17 October 2000
PAGE 1 | PAGE 2

Referendum Survey Results, Continued...

Why We Asked

Well, partly morbid curiosity; would our favorite sections be the ones no one read? Fortunately that didn't seem to be the case. (And not that that would stop us.)

Anyway, asking how often visitors read various sections of the website is in a sense redundant. Our web server statistics tell us precisely how many accesses there are to each page, every day. [Added note 18 Oct: For instance, there were 188 accesses to the first page of this editorial on Tuesday 10/17; only 103 to this second page; fewer than 21 to the third sidebar page.]

Based on those statistics, the site is doing well. Visits to Locus Online have increased by over 100% from the first quarter of 2000 to the third quarter -- from roughly 11,000 per week at the end of March, to 27,000 per week at the end of September. (A 'visit' is any number of accesses from the same computer until 15 minutes elapses without another access.) The addition in June of the "Locus Index to Science Fiction" to the locusmag.com domain brought a substantial increase in visits, as much as 20% of accesses some weeks, but even accounting for that, visits to the site have doubled in six months.

Unique visitors is more difficult to get a handle on. Our server used to accumulate statistics by week. When it did, the ratio of total visits to unique visitors was 2 to 1 or more -- 19,000 visits, 8,000 different visitors, the last week in July, for example. Now our server only posts daily totals, and the ratio is more like 1.4. There's no reason to suspect the weekly ratio has changed, which would suggest unique visitors lately in the 13,000+ range.

But these statistics are slippery and exhibit befuddling blips. The week just finished, Saturday through Sunday, 10/8 through 10/14, the number of visits was over 41,000 -- an astonishing 50% jump from the week before (27,700). What could possibly be the explanation? True, there was a bunch of new material posted on Monday the 9th -- but visitors don't know if there's new material or not, until they visit the site. The number of unique visitors was also up 50% or more each day last week compared to the corresponding day the week before. One possibility is, that when a lot of new material is posted, Locus Online fans catch up both from work and at home -- counting as two visitors each, or more if they let that 15-minute gap pass between clicks. Then again, we're aware that a substantial portion of hits to all websites (we read this somewhere -- something like 20%? 30%? -- a lot) come from robots, those automated programs that compile content for search engines. So we shouldn't read too much into these gross figures.

Getting back to the matter at hand... this increase in traffic to the website, however, has not translated into increased revenue. Rather the reverse, in fact. Locus Online is largely independent of Locus Magazine -- that is, your subscriptions to Locus Magazine support, roughly speaking, only the portions of this site explicitly related to the magazine -- the subscription form, the Locus bestseller lists, interview excerpts, etc. (Just because you subscribe to the magazine doesn't mean you're entitled to special sections of the site, as one survey responder implied. It means you get the magazine.) The rest of the website is produced by volunteer labor and compensated for only by the online advertising and Amazon revenue, as we explained in an editorial earlier this year. Yet, though visits to the site have doubled in six months, commissions from Amazon have dropped 34% over the same period. We don't know why. [Sidebar: the ranked clicks and sales from Amazon.com for the 3rd quarter 2000.]

The point in the context of this survey is that, however attractive many of the ideas of added sections and additional content of the site may be, their development is limited by these practical constraints. We're not in a position to be subsidized by, say, a software corporation or a cable TV channel; we're not inclined to change the content of the site to attract media fans and thereby massively increase advertising. Therefore the state of this site, the amount of material that appears from week to week, will remain about as it has been for the indefinite future...assuming we don't expire from exhaustion entirely. (One reason for the survey was to see if there were sections of the site that could be eliminated.) By the same token, projects like converting the archive of Locus reviews to HTML, or building the "Locus Index to Science Fiction Awards", will come only gradually, largely through the volunteer efforts of Bill Contento and Mark Kelly, with only the promise of eventual CD ROM sales as compensation.

That said, we hope to not sound too crass by reminding you that you can support this site -- specifically, to help finance the addition of guest reviewers -- by using links to Amazon.com from Locus Online to buy books or other goods. As explained in that earlier editorial, Locus Online gets a commission whenever you buy a book from a link here on Locus Online (such as those on our cumulative 2000 books listings pages). In fact, we get a (smaller) commission on anything you buy from Amazon (electronics! garden equipment!) if you arrive at Amazon through a link from Locus Online, even if you then browse around Amazon and buy an item that wasn't directly linked by us. Your cost is not affected; you pay the same price Amazon charges regardless of whether you use Locus Online's links or not.

Yes, we realize that you may fine reasons not to buy from Amazon. Specialty bookstores need and deserve our support. That's why we list their events. You may well have allegiances to other online booksellers. Why, some of Locus Online's best friends work for Borders or Barnes & Noble. But where those considerations don't apply -- buying non-genre books, or gardening equipment -- using Locus Online's link to Amazon.com will support this site at no extra cost to yourself. (Just bookmark this link to Amazon and use it from now on.)

In Conclusion

Thanks to those who participated in our survey. We appreciate your feedback. We will continue to continue...