S P O N S O R S
- June 22 1996
- June 15 1996
- June 08 1996
- June 1 1996
- May 25 1996
- May 18 1996
- May 11 1996
- May 04 1996
- Apr 27 1996
- Apr 20 1996
- Apr 13 1996
- Apr 06 1996
- Mar 30 1996
- Mar 23 1996
- Mar 16 1996
- Mar 09 1996
- Mar 02 1996
- Feb 24, 1996
- Feb 17, 1996
- Feb 10, 1996
- Feb 3, 1996
- Jan 27, 1996
- Jan 20, 1996
- Jan 13, 1996
- Jan 06, 1996
- Dec 30, 1995
- Dec 23, 1995
- Dec 16, 1995
- Nov 25, 1995
- Nov 18, 1995
- Nov 11, 1995
- Nov 04, 1995
- Oct 28, 1995
- Oct 21, 1995
- Oct 14, 1995
- Sep 09, 1995
- Sep 02, 1995
- Aug 26, 1995
- Aug 19, 1995
- Aug 12, 1995
- Aug 05, 1995
- Aug 05, 1995
All material on this
website is the
property of IBN
(The Internet Business Network)
You may download
a copy for personal
All material on
this site is
© 1995. 1996 by IBN
- Back Issues, Weekly
- Week Ending: June 01, 1996
- June 1, 1996
- More Notes On Advertising Effectiveness
Advertising is one of the foundations of the continued advancement of the Web. Further moves ahead depend on ready sources of money to fuel continued innovation and experimentation. Funding by advertisers is certainly a large component of our continued existence. We're deeply thankful for the freedom we get from our advertisers' generosity and faith in our newsletters and readers.
On Thursday, we talked about the emerging capability to filter ads on the Web. Here's why giving users more control over the information that they consume is a good thing. (No, this isn't another mindless regurgitation of the tired "The Net Is Supposed to Be Free". Hardly)
Implicit in the free market consumption of information is the ability to chose. We're quite sure that anyone who doesn't want to see ads isn't going to be a likely customer for our advertisers. (Given the businesses we're in, we wonder if there is anybody who pays attention to us who feels this way). As an advertiser, you can be sure that we'd rather not pay for the privelege of reaching people who were not going to be interested in our products.
A significant problem with advertising, as an industry, has been its patterned inability to reach precise audiences. Advertising effectiveness rates are always single digit percentages. This is because one of the fundamental principles of advertising is to hit a broad spectrum, hoping to reach the interested few. As long as "carpet bombing" was the best precision available, everyone had to settle for minimal results. We're very sure that the ad-industry would change fairly quickly if there were reliable tools for forecasting the interests of specific concumers. After all, wouldn't you rather invest in reaching very likely targets?
So, when consumers "opt out" of a stream of advertising information, it simply helps to make the advertiser's investment more productive. The remaining audience is inherently more valuable. That makes us big fans of the potentials of performance based ads (payment for clicks).
Until the models for performance based payment are perfected, there are some simple things that can be done to ensure the reach of an advertiser's message. It's entirely possible to ensure that a web page is readable only if the graphics are downloaded. Without solid, and well understood principles for performance based advertising, this is a simple service to offer your advertisers.
- May 31, 1996
Alliances, on and off the web, can become the source of enduring success for your enterprise. In the Electronic Recruiting Industry (which we track very closely) E-Span is leading the pack in establishing innovative relationships centered around "distribution". That is, they've looked at their target markets very closely and found existing organizations that reach niches inside those targets. E-Span then makes royalty based agreements with the existing outlet. We're starting to call these background positions "engines". Our current darling, Tropical Jim's is providing similar services as an HTML / Graphics / Java backroom. We've used them and are tremendously impressed.
We expect that the distributed nature of the web, combined with the trend towards small and flat organizations, will make this strategy a succesful approach to focusing on core competence while achieving greater market penetration.
- May 30, 1996
- Filtering Ads
There's a brewing controversy over products like Internet Fast Forward. In beta testing currently, the product allows viewers to block certain images from being downloaded. It's an "ad filter".
From a design perspective, it would be smart to assume that this kind of application will become commonplace (just like remote controls for television sets). It simply means that the quality of web ads will be forced to improve. Giving users more control over the information they consume is generally a good thing.
The Federal Trade Commission will be holding a workshop on Consumer Privacy on the Global Information Infrastructure. There are also background rumblings about regulation coming to Internet Advertising. It's an election year in the States, expect anything.
- May 29, 1996
- Two Tugs, One Budget
We save the late night for worrying about whether or not we'll lose our technology edge. The web is an amazing world in which a project looks outdated as soon as its finished. The technical development side says "Don't release it yet, the latest release of (Name your favorite next greatest thing) is going to hit the streets next week." The marketing side says "Build mindshare now, an audience is forever." Where's the middle path?
The answer is unfortunately vague (it's a good thing that Tagamet has become an over the counter drug). Development has to remain a step ahead of the audience (not two, mind you). Marketing has to straddle both the "here and now" and the "what's coming soon". In this marketplace, the odds are high that a competitor will emerge tomorrow with technology that shames you. So, how do you make investment decisions.
One thing is beginning to be clear. Though Web technology and applications evolve in rapid cycles, harware investments run on consumer and business budget cycles. That means, for instance, that Java will become an institutional part of the web environment in the first quarter of 1997 (because of Holiday purchases and corporate budgets). The players who hang on through that shift will be releasing their "beta sites" in mid to late fall.
The most interesting thing to watch will be the rate at which browser usage converts from the 2.0 series to 3.0 (Microsoft is promising a product that is fully competitive with Netscape's 3.0 betas by mid June). You might look for signs of niche fragmentation in browser usage as well.
If Web usage continues to grow at current rates (that's a big if), we're liable to see a market that looks like PCs in the 80s (newer hardware purchasers get better technology cheaper and drive trhe market). If, on the other hand, the market behaves like publishing, old loyal customers can't be taken for granted in the same way that the software industry did back then. We think that this tension will force new business models out of the woodwork. It might even create new development models in which "legacy" isn't seen as an encumbrance but is viewed as an asset.
- May 28, 1996
- Advertising Effectiveness
We're learning a lot about what works and doesn't work in web advertising. It's a complex mix of timing, content, size and placement, duration and the goals of the advertiser. We're presenting the summary advertising report for our current clients over the last three weeks (It was generated with the tools we talked about yesterday.) Here are a few of the things we're figuring out
Holidays and seasons affect traffic in mix, volume and performance. For instance, our traffic was about 5% off last week due to the Memorial Day weekend in the United States. But, click-through rates (the number of people who click on an ad) were significantly higher. We think this is because the pace in offices slows as the holiday approaches.
Click-Through performance is clearly a function of the relevance of the ad to the audience that sees it. It's also shaped by the content of the ad. Direct and nearly blunt seems to work much more effectively than subtle and thought provoking.
Size and Placement
We've been surprised to discover that size and placement are not as important, overall. We have small ads, tucked well into the "button" section, that consistently produce very high click rates. We're starting to wonder if moving the ad around over time increases its productivity. The top and bottom of the page seem to work equally well for banner ads.
The number of times that a reader sees an ad seems to be related to click-through rates. We think that the optimal number of exposures per reader is somewhere above 10 over a period of months. Our longest running advertisers have shown steady growth in click rates in months four and five of their contracts. We expected that there would be saturation points after which returns diminished. Ineffective content and muted tones seem to offset the trend.
Goals of the Advertiser
Click-rates are not the only motivators. Many advertisers are trying to build brand awareness and name recognition for reasons other than site traffic. Display prominence and consistency in positioning with a constantly high number of impressions per visitor become the important issues in this case. It's critical to understand your advertiser's objectives
- May 27, 1996
- Tools That Work From Microsoft
Slowly but surely, Microsoft is winning us over. They're starting to provide real work tools for the web environment: Microsoft Internet Assistant for Word and the Internet Assistant for Excel. Both seem to be indicators of things to come and do require that you have the latest revisions of Word and Excel.
Each week, we bravely face the chore of analyzing the statistics for the overall interbiznet.com Website. With thousands of files, tens of thousands of visitors from thousands of discrete domains and some "legacy" problems, we plunge into the data trying to make sense. (It's worth noting that we do it manually because we think that most of the automated analysis tools miss key information. The top 100 domains requesting material from us only account for about 5% of our overall traffic.).
Assistant for Word is terribly slow as a daily Web Browser (it does read HTML and display graphics files). It does, though, provides a unified work environment for sorting and viewing server statistics logs. Those logs, put out by a generic "stats package" require additional processing before they're really ready for analysis. That's where Assistant for Word shines. Parsing and sifting the data is much easier without having to cut and paste between applications. (To get the data into the right shape, we take out a lot of spaces, add a couple and add a bunch of tabs so it can be poured into a spreadsheet.)
Then, we cut and paste into Microsoft Excel for some real analysis and standard reports
We've been spreadsheet "junkies" for 15 years (anyone remember Visicalc?) and find that the format is incredibly useful for delivering complex information that needs to be studied. Until now, delivering spreadsheet data on the web has been an awful and time consuming process. Internet Assistant for Excel does one thing pretty well (and it looks like Microsoft will continue to improve it). It translates data from Excel into HTML tables. Now, we can deliver data to our customers on the web, easily.
If you sift through data as a part of your marketing efforts, both of these free add-ins (for Wintel and Mac Systems) will serve you well. The download and installation instructions are clear and simple.
- May 26, 1996
- Graphic Mileage
Describe the last graphic you saw on the Web that you really needed. It's a minor percentage of what's out there. Given the precious nature of your customers' time, it's critical to ask yourself what each graphic adds to your message.
There is very little in the way of substantial information about how to effectively use graphics as communication tools on the web. The best approach is to develop your ability to tell what works and what doesn't by constantly reviewing and evaluating sites.
We're big fans of Edward Tufte (author of Envisioning Information and The Visual Display of Quantitative Information) He says:
The interesting thing about Tufte is that he writes about the effective use of paper, not computer screens or multimedia. He routinely gives seminars around the United States and is worth the time and price of admission. HIs books (at about $50 each) are widely available in the design section of the major bookstores. While you're there, take a moment to immerse yourself in the available literature on design. It's not much and thumbing through it may well improve your "eye".
- Low resolution graphics are totalitarian environments where information is tightly controlled.
- Clutter is not a problem of information, it is a problem of design. Don't throw away the
information, improve the design. Keep the richness and complexity of the information, aim for
clarity and simplicity of design.
- Gain screen real estate by cutting down on administrative overhead (stuff like key help and menu
Try this link: Edward Tufte. It should take you ta an AltaVista search for all references to Tufte on the Web. It can be a useful type of outbound link in certain settings.
Al Globus' short essay Principles of Information Display for Visualization Practitioners is densely packed and oriented towards the presentation of raw data. It's worth printing out and having handy.
- Contacting Us
- Call, fax, write, email. We'd love to talk about your project.