Back in September 2003, I wrote about several new revenue sources we were testing in an attempt to supplement our waning income stream. I want to report back on how well they and some other projects have performed, particularly since a number of people had inquired.
DealBITS Drawings -- We've done five DealBITS drawings so far, and overall, I'm quite pleased. After the initial setup work, we've managed to streamline the process of setting up a new drawing to minimize the effort on our end, and they seem popular with readers, thanks to the chance to win a free product and the discount for entrants who don't win. In terms of revenue, DealBITS has performed acceptably, generating roughly the same amount in a week as we'd receive from a sponsor for that week.
Two other positive aspects of DealBITS surprised us. First, the number of entrants for each drawing has been fascinating to watch, since it's essentially a way of determining how attractive that product is to TidBITS readers. It's not a statistically significant number, of course, since some number of people would enter a drawing to win floor lint, other people wouldn't enter a drawing if the odds of winning something good were 1 in 2, and unrelated events (like the holidays) undoubtedly affect whether or not some readers will find the time to enter.
Second, because a particular DealBITS drawing runs for only a single week, it ends up being significantly less expensive overall than a longer-term sponsorship, which makes it more accessible to smaller companies. We've long been troubled by our inability to figure out ways to help small companies promote their products while not diluting the value of sponsorships, and DealBITS seems to be a good step in that direction.
Spotlighting Peachpit Books -- Another thing we tried was putting tiny nano-reviews of particularly interesting Peachpit books in our sponsorship section; revenue was generated when readers purchased those books through our affiliate program. From an income standpoint, this approach was fairly dismal, generating in a month less than a tenth of what we'd earn from a sponsor in that time. As we brought more sponsors online toward the end of 2003, it became clear that the nano-reviews weren't earning their keep in that space. The effort of selecting and writing up a book within the space constraints of our sponsorship area was more onerous than I'd expected, and the InformIT affiliate program run by Peachpit's parent company lacked both ease-of-use and any kind of real-time reporting, making it a rather frustrating experience.
Despite its poor performance, I'm not writing this idea off entirely, because I think there are ways an affiliate-based revenue stream can prove useful both for readers and for our bottom line. If and when we try again, we'll rejigger the approach.
Google AdSense -- We had high hopes for Google's AdSense program, which publishers like us can use to place targeted AdWords ads on their sites. AdSense uses a click-through model, so the publisher earns a few cents per click. Thus, earning any kind of real money via AdSense requires a vast number of impressions from which to generate a non-trivial number of click-throughs.
Our AdSense numbers started out promising, as visitors to our Web site were initially intrigued by the novelty and clicked through at a good clip. The click-through rates tracked our publication schedule, peaking on Tuesday with a new issue and trailing off throughout the week. Unfortunately, the numbers, even on Tuesdays, started to drop after a few weeks, and by the time we removed the AdSense ads from our pages at the end of 2003, we were down to earning, in Seattle coffee terms, roughly a latte per day. We replaced the AdSense ads with ads for our Take Control ebooks, since selling even a single ebook would buy us that latte.
In retrospect, we shouldn't have been surprised at the lousy performance for Google AdSense, not through any particular failing of the AdSense program, but because our audience, though large, is relatively static. Sites that do well with AdSense, I suspect, have a very large and changing audience, so the AdSense ads remain interesting and fresh. For our loyal readers visiting the site each week, the AdSense skyscraper ads faded into the page background. As much as Google does a fairly good job matching the ads to the content of the page that contains them, we also found that we seemed to end up mainly with generic ads for Macs, which we recommend buying through our long-time sponsor Small Dog Electronics, or with ads for non-Mac products.
Take Control -- Although revenue from our Take Control series of electronic books doesn't directly support TidBITS, it's all intertwingled, and in that regard, Take Control was our home run for 2003. Despite a bunch of wasted effort on an abortive attempt to use our own merchant account (demonstrating why most people should stick with online stores like Kagi or eSellerate), Take Control sales have been amazing. Joe Kissell's "Take Control of Upgrading to Panther" has sold more than 5,000 copies, Matt Neuburg's "Take Control of Customizing Panther" has sold more than 3,600 copies, Kirk McElhearn's "Take Control of Users & Accounts in Panther" has just hit 1,500 copies, and Glenn Fleishman's "Take Control of File Sharing in Panther" is off to a promising start with over 600 copies sold in the first few days (see the excerpt elsewhere in this issue).
Although the Take Control ebooks ended up being more work than we anticipated (mostly due to the effort of living up to our own standards), they're both financially worthwhile and tremendously rewarding, since they're meeting our goals of being highly practical, tightly focused on specific topics, and extremely affordable. It's great fun creating something that people really like and that actually helps put food on the table for both us and worthy authors.
PayBITS -- Lastly, although it's not a new revenue source, I wanted to report back in on our PayBITS experiment, which helps readers acknowledge the value of information in an article via direct payments to the author. So far, it's defied coherent analysis because there are too many variables that arise when someone is deciding to pay for an article. For instance, some people see it as a way of supporting TidBITS in general, with a particular article as the impetus for making the payment. Others may pay for one article, but then not for another equally useful article because they feel that they've spent enough on TidBITS or on that author (particularly me, since I write so many of the articles) for a while.
Some of our outside authors have done well, with David Shayer receiving over $400 for his ultra-detailed comparison of disk repair utilities (see "Shootout at the Disk Repair Corral" in TidBITS-707). Some articles have approached $100 in PayBITS payments, but others have netted almost nothing, and we haven't been able to identify patterns.
As a result, I've been working entirely on instinct with regard to when to include PayBITS blocks underneath articles. Here's my current approach. Unless we believe that a staff-written article is likely to be particularly useful, we're going to publish it without a PayBITS block. The problem is that PayBITS can succeed only if people don't feel as though they're being pinged incessantly by it. There's nothing special about the PayBITS block text, though, so if you want to reward an author for an article that lacks an explicit PayBITS block, you can likely use that person's email address from the byline with PayPal.
My hope is that by limiting our use of PayBITS blocks to just those articles written by outside authors (and only those who want or are allowed to accept payments) and staff-written articles that are especially worthy, we can focus enough more attention on PayBITS that the amounts per article will increase. We've never promised that PayBITS would provide anything more than money you wouldn't pass by on the sidewalk, but it's always nicer if an author can end up with $40 instead of $3 - it's that difference between a couple of music CDs and a latte.
Eventually, when we have everything moved over to Web Crossing (see "The Web Crossing Begins" in TidBITS-711), we hope to integrate PayBITS better, and perhaps even start supporting one of the new micropayment companies, like BitPass.
The Bottom Line -- Despite the varied levels of success of these different efforts, it's clear to us that we must continue to concentrate on our corporate sponsorship program, since that's where we can most effectively generate the funds necessary to keep TidBITS functioning, while at the same time helping worthy companies introduce their products and services to readers in ways that everyone finds helpful.