Banner Advertising - is it worthwhile at all?

[derekwong28]

We ran a very successful advertising campaign last month where we spent almost $3000 on Google, Overture, Sprinks and other PPC engines. Our conversion rates were as follows.

Google Search Results - 1 in 200
Yahoo directory - 1 in 100
PPC including Google adwords - 1 in 20 to 1 in 40
Highly Targeted text ads on content sites - 1 in 10 to 1 in 20

This month, we started to run banner campaigns on targeted sites with and the results have been disastrous.

A CPM campaign with tribal fusion ended with a clickthrough rate of 0.32% and an overall Cost per click of US$2.63. Needless to say, we have abandoned plans for a $5000 campaign before christmas.

After receiving 1000 clicks from Fastclick at a rate of $0.35 per click, we have only been able to track a single conversion on their reporting.

As a conversion rate of more than 1 in 1000, this would be far worse than a click from a general Google search result. I have really expected a conversion rate of around 1 in 20 because this was what we were getting from highly targed text ads. I wonder whether the fastclick tracking pixel is missing some sales, or really a click from banner advertising is really that bad?

[khuntley]

derekwong28,
There are very few situations in which banner advertising will be profitable. Two factors generally need to be in place. There must be a nationally known brand or company involved. There must also be an irresistable call to action. An example ad might be something like:

Nationallly Known Brand 40% off this weekend at NationallyKnownRetailCompany.com

Kevin

[derekwong28]

Hi, I am begginning to come round to this. What I can't get is why a click from a text ad have a conversion rate of 1 in 20, whereas a click from a banner has a rate of more than 1 in 1000?

[mack]

"What I can't get is why a click from a text ad have a conversion rate of 1 in 20, whereas a click from a banner has a rate of more than 1 in 1000?"

I think users are becoming so used to banners that they might as well be invisable. What they are looking for is content. If you can have your advert as text within the content then it is more likely to catch their imagination.
Ask yourself. when did you last click a banner and when did you last click a text link.

Mack.

[derekwong28]

The issue is that once a user have clicked on a banner and got to the website, why would they be less likely to buy then after having clicked on a text link? I have expected a purchase rate of around 1 in 20 to 1 in 50 per click but have thought it would be as bad as 1 in 1000.

The service from Fastclick has been excellent, they are honest enough to provide a tracking pixel. I have nowsuspended the campaign pending a redesign of the creative.

[powerstar]

If you convert 1/20 or 1/40 then 1/1000 could not and will not be real traffic.

[Perplexed]

Do most people even look at banners anymore? The only time I read one is if it loads much quicker than the rest of the page... and then I tend to think it was designed that way on purpose and I would not click on it out of sheer principal.

[Jenstar]

It also depends on the site and how well the banners fit the content. I have one site that has banner ad spaces available for many advertisers, and I get many repeat advertisers saying they got an excellent response, which goes against what I know about banners and their effectiveness ;) But they are banners perfectly suited to the audience and the pages. So there are always exceptions to the rule.

[rcjordan]

>get many repeat advertisers saying they got an excellent response

(re 468x60 top banners) Same here. Effectiveness is down from 1998, to be sure, but CTR really bottomed out several years ago and has been holding steady since then. I've always used the cost of acquisition from direct mail campaigns for cost comparisons and that still seems to work as a realistic benchmark.

Because I had 468x60 top banners set up in the layout, that's what I used in Google's adsense program. Performance has been good, slightly better than most graphic creative, but not out of range.

[Jenstar]

That is exactly what I did, it was quite convenient the AdSense size was the same. Ironically, advertisers have booked nearly all those banner ad spaces for next month. But I should be able to tell the difference between revenue from AdSense and regular banner advertisers by the end of the month.

[edit_g]

Banner ads are still selling like hotcakes on some sites, but ctr and conversion - forget about it... Banners can be used well to promote a very good offer from a well known brand (like khuntley said) and they can be used for branding. This is the way the market is going at the moment IMH.

Rich media and expanding banners are taking over from the standard banners anyhow - on most of the big sites (yahoo, wired, msn etc). I've seen some astonishing clickthrough rates on rich media and expanding banners.

[rcjordan]

>Banners can be used well to promote a very good offer from a well known brand

Actually, I've found (both as a publisher and an advertiser) that banner-blindness has less to do with banner position and footprint and is more dependent upon the creative than many want to believe. I'm not running any "known brands" right now but custom, traditional-style banners do the job quite well enough (again, as compared to acquisition costs in other media) in my slots.

[edit_g]

more dependent upon the creative than many want to believe.

I agree completely - I'm coming at this as both a publisher and a buyer and I've found the same thing - the creative makes the difference.

But - the highest CTR I've seen on a banner is around 4% - the highest I've seen on a rich media creative is over 30%. So yeah - it is the creative - I just think that other creatives are the way forward.

[Jenstar]

custom, traditional-style banners do the job quite well enough

The same here. All the advertisers running banner ads are pretty small businesses without a brand behind them. But the ones that do well are the ones that look good. No annoying animated gif banners, no addesigner banners, no inferior looking banners. Some of them may have only spent $10 or $20 getting a banner designed, but they look good and so people notice them. The look of the banner will go a long way in terms of the banner's success on a site. You just have to find that perfect site to advertise on.

[rcjordan]

>quite well enough

Quite well enough? Aarrgh! Where's my proofreader?!?

>I just think that other creatives are the way forward.

Agreed. But two constraints, existing layout space and development budget, will keep the banner around for quite a while, imo. But, using that space (and back on topic), derekwong28 poses an interesting question:
What I can't get is why a click from a text ad have a conversion rate of 1 in 20, whereas a click from a (graphic -rcj) banner has a rate of more than 1 in 1000?

Note that he's not talking about the CTR, he's already got drawn the click --but he's seeing a difference in the conversion rates relative to the type of creative that brought the traffic to his site.

[edit_g]

It is a very hard question to answer.

But I'll take a stab! :)

Banner advertising:

The majority of banner advertising on the internet is in the vein of "click the monkey" or "internet alert - you've been spotted" adverts IMHO. Users may think: "ugh, if I click on that ad I might get 10 pop-up windows and I'll be asked to add the target site to my favourites". So - users, by instinct, distrust banner advertisments.

Banner ads are also common. By this I mean that banner ads are like TV advertising: they are easily ignored.

If you've ever tracked banner ads you'll know that the session times generated from banner ads are minute and the "back button" rate is through the roof.

Text based advertising:

Maybe users feel that they "trust" it a bit more.

It is relatively new - and anything new online has novelty value.

Also - the text ads on content sites are sometimes seen as part of the content - the concept of "implied endorsement".

So that's my stab... Perhaps text ads are just innovative for the moment and they'll go the way of other ads soon - or perhaps they are just superior to standard media placements... Time will tell.

[netunity]

Why did text ads get better conversion rates then banners in this example?

What is the example of a text ad? What is the example of a banner?

In the absence of that information my guess is that the people who click on the text ads actually read them. Thus visitors who arrive to your site through the text ads are 'more right' for your site then the ones that arrive through banners.

[derekwong28]

Thanks for your input. I have suspended the fastclick campaign pending the creation of a new creative. I should have done it much earlier, but we just didn't believe the poor figures initially and suspected that the tracking pixel wasn't working properly. It is only after that we have compared our logs with the ip addresses of our customers that we knew the results were genuine.

I do not believe that we are receiving fraudulent clicks. Fastclick is a reputable company and have been very helpful. Otherwise, they would not have provided us a means of tracking the campaign with the use of a tracking pixel.

I think the ad we are running (120x600 skycraper) may have something to do with it. Because the ad integrates so well into the web sites, it may be prone to surfers clicking on it by accident.

Therefore the new creative will be much simpler. It will have fewer graphics but will stand out much more easily. It will promote a single product with a simple slogan, just as what we have been doing with text ads.

On that note, the text ads on content sites by Sprinks had been extremely effective, although we no longer receive much traffic from them because their ads are no longer shown on an important site. I will keep you posted as to whether there are improvements after the changes.

Derek