, it is important to awaken interest in the
object in a way which establishes a desire for it. Desire is one of the
emotional appeals responsible for the buying impulse which leads to the
action, i.e. a purchase or similar. All these steps can be accomplished in
different modes, linguistically and by means of graphic elements.
Persuasion can take place through rational information and emotional
appeal. The author then analyzes in details the following categories of
traditional advertising: print advertisements, TV commercials, direct
mailings and coupon ads.
In Chapter 3 ("Online advertising"), after having introduced the World
Wide Web and the main activities carried out on it, the author analyzes
web advertising. Web advertising generally follows the same principles as
traditional advertising. Usually, only part of a web site serves the
advertising purposes, because the spatial limitation is identical to
traditional media advertising. Nevertheless, there are remarkable
differences. Online advertising is interaction-orientated, which means
that digital ads are meant to be directly activated. This activation is a
form of interaction that provides evidence for the new role of addressees.
Different types of web ads allow different degrees of interactivity. The
AIDA formula also works for the explanation and structure of online
advertising, although greater emphasis is placed on the first step,
getting attention, because of the competitive environment represented by
the web page where the ad is placed. The author points out that the
term "web ad" is used to subsume different types of advertising messages,
including banners, buttons, pop-up windows, etc. These different forms all
aim at informing users about the existence of certain web sites and
persuading them to visit these sites. The important remark made here is
that web ads can be seen as hyperlinks which enable activation through
their users. Once users have clicked on them, they are taken to another
connected web page, the linked target source. Similar to hyperlinks, web
ads consist of three elements, namely the web advertisement which is
located on a web page, the link enacting the connection, and the target
web site. Web ads need to fulfil at least three important functions to be
effective. First, they should attract the user's attention; second, they
should motivate the users to click; third, they should meet users'
expectation. The author then analyzes different types of web ads: static,
animated, and interactive web ads, and also special types of web ad, such
as pop-up ads and web ad traps, which are considered to be a form of
miscommunication. A number of examples serve to elucidate how web ads are
structured, how they function and linguistically address their audience.
The phenomenon of "banner blindness" (i.e. the tendency of web users to
ignore banners, even when they contain information they were looking for)
is then explained.
The focus of Chapter 4 ("Communication") is on communication patterns in
traditional and online communication. The aim is to show that
communication on the Internet is hybrid, i.e. it makes use of traditional
communication forms (mass media and interpersonal communication), but also
develops its own communicative features that have not been previously
used. Although the boundaries between traditional and new forms of
communication are not clear-cut, the author suggests a new communication
model, the "interactive mass communication", which incorporates mass and
interpersonal communication in online communication (Figure 4.5, p. 98).
The model not only combines characteristics of interpersonal communication
and mass communication, but it also allows communication between a sender
and the mass audience. The mutual interaction between the sender and
receivers can be described as a "multiway-communication", combining the
one-to-one and the one-to-many flow of messages. Interactive mass
communication has also some effects on the language used to communicate.
Web ads are written text categories that are characterised by "conceptual
orality", i.e. they are written texts which have been orally
conceptualized. Based on the findings of her study, the author argues that
the language used in written ads can be described as similar to chats
(even though their message transfer is asynchronous, and not synchronous
like in chats). This is because new technical facilities turn uni-
directional mass advertising into an interactive means of communication
that relies on the addressees' contribution. Other elements, such as
interaction, feedback and interactivity, also affect the roles of both
communication partners. The audience is not any more a mass of passive
receivers, but a target group of active interactants.
Chapter 5 ("The language of web ads") is devoted to language analysis. Web
ads employ particular linguistic means in order to attract attention and
to persuade users to click. Persuasion is realized at syntactic, lexical
and pragmatic level together with emotionally motivating strategies.
Although certain linguistic similarities can be found in both traditional
and online advertising, there are particular means which are more
frequently used in online advertising and others which have been recently
established, due to the requirements of interactive advertising
instruments. Although online ads are written texts, they make use of
elements that are typically found in spoken language, for instance
simplified, abbreviated language. Typical linguistic devices are found in
web ads are questions (different types of interrogatives), imperatives
(imperative verb forms and directive speech acts), personal and possessive
pronouns (as in a conversation between the sender and addressee), spatial
and temporal deixis (such as "here" and "there"), abbreviated sentences
(for example, "Travel a lot?"). The conceptual orality, often expressed by
directives, creates a communicative immediacy, imitating interpersonal
communication. There are linguistic means that aim at persuading on an
emotional level, such as motivation strategies, "trigger words" (i.e.
different word classes -- mainly verbs and adjectives, but also nouns --
that can be semantically explicit or provoke the users' attention and
interest by their implicit meaning in the form of emoticons, acronyms,
etc.), and linguistic simplification (simplified items which have a time-
saving aspect because they speed up writing). Web ads do not concentrate
on one single persuasive strategy. They combine several, and any
combination is possible.
Chapter 6 ("Hyperadvertising") is a crucial chapter which collects
important observations and original suggestions. Among other things, the
author analyzes two central criteria of textuality, cohesion and
coherence. While cohesion is based on grammatical dependencies, coherence
focuses on the meaning of text, i.e. the semantic junction of language
units beyond the syntax level. There are types of text like poems which go
beyond cohesion, but they remain coherent. Ads, which may consist of
various linguistic and non-linguistic elements, are understood and are
coherent, although they often lack grammatical cohesion. Coherence
interacts also with linearity. In printed texts, the text producer
establishes and determines the text structure and the receiver's path of
text perception. In this sense, texts are linearly structured and
perceived. However, this concept of linearity is only partially true.
Although the majority of printed texts are constructed in a linear way,
printed linearity is not a text inherent or exclusive criterion for
perception and comprehension. For instance, dictionaries, encyclopedias,
or telephone books are printed documents, but they are not intended to be
read from the beginning to the end. Jucker (p. 163) calls these kind of
printed texts "printed hypertexts". The structure of an ad in traditional
print advertising principally follows the structure of printed hypertexts.
Perception follows certain rules of attention, especially in advertising.
In the interplay of text and picture, the majority of receivers first pay
attention to the picture and then to the text. Since images are perceived
faster than text, they are often implemented to create an initial contact.
Illustrations and pictures have at least two functions: first they should
catch the addressees' attention and keep it on the ad; secondly, they are
responsible for telling the story of the advertising message. Therefore,
advertisers can, to some extent, direct attention and basically determine
the way in which they want addressees to visually access the message.
Nevertheless, after this first visual contact, recipients have the option
of deciding whether to read the body of the message, looking at the
background picture or to concentrating on the headline or slogan. This
freedom makes the access to the message non-linear.
In hypertexts, instead, non-linearity results in individual reading paths
with ubiquitous access and this in turn leads to different ways of
perception and comprehension. The author suggests a new model of hypertext
structure (Figure 6.5, p. 173), adding new elements to the model
previously suggested by Fritz (p. 172). The new model includes the
following considerations: 1) Hypertexts are open-ended; 2) Textual
elements are embedded in other textual elements; 3) Not all hypertexts
contain a starting page or a textual elements like a homepage. In fact,
users do not need to access a web site by its homepage; there are optional
links which lead users directly to the information unit they want.
Nevertheless, the homepage is a means of orientation, structurally and
contextually. It provides an overview of the structural content of the web
site by the main navigation menu and should give users an idea of what the
web site is about. The homepage gives a hypertext a contextual setting, a
starting point, even if there is no end. Hypertexts are decentralized
structures. They are not conceptualized to be followed linearly. The multi-
linear way of presenting information affects the user's reading behaviour.
Online readers have control over the order in which they extract
information from the Web: readers of hypertexts may follow their own path,
create their own order, their own meaning out of the material. The
creation of an individual path entails that for the users the reading
process is always linear. This means that coherence in hypertexts is
process-oriented, determined by the users, and it is not a text intrinsic
quality. Also, although hypertexts are informative, appellative,
obligatory, fulfil a contact function, are declarative and entertaining
(following Brinker's categorization, p. 164), they seldom fulfil only one
function, most often they incorporate many. One explanation for these
blending is their flexible text structure and their interactive function
due to new technologies. While finite texts tend to concentrate on one
function in order to be most effective, infinite hypertexts are difficult
to restrict to just one text function since it seems impossible to define
where a hypertext starts and where it ends. Due to the inherent
characteristic elements of hypertexts, users can easily jump from one type
of text to another by using hyperlinks.
Hyperlinks are extremely important elements in hypertexts. After
mentioning the classification of hyperlinks suggested by Storrer (p. 183)
in intratextual, intertextual, and extratextual, the author proposes a
classification of web ads based on their visual perception. In fact,
hyperlinks can be text, image, and interactive (e.g., pull-down menus, or
search boxes). Web ads show in most cases a combination of these.
Furthermore, hyperlinks have mainly three functions: 1) persuade and
motivate users to click, which is especially important for commercial web
ads; 2) create a connection between two units; 3) meet users' expectations
on the linked web page. The meaning of hyperlinks creates and influences
comprehension and the selection of users' path. If users do not understand
the meaning of a trigger, they will probably turn their attention
somewhere else. Since hypertexts provide several paths, grammatical
cohesion across sentences or textual units is difficult to accomplish.
Moreover, most hyperlinks are not expressed as whole sentences, but often
as single words or short phrases. As coherence in hypertexts is process-
oriented, hyperlinks work as information points which create a semantic
chain when connected. In hypertext coherence depends on the users creating
their individual coherent path along these semantic points, i.e. it is a
self-selected path. However, this freedom of choice is only partly true.
The selection and activation of hyperlinks is restricted to those
hyperlinks which are actually provided, and therefore, pre-selected by the
Chapter 7 ("Summary and Conclusions") sums up and gives the overall
picture of the study and concludes with a sensible statement: "the
structure of information needs to follow the users' cognitive capacity.
Not everything that is technically feasible corresponds to what users can
or want to cope with" (p. 198).
This book has many merits. It offers a useful tool of analysis of
textuality on the Web. Even though it focuses on one type of texts only,
web ads, by a contrastive and similarity approach it makes useful
connections with other types of textuality, both traditional, e.g.
printed, and non-traditional, such as emails or chats. Concepts
like "conceptual orality", "trigger words", "self-selected path", "open-
endness", "banner blindness", etc. represent useful points of reference
for the description of hypertexts in general. A classification of
traditional advertising (which includes print ads, TV commercials, direct
mailing and coupon ads) and online advertising (which includes static,
animated and interactive web ads, plus special types such as pop-up ads
and web ad traps) is an important effort of generalization and a useful
starting point for further research. Linguistic features associated with
web ads are a valuable asset for further investigations on (automatic) web
genre analysis. Also the analysis of hyperlinks is important to integrate
studies where the attributes of functionality and interactivity (both
based on the technical possibilities offered by hyperlinking on the Web)
are suggested as special features of web documents (see for example,
Shepherd and Watters 1999 or Crowston and Williams 1999).Original
contributions include a new model for communication (the interactive mass
communication model, p. 96 ff.); a new model for hypertext structure (p.
172 ff.), a visual conception of hyperlinks (p. 178).
What seems lacking is a more detailed description of the corpus of the web
ads analyzed in the study. Even if the analysis is merely qualitative (p.
4 ff.), it would be an added value to know, not only the range of years
covered by the web ads, but also the number of them. It would be useful to
have them collected and described in an appendix and grouped them into
families, where different types and persuasive strategies are highlighted
and illustrated. The book is virtually flawless. I could spot only a
single typo on p. 177 ("its seems impossible").
Crowston, K. and Williams M. (1999), "The effects of linking on genres of
Web documents" Presented at HICSS-99, Kilea, Hawai'i, January 1999.
Shepherd M. and Watters C. (1999), "The Functionality Attribute of
Cybergenres",Proceedings of the 32nd Hawaii International Conference on
System Sciences (HICSS-32).
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Marina Santini is a PhD candidate in Computational Linguistics at ITRI
(Information Technology Research Institute), University of Brighton, UK.
Her general research interests include computational analysis of text
types and genres, and her specific research project focuses on automatic
identification of web genres.