
		e-Marketing

Papers & Articles > Web Site Promotion > Banner ads

...Web Site Promotion

BANNER ADVERTISING
Banner advertising has attracted much heated debate - for example, whether it is overpriced, or simply ineffective.

Nonetheless, according to the Internet Advertising Bureau (27), the Internet advertising industry in the USA has grown every quarter for the last four and a half years, reaching $4.1 billion (£3.0 bn) for the first half of 2000. Banner ads currently account for 50% of all USA advertising (followed by sponsorships at 27%, classified ads at 7%, referrals 4%, interstitials 3%).

Ads appear in many different sizes (the 468 x 60 pixel still by far the most popular) in many different forms - e.g. static, animated, interactive, HTML, and incorporating rich media. As reported by CyberAtlas (M4 - European Ad Expenditures Up In August, September 2000), Forrester Research found the UK generated $65 million (£44.8 m) of the total $110 million (£75.9 m) spend in Europe for August 2000. Advertising rates however have dropped (an average of $2.70 or £1.86 for a full size banner down to $30.52 CPM, or ££21.05 CPM over the last six months). In the same period though European advertisers have increased their spending by 14% to an average of $299,000 (£206,210) per company.

Banner ads have come under increasing pressure from advertisers, because of 'click-through fatigue'. DoubleClick reported that in 1996 the top 25 ads in their network averaged 8% click-throughs, with some achieving 15%, and an average for all ads of 4%. This has now declined to an average of 0.36% overall in the USA (March 2000), according to figures from Nielsen/NetRatings (M18), a cause for much consternation among advertisers. Click-through rates are believed to be higher in Europe, possibly because of a 'fresher' audience.

There is a clearly measurable effect of declining interest in banner ads - once the novelty has worn off people seem to ignore banners. Click-through rates drop to 2% for the second exposure, then to 1% or less for the fourth exposure (Hoffman and Novak - 25).

Ad publishers however, claim that most of the beneficial effect of a banner ad is in its display, not the click-through. An Internet Advertising Bureau study (28) apparently showed that banner exposure alone was responsible for 96% of the 'brand enhancement', while a click-through only contributed 4%. What should not be forgotten is that a banner ad, in the world of the Internet, is a relatively crude expression - it is very limited in size, and must compete for attention against the real information content on its host page.

A banner ad display is a passive medium; only when a user clicks-through is the target material presented, which can be as complex and detailed as the medium of the WWW permits. Clearly advertisers want more than just a superficial brand image to be shown - the real message is on the other side of the banner.

A recent development in the industry is the full screen or pop up display known as an interstitial, or its further refinement, the superstitial. This presents a more TV like media rich advertisement, and while interstitials make the viewer wait for the maximum 100Kb file to download, superstitials are only sent in the 'quiet' periods between page requests, and are presented during a logical user initiated break during surfing.

Superstitials are very expensive - £8,000 or more for an agency to produce, and £60 to £80 CPM to publish, according to Costello (10). Early results have been very encouraging, British Airways for example reporting click-throughs for its new branding campaign in early 2000 of up to 30%, and Agency.com commenting that they achieved a 400% better response rate with the superstitial than a banner ad (Costello). These high response rates will decline as the novelty wears off, and it remains to be seen whether users will tolerate such interruptive advertising once they become more familiar with it.

Novak and Hoffman (26) make a very strong case for a radical shift in the advertising pricing model - away from the traditional exposure based pricing, to a 'pay for performance' model where ad costs are tied to resulting sales.

The WWW is the most measurable and accountable advertising medium ever, yet ad publishers are strenuously resisting any change from the traditional metric of flat rate, or audience reach pricing (i.e. CPM or 'per thousand impressions'). They claim they would be hostage to the creative element of the ad which would be beyond their control, if they were only paid on a CPC (Cost Per Click) or similar basis.

Schwartz (45) also questions the notion of charging for advertising 'space', and selling online 'inventory', when the amount of space online is infinite, and bears no comparison to traditional media.

With the increasing democratisation of the web, and the balance of power shifting towards the consumer, it seems reasonable that both advertisers and publishers should share the risks and the rewards.

If the right metrics are used, then much more targeted ads (in both design and distribution) can be developed to yield high response rates, and this would force both parties to make intelligent use of the vast amount of data that is available from tracking web traffic and behaviour. This would also make it much easier for advertisers to calculate their customer acquisition costs before making purchase decisions, and these would be much easier to justify on rational and financial grounds than at present.

Forrester Research estimate that by 2004, half of the projected $33 billion (£22.76 bn) online ad spend worldwide will be performance based. Perhaps the ultimate example of performance related low cost advertising is that of affiliate schemes, which are truly participative, unlike the present banner ad system.

Customer acquisition costs for a well run affiliates scheme tend to be much lower than for banner advertising. The current structure and method of pricing banner ads is rather irrational, and will doubtless have to change. In the mean time banner ads are still very successful for many, but require a higher budget than most other online promotional methods.

For example, the UK and Ireland Yahoo! site serves 140 million page views per month, and commands £4,000 for a banner on their home page for a week. Cheaper run of network ads start at £15 CPM, or £30 for specified categories, rising to £60 CPM for keyword dependent placement.

The most important consideration however, is precise targeting. It is better to have 100 page views with a click-through of 10%, than 1,000 views at the same price with a click-through of only 0.5%. Buying a keyword dependent placement at a higher price, which reaches a more select audience can be far more effective, but requires accurate knowledge of the audience profile and statistics.

As reported by Costello (9) for example, the wedding site www.confetti.co.uk advertised with Yahoo! and achieved click-through rates of over 10% for keyword targeted banner ads - by carefully segmenting their audience reach they were able to mount a very successful campaign.

The disparity between so many advertisers who suffer less than 1% click-throughs, and those who manage over 10% appears to be due primarily to the precision of audience segmentation and targeting. When better measurement standards and visitor segmentation data are introduced, as well as results oriented pricing schemes, then marketing management will be able to focus on long term promotion strategies, rather than the short term tactical media buys that are currently the norm.

Better segmentation information will also open up the market for small niche businesses, who will be able to advertise selectively to very narrow audience profiles at much more affordable rates.

Joining a banner exchange programme is an alternative way to advertise banners. This can be completely free, but operates on a barter principle, often on the basis of 2 for 1. Thus an advertiser will have one banner displayed on another member's site for every two banners shown on his own. The exchange organisation (e.g. Link Exchange, www.linkexchange.com) makes its money on the difference. For large schemes with sufficient members to permit well targeted advertising, e.g. by industry category at least, this can be a worthwhile exercise.

A very informative and comprehensive account of online advertising issues has been produced by Zeff and Aronson (53).