By: Stan Bogdanov

September 8th, 2003

Reader Rating: 5.5

Almost every Webmaster has tried a banner exchange at least once. While many have ideas about what they might expect the exchange to provide, only a few know how these exchanges really work. Only these few can evaluate the benefits the exchange promotes against the information that remains undisclosed to public. The purpose of this article is to make the 'black box' of banner exchange more transparent to those who want to know.

To begin, let's sum up the main features a typical Webmaster is likely to consider attractive in a banner exchange network.

The most popular seem to be:

• Extensive reach
• Tight quality control
• High exchange ratio
• Good targeting
• Handles multiple banner sizes
• Provides detailed statistics
• Fast (or at the very lest flawless) ad loading

Let's review each of these in turn, and find out what can potentially lurk behind each seemingly attractive feature.

Extensive Reach and Tight Quality Control

These two features are closely related. Quality control becomes extremely time consuming and costly as the network (and, respectively, its reach) grows.

It's not difficult at all for the banner exchange owner to verify banner placement of their ads on the static pages of several dozen sites. But when the sites number in the hundreds, an advertiser’s banner may be spread across many thousands of pages. How often can a single person (or two) check them all, even if, between the hours of 9 and 6 they do nothing else?

The calculation is not difficult. Let's assume that it takes the banner exchange owner half a minute to check the page. That makes 120 pages per hour or 960 (or 1,000 at the very best) during the working day. For a network with, say, 100,000 Web pages, one QA staff member will only be able to check a particular page about tree times a year… Not frequent enough to maintain any real quality control. Four staff members will lower this frequency to once a month -- still not enough! In fact, only when the banner exchange has 16 full time QA staff members can the network owner have the right to claim that banner placement is under constant monitoring.

Definitely, manual labor needs to be enhanced by automated controls, but any script -- even the most intelligent -- can be fooled with enough skill, knowledge, time and patience. Thus, to achieve tight quality control levels, manual checks are irreplaceable. More sophisticated methods such as, for example, double random sampling can save a lot of moderator effort. However, such techniques require staff to code and support them, as well skilled specialists to interpret the results and direct the QA checks further. Thus, whether the exchange owners has 16 moderators with decent scripts, or top-notch scripts, a skilled administrator and 4 or 5 moderators, the expense of running the system is likely to be considerable.

Few networks will make their reach data publicly available, however, there is still a way to evaluate it. All you need to do is to find out what software a particular exchange employs, and what hosting option they use. The use of standard, publicly-available cgi scripts based on a MySQL database signals that the network may be incapable of achieving decent reach and effectively managing campaigns at the same time.

Need proof? Let's look at the numbers.

Let’s assume that at peak load the network needs to process 1000 impression requests per second (which is quite common for a network that provides 30 million impressions daily). In this case, about 50,000 connections must be open and served flawlessly at any given moment.

What might this mean in terms of standard scripts and hosting options?

First and foremost, it means that the system kernel has to be optimized to enable the handling of such a large number of requests at one time. The base configuration of any OS simply cannot handle that many file descriptors simultaneously.

Second, most publicly available scripts use the default configuration apache server, which can only handle up to 256 connections simultaneously. Is this enough to provide decent reach? It’s doubtful. Moreover, apache forks on a response, taking up to 1MB of memory per copy (i.e. per request). Is this kind of space likely to be available from that network’s particular provider?

Cgi request processing under a standard SQL server puts additional strain upon a banner exchange. Find out where the exchange is hosted and ask the provider's support team how many times per second you might be allowed to run your cgi script. It's almost safe to predict that the figure will be something like 10 per second at the most for co-located exchanges, though options may vary significantly for those with dedicated hosting. Compare these values with those presented above, and draw your own conclusions.

The issues of reach and quality control can be partially addressed if the exchange owners increase the number of servers they use and implement some kind of balancer software. However, in the case of banner exchanges, such techniques can result in a lot of extra (and irresolvable) problems in the long run.

High Exchange Ratio

Commission is the lifeblood of any exchange network. If a network operates honestly, the only source of income it has is the sale of commissions. When considering which network to join, you should always remember to compare the rates offered with their affordability for the exchange owner. Only sign up if the figures meet your expectations and don't look contradictive.

But let's start with the evaluation technique. No prominent exchange can ever be run efficiently (or for long) by fewer than 4 people: an administrator, a moderator/support person, a programmer and a promo/development staff member. And each of these people expects to be paid at the very least a competitive rate for their hard work. As a potential advertiser, you need to find out the price for placing commercial ads with that network, and evaluate the network’s profits against its operation costs.

A word of caution here -- the Internet spreads not only across national borders, but also across different economies. The quality of IT expertise does not consistently correlate with the expense to business of employing those skills; therefore, the cost of what are apparently comparable services to customers can vary widely between countries and service providers. Look at it this way: a high-quality IT professional might be able to earn $10,000 per month in one location, yet another person with the same skills and expertise might be content, if not happy, with a monthly wage of $500 in another country.

Thus, it's a very difficult task to evaluate the value of commissions against the business’s expense figures. Yet there is a rule of a thumb -- no exchange, wherever it might be based, can survive without receiving at least 20% commission at a decent reach.

Note that it's not always an easy task to derive the actual commission rate an exchange delivers, as it can vary depending upon different factors (CTR being the most popular one). Only the relative traffic volumes of differently "taxed" groups of sites can actually help you identify the true rate.

If it’s not possible to estimate the actual (as opposed to advertised) commission rate of a particular exchange, your best bet is to consider a different network. If, on the other hand, you can estimate the real commission, but the figure looks too lucrative to be true, there are two possible explanations:

1. The project is on budget and does not need to break even for the time being. This is most likely the case if another profitable business is behind the exchange. Such cases are usually identified as such quite plainly on the site as "limited time offers", or "promotional offers" etc.

2. The exchange breaks even at the expense of its participants by selling more than it has a right to. This means the actual exchange rates differ from those declared on the site.

For example, if the network only shows 800,000 banners daily (24 million per month) and needs to sell 10 million just to survive, it means that the actual exchange rate for its members is less than 2 to 3. Members may receive 3 credits for each 4 banners shown, but some of them are just accumulated on accounts and cannot be spent. The longer things continue like that, the less likely it is that these credits will ever be useable at all.

Targeting

Another feature on the advertiser’s “most wanted” list, and rightfully so, is targeting for banner exchange members. The general attitude is "the more options, the better". But more options come at a price…

The root of the problem lies in the fact that every member contributes untargeted traffic to the network, but every campaign that’s run within the network targets a particular audience.

Let's assume, for example, that we have a group of Canadian sites in a network oriented heavily towards a Canadian audience. In reality, less than 80% of the combined audience for these sites will comprise Canadians. The other 80% will arrive at the site from all over the world. Assuming the GEO targeting option is available to every exchange member (and each has it set to “Canadians only”), if the sites receive aggregate daily traffic of 100,000 users, 20,000 of these users represent traffic that members produce but do not utilize.

Now let’s add the network itself, which has a sales department that’s also based in Canada, and assume it offers a meager commission of 20% (4 to 5 ratio). It should be obvious that this share is most likely of interest to Canadian companies who want to reach Canadian audience. The network needs to sell the entire lot just to survive. This cuts the total members pool down to 80,000, one quarter of which is 'unwanted' traffic that’s used by neither network nor its members.

The situation is similar to one described earlier: the members receive 4 credits for each 5 banners shown, but only 3 of these credits can be spent without violating the targeting settings; the remaining credits become a dead weight on accounts.

There are 3 ways out of such a situation:

1. Increase the network commission (thus lowering the exchange rate), but do not sell the extras. Instead, use them for balancing purpose only. In our example, this would mean that the official rate goes down to 3 for 5, but the network sells only half the commission and the rest are used to show some general ads (usually promoting the exchange itself) to the untargeted audience.

2. Charge an extra commission for the use of targeting. This is a derivative of the above solution that differentiates between members on the basis of their campaign settings. So, if a member site wants Canadian traffic only, that site gets it at 3 for 4 instead of basic 4 for 5, which provides the network with balancing capabilities. Those sites that are ready to accept any traffic get (and more importantly, spend) 4 for 5, while helping the network balance ads and audiences.

3. Force network members to show some untargeted banners, and provide free targeting option for all other ads. Returning to our example, 3 banners out of the earned 4 are shown to Canadians, while the fourth is forcefully shown whenever a banner request comes from a foreign visitor.

Other kinds of targeting work in almost the same way. The only notable exception is frequency capping, which calls for a separate database (and quite a "weighty" one, if the solution is implemented properly). In addition to aforementioned impacts on a network, this option also requires extra investments into memory, disc space, or even a separate server.

Multiple Banner Sizes

Another feature that’s as tricky as it is useful. The trick is that it is almost impossible to balance a network that operates with multiple banner sizes.

Once again, let's consider a very simple example. Imagine we have three sites and three banner sizes (large, medium, and small). The first site shows only small banners, but wants to advertise with large ones. The second site opts for medium banners on its pages. The third site offers space for large ads, but wants to run a campaign using medium sized ads. Is it possible to run such a network smoothly?

Actually, there are 2 types of multi-size networks:

1. truly multi-size, like the one described above (which can never guarantee smooth running of campaigns, nor complete use of ad spaces)

2. pseudo-multi-size (which is actually a number of separate networks united under the same brand with the interchange of impression credits between the networks)

The latter is a much more balanced network, because the possibility for members to use conversion rates for credit transfer between different banner sizes gives the network administration an added degree of control. On the other hand, when considering participation in such a network, Webmasters should evaluate each of the sub-networks separately, unless they are ready to accept numerous similar plug banners on their pages, or face problems when carrying out their own campaigns.

Thus, size doesn’t matter. What matters is how you use it. Or, more specifically, what matters is how they let you use it. It’s no secret that Web surfers are always evolving, which is why those "hit the monkey" campaigns work no more. Internet advertising has changed. Graphic banners can still perform well …if we're talking brand awareness. However, if you need to build recurring traffic, you might want to enhance your online ad campaign with capabilities offered by other media.

Now, the banner exchange site might make all manner of claims about supported ad sizes and formats, but if you want to be sure, look at the network’s HTML code. If it is nothing other than <a href src=> and <image src=> tags, the network is limited to simple graphics -- whatever the site or support team might claim. If the HTML contains the iframe tag, then take a closer look The iframe tag is supported by almost all major browsers and gives you two substantial advantages:

1. it lets you use advanced media types (frequently referred to as "rich-media") including HTML, Flash, javascript, cgi forms and more

2. your page loads before the content of iframe. So, even if your ad server is inaccessible to a particular visitor, it will not hamper their ad surfing

There’s an added benefit as well. Often, the Google bot indexes sites it finds via another site’s iframe. This may not be the case with every site, but it happens nevertheless.

Additionally, you should look carefully at the number of banners allowed by an exchange network -- the lower the number, the more trouble you'll have running your campaigns. The reason for this is very simple. For the sake of the exchange’s quality, each banner must be moderated. If, in order to run a new banner, you're forced to replace an old one, your campaign will suffer from being stopped (or slowed down) until the new ad is approved. The higher the number of banners that are allowed simultaneously, the lesser the impact that any ad moderation delays will have. Only the allowance of an unlimited (or almost unlimited) number of banners per campaign will grant you the flexibility to avoid the ill-effects of potential delays in moderation.

Detailed Statistics

This one is plain and simple:

• The more statistic an exchange provides, and the longer these stats are available to members, the more it costs the network in terms of storage space (for the stats archive) and operating memory (for the presentation of real-time data). This should be taken into account in your evaluation of the network’s ability to deliver the exchange rate promised.
• A greater number of targeting options granted by the network means more stats for member sites. As such, the availability of statistics should be evaluated against the targeting options available.

Fast Loading of Ads

The main factor that contributes to overall ad loading time is the process by which ads are chosen. There are other factors as well, but generally these are beyond control of exchange creators.

There are two ways to speed up ad choice processing.

The first is pre-calculation. Here, the system knows what banner is to be shown to a particular visitor long before he opens the page. In terms of speed, it's fine. In terms of flexibility, however, such an approach hits the exchange member badly. Advertisers can only stop or alter a particular campaign after all the pre-calculated impressions have been displayed, which means the delays in your requests can last from several hours, to a day -- or even more.

The second option allows real time management of ads, but places tremendous demands upon exchange creators. Why? Because this option requires:

• Fine-tuning of the OS kernel to adapt it to the real-time processing of multiple (and I mean tens of thousands of) connections that are served at the same time.
• A specialized Web-server that doesn't fork to process requests and can support many thousands of processes simultaneously.
• Custom-written databases, especially tailored to serve huge numbers of requests quickly, with flawless replication and scalability options far beyond those provided by general MySQL.
• Very specific hosting conditions (more often than not, including a separate data center with several servers for optimal scalability).

Conclusion

The aim of this article was to provide you with a tool to evaluate your future in a particular exchange network against the promotional statements and marketing promises that appear on the network’s site.

To simplify the task further, a brief evaluation of several exchange networks is presented overleaf.

Due to the fact that almost no networks provide any tangible information about their reach, or make their member lists publicly available, the reach and quality evaluation is omitted from this evaluation.

The network listing is in alphabetical order, and the evaluations are either absolute or grade-based.

The grades include:

1. non-existent
   
2. primitive/amateurish
   
3. adequate
   
4. professional

Network: 1adexchange
URL: www.1adexchange.com [1]
Exchange Ratio: 1 for 2
Statistics: 4
Targeting: 2
Banner Sizes: 468x60
Rich Media: not supported
Max Banners: na
Interface: 4

Network: adbunker
URL: www.adbunker.com [2]
Exchange Ratio: 1 for 2, 4 for 5
Statistics: 2
Targeting: 2
Banner Sizes: 468x60
Rich Media: not supported
Max Banners: na
Interface: 2

Network: adendum
URL: www.adendum.com [3]
Exchange Ratio: 2 for 3
Statistics: 2
Targeting: 3**
Banner Sizes: 468x60
Rich Media: not supported
Max Banners: na
Interface: 1

Network: bcentral
URL: www.bcentral.com [4]
Exchange Ratio: 1 for 2
Statistics: 2
Targeting: 3
Banner Sizes: 468x60
Rich Media: not supported
Max Banners: na
Interface: 1

Network: e-bannerx
URL: www.ebannerx.com [5]
Exchange Ratio: 2 for 3
Statistics: 2
Targeting: 4
Banner Sizes: 468x60, 120x60, 120x600, 125x125
Rich Media: supported
Max Banners: na
Interface: 1

Network: neobanners
URL: www.neobanners.com [6]
Exchange Ratio: 1 for 2
Statistics: 3
Targeting: na
Banner Sizes: 468x60
Rich Media: not supported
Max Banners: 8
Interface: 3

Network: hitexchange
URL: www.hitx.com [7]
Exchange Ratio: Flexible
Statistics: 1
Targeting: 2
Banner Sizes: 468x60, 400x40, 400x50
Rich Media: not supported
Max Banners: 8
Interface: 3

Network: exchangead
URL: www.exchangead.com [8]
Exchange Ratio: 1 for 2, 3 for 4
Statistics: 2
Targeting: 2
Banner Sizes: 468x60
Rich Media: not supported
Max Banners: 8
Interface: 2

Network: free-banners
URL: www.free-banners.com [9]
Exchange Ratio: 1 for 2
Statistics: 3
Targeting: 3
Banner Sizes: 468x60
Rich Media: not supported
Max Banners: na
Interface: 2

Network: linkbuddies
URL: www.linkbuddies.com [10]
Exchange Ratio: 1 for 2
Statistics: 2
Targeting: 3
Banner Sizes: 468x60
Rich Media: not supported
Max Banners: na
Interface: 2

Network: probe.prohosting
URL: www.probe.prohosting.com [11]
Exchange Ratio: 1 for 2
Statistics: 3
Targeting: 1
Banner Sizes: 468x60
Rich Media: not supported
Max Banners: 5
Interface: 1

Network: rle-network
URL: www.rle-network.com [12]
Exchange Ratio: 4 for 5
Statistics: 4
Targeting: 4
Banner Sizes: 468x60
Rich Media: supported
Max Banners: unlimited
Interface: 4

Network: trafficflame
URL: www.trafficflame.com [13]
Exchange Ratio: 4 for 5
Statistics: 2
Targeting: na
Banner Sizes: 468x60, 120x600
Rich Media: not supported
Max Banners: na
Interface: 1

Network: vesperexchange
URL: www.vesperexchange.com [14]
Exchange Ratio: 1 for 2 (basic)
Statistics: 2
Targeting: 2
Banner Sizes: 468x60
Rich Media: not supported
Max Banners: na
Interface: 2

Network: worldbe
URL: www.worldbe.com [15]
Exchange Ratio: 1 for 2, 1 for CTR-dependent
Statistics: 2
Targeting: 2
Banner Sizes: 468x60
Rich Media: not supported
Max Banners: unlimited
Interface: 2

* Maximum number of banners allowed by the network is not only a limitation to the participant, but also an indication that the network is likely to be facing (or might soon face) productivity problems. Setting limits for the number of banners allowed per member is an obvious attempt to resolve such problems at the expense of members, which is never justified.

** I can’t help wondering how targeting by age can possibly work… there’s something wrong with that statement!

[1] http://www.1adexchange.com
[2] http://www.adbunker.com
[3] http://www.adendum.com
[4] http://www.bcentral.com
[5] http://www.e-bannerx.com
[6] http://www.neobanners.com
[7] http://www.hitx.com
[8] http://www.exchangead.com
[9] http://www.free-banners.com
[10] http://www.linkbuddies.com
[11] http://www.probe.prohosting.com
[12] http://www.rle-network.com
[13] http://trafficflame.com
[14] http://vesperexchange.com
[15] http://worldbe.com