Do you suffer from ‘banner blindness’? If so, you are a typical web user who doesn’t click on the standard 460 by 68 pixel ad banner or even register their existence. Eye tracking studies such as that for online newspapers (http://www.poynterextra.org/et/i.htm) shows that when we first load a web page, our eyes are drawn first to headlines and then graphics.
Partly as a result of banner blindness, there has been a dramatic decline in average click-through rates (the percentage of people viewing an ad who click on it) from well over 1% in the mid 1990s to a fraction of a percent today. The latest average click-through rate for traditional banners is just 0.27% according to Q3 2002 Doubleclick Ad serving trend report for the US.
Given this decline in click-through rate, it is unsurprising that many marketers ignore online advertising as a means of reaching potential and existing customers. However, this article will suggest this may be misguided.
We will see that there is more to online ad effectiveness than click-through. Furthermore, new rich media formats are producing significantly higher responses – an average click-through of 2.5% was achieved in the 400 million rich media ads delivered by Doubleclick in Q3 2002.
Why advertise online?
This may seem like a strange question, but we have become so pre-occupied by click-through rates, that I think the question needs to be asked.
Reach and media consumption
One reason for advertising online is that more and more of us are spending more and more of our time online.
In 2001, Freeserve commissioned a year long study of 1,000 Internet users over 16. It found that of the 6 hours average daily media consumption, using the web at 45 minutes was way ahead of newspapers and magazine readership, although as expected, trailing TV and radio.
Perhaps more significantly we can closely target our advertising according to specific channels, content sections or keywords typed in by visitors to the type of sites in the table. The Doubleclick survey referred to above showed that three quarters of ads placed are keyword or content-based.
Targeting according to time of week, day and browser type are also possible, but much less widely used. Targeting by geography is relatively straightforward at a national level, but is less easy to target regions within a country unless a site servers a local area or visitors to a site have been profiled according to geography.
Direct response action
Returning to the question we posed at the start of this section, achieving click-through to a site was the main reason for the original upsurge in banner advertising in the mid 1990s. Online retailers will be looking to achieve a sale when the viewer of the banner clicks through to the destination site. For many other companies, particularly B2B companies, a direct sale will not be possible, so the purpose may be to obtain a sales lead. For example, technology companies such as Oracle and IBM use banner ads often. A typical approach is to offer a report or industry analysis in return for contact details and profile information being provided.
Post-impression action
Although the low to average click-through rate of banner ads is well known, one of the best-kept secrets about banner ads is that although immediate click-throughs may be low, later visits to a web site may give at least as many visitors again. By using cookies, it is possible to track whether ad viewers subsequently visit a web site. Such analysis suggests this is a common occurrence, so it should be accounted for when budgeting and assessing the return on investment for banner advertising.
The number of post-impression visits to destination sites suggests that banner ads do have some effect in raising awareness and this is indeed the case.
In one of the most extensive studies conducted, involving 13,000 consumers, the effect of cross media online, print and TV advertising for the Dove soap bar showed that spending 15% of the overall marketing budget on the Web resulted in a 24% lift in branding impact.
Further evidence of the capability of the Internet to increase awareness is provided by a branding campaign run by Lloyds TSB on Freeserve. The aim of this was to re-inforce the perception of brand values communicated in offline ads, i.e.
‘a modern, innovative and helpful source in the financial stages of ones life’
There are now a staggering range of online ad formats with confusing proprietary names such as Eyeblaster, Shoshkele and Overts. Essentially these are animated ads appearing over the page which then move to give a final message. They definitely can’t be ignored as can be seen by the rich media examples below. Media owners tend to limit their use to one or two times per user per page to avoid the annoyance factor being too high.
The main options for online ad formats are:
- Static – they don’t change through time – commonly used for sponsorship of a content section.
- Animated – the norm for keyword related placements on portals with a typical a rotation of three to five different images
- Interactive – the user can type in an e-mail address to register for information or use a calculator for a loan for example.
- Rich Media – using a combination of animation, or less commonly video or even sound. One of the most popular providers of these ads is Eyeblaster (LINK: http://www.eyeblaster.com/WebSite/Gallery.htm) while in the UK, the Tangozebra agency (LINK: www.tangozebra.com/adformat) has been active in using these formats with major brands.
- Pop-ups – these include interstitials which appear between pages and superstitials that appear in an additional browser window. Pop-unders that appear below the current window are another variant.
The effect of ad size on effectiveness
Static and animated ads come in a range of sizes. The standard 468 by 60 pixel is still most important, but large rectangles of 120 by 140 and 120 by 600 ‘skyscarapers’ are becoming more important. Research suggests these larger adverts are much more effective in generating awareness and association.
According to the Avantmarketer report ‘What works in Internet advertising’, the effectiveness of advertising in terms of message association can be increased as follows:
· Banner: 19%
· Skyscraper: 60%
· Large rectangle: 73%
Expandable banners such as those at MSN (www.msn.co.uk) exploit these differences.
The effect of ad type on effectiveness
The rich media ads referred to above have even more impact than large ad formats. A study commissioned by Microsoft published at: LINK: http://advantage.msn.com/docs/Ad_Effectiveness.pdf shows the following increase in message association with ad type:
· Banner (18%)
· Large format rectangle (31%)
· Flash (34%)
· DHTML (39%) – an that moves across the screen
Creative guidelines
The execution of individual campaigns will obviously also significantly affect their effectiveness. Since creative elements are limited online, it is possible to suggest general rules for creative strategies. These from Rex Briggs given in the Avantmarketer Internet advertising effectiveness report (www.avantmarketer.com) are instructive:
· Brand the first frame
· Brand top of skyscrapers
· Ditch ‘Click here!’ Although once effective, the evidence suggests that this signals to consumers to ignore Internet ads.
· Use High contrast of text and background.
· Use < 15 ad elements
· Use large logos. At least 14% of the creative space is suggested.
· Include a human face if appropriate
However, the ‘6 times rule’ from Internet advertising specialists at AtlasDMT (LINK: www.atlasdmt.com) should be remembered. This states that CTRs vary 6 times more according to placement than creative.
Buying online ads
Banner advertising is typically paid for according to the number of site visitors who view a web page and the advertisement on it. These are the ‘ad impressions’ referred to earlier. Cost is then calculated as CPM or cost per thousand (mille) ad impressions. This still rules the rate cards. Since there is much spare inventory, the rate cards are not meaningful, and some larger advertisers may be able to negotiate a CPC (cost per click) deal.
One of the benefits of online ads is that it is possible to change the creative and placements through time. If a placement or creative is not performing, it can be removed from the campaign. So, a typical approach is to start wide and then narrow the focus. Alternatively, 20% of budget can be reserved for high performers (with high CTR and conversion rates). As with ads in other media, the frequency needs to be assessed effectively – the studies referred to above show that ads are most effective when viewed at least 3 times by an individual.
Fellow contributor to WNIM, James Colborn [james.colborn@btinternet.com], advises using an autorouter to find the most effective landing page on a destination site. Say for example, a retailer was running a Christmas Campaign Pay Per Click Campaign in eSpotting - ‘Christmas gifts for her’ an autorouter could direct different visitors as follows:
· 1st person goes to – perfume
· 2nd person goes to – bathroom products (soap etc)
· 3rd person goes to – christmas gift finder
This is then run over a sample number of people, 999 usually (333 per page) to find out which one drives the greatest number of sales. That destination page is then used for all subsequent ads served. Taking this one step further, different search engines have different audiences and demographic profiles, so James uses this technique to determine the most successful landing pages for each and every type of search engine, banner ad, affiliate network, email etc.
Pop-ups
No exploration of online ads would be complete without a quick look at pop-up ads since many web site users are vehement in their dislike of these ads. Indeed, the US arm of women's portal iVillage (www.ivillage.com) recently announced it was to ban pop-ups, due to a consumer backlash against what is arguably the most intrusive online advertising format. However, in the UK arm (www.ivillage.co.uk) has not followed suit. In the UK, well-respected sites such as iVillage, The Guardian and FT.com continue to use them since they are dependent on the revenue from advertisers and advertisers are still willing to pay for this form of placement. With some sites such as the FT the restricting free content and extending subscription services, there may be a price for avoiding ads. Alternatively you could become one of the handful of technical whizzes who download ad-blocking software.
Pop-ups will continue evolving and annoying. Recently some advertisers have started using Gator (www.gator.com) and Whenu (www.whenu.com/demo) which require software downloads, to display an ad offering to beat a price when a visitor is on a competitor site!
Conclusion
If you are still not convinced about using online ads, remember that many companies are effectively using ad-style graphics and pop-ups on their own sites to good effect. For example, Lastminute.com, which has collected over 5 million e-mail addresses, and incidentally has just entered profitability and the FTSE 250, uses pop-ups to encourage e-newsletter subscription. Meanwhile, Amazon tempts first time visitors to its site (which it recognises since their PC does not have an Amazon cookie), with a pop-up window offering a £3 discount on the first order. Well targeted ad-style messages and pop-ups with relevant offers definitely work, whatever your opinion about them.