The E-Print Experiment
A short while ago, I blogged about an experiment I was planning — to switch from mailing out reprints of my recently-published papers to sending out emails offering a variety of ways to obtain the reprint (electronic copy via email attachment, SSRN download, or hard copy reprint via snail mail). I decided to go ahead and try the experiment, and many folks asked me to report back the results.
I sent out the following email:
I’m writing to distribute a final reprint of my recently-published essay, Fourth Amendment Codification and Professor Kerr’s Misguided Call for Judicial Deference, 74 Fordham L. Rev. 747 (2005). The essay critiques Professor Orin Kerr’s argument that legislatures are superior to courts in creating the rules governing law enforcement and new technologies. Kerr recommends a deferential judicial approach to the Fourth Amendment when new technologies are involved. I contend that Kerr is wrong.
I’m trying a new experiment with reprints. I used to send out a mass mailing of reprints, a system that was expensive, not environment-friendly, and not very efficient, as I bet that only a percentage of recipients were interested in reading the reprint. Therefore, I’m trying out a new system, one that is designed so that copies of the reprint go only to those interested in reading it.
If you’re interested in reading a copy of my essay, just reply to this email and let me know which version you want:
1. Electronic Copy: I have an electronic PDF copy of the final published version, which prints out looking identical to how the piece looks in its final published format. I’d be happy to send you this version as an email attachment. Or you can download a copy for free at this link: http://ssrn.com/abstract=786266
2. Snail Mail Reprint: I have ordered a bunch of reprints for those who prefer them, and I’d be happy to send you one if you want.
If you prefer not to receive emails from me about my reprints in the future, just send me an email asking to be removed from the list. I promise that my emails will not be very frequent, as I can only publish so many papers in any one year!
Since my reprint list has been growing over the years — and since I could distribute my paper without incurring great expenses for my law school — I sent the email to approximately 350 people.
I received 60 replies — 19 wanted the PDF attachment, 15 said they would download from SSRN, and 20 wanted the snail mail reprint. Another 4 said that they had already read the piece because they received a copy of the Fordham Law Review issue (it was a symposium issue), and 2 politely declined (they focused on consumer privacy issues not law enforcement and privacy issues).
One of the replies was from an individual not even on my list — he was interested because a friend forwarded him my email.
I received replies from several folks who really liked the idea of the experiment. Nobody told me that they didn’t like it. Nobody has opted out of my list either. If I were a marketer, I would surely assume that everybody just loves being on the list. Of course, I’ve critiqued opt out approaches before as not really reflecting such a sentiment, but I want to feel good about myself, so I’ll just believe it anyway.
These stats indicate a relatively even distribution between those who prefer PDF attachment, SSRN, and snail mail reprint. I checked my SSRN download counts for the article, and since sending out my email, it logged an increase of 31 downloads. I am assuming that the vast majority of those downloads were inspired by the email, and that there were a few people who downloaded it from SSRN but didn’t inform me via email.
Some thoughts on the experiment:
1. Counting SSRN downloads (31), PDF attachment (19), snail mail (20), and those who already read the piece (4), this yields 74 people who were interested in the piece. That’s a little over 20%. For direct marketing standards, I’m doing quite well, since direct marketers have a yield often lower than 2%. Maybe I’m in the wrong business! On the glass is half empty side, 80% of my list didn’t want to read the article. I’m curious about how many more would have read the article had I just sent them the reprint. Perhaps some folks didn’t want to bother to request a copy or felt it would be an imposition to me to ask for a snail mail reprint. Did this new method of distribution decrease readership? Turning back to the glass is half full side, perhaps more people might read the piece since some might prefer the electronic version and might pay less attention to the article had I sent it to them in the mail. Maybe people who get the reprints in the mail just put them in a pile, which stacks up over time, and then the stuff in the pile eventually finds itself unread and in the trash. I guess I’ll never really know.
2. Perhaps the 20% response rate differs little from the percentage of people who would actually read the reprint had I just sent it to them. In this regard, the experiment showed me the truth about how many folks actually read my reprints. A 20% response rate is about what I expected — I had thought that most people don’t read reprints. I get a ton of reprints and they often go in a pile. I often don’t read them until I’m working on a project, and then I frequently forget I’ve got the reprint and I just pull the article from Westlaw. So it was my expectation that the response rate would be low, although I had dreams of a 70% or 80% response rate. Maybe the truth is better left unknown!
3. I gave a lot of thought to providing the SSRN link. At one point, I was disinclined to do it, and was planning just to offer the PDF attachment or snail mail option. But I think that the SSRN link did serve a good purpose. First, there were about 15 people who just downloaded it from SSRN and who might not have wanted to bother me by requesting it. Second, the SSRN was a nice “out” for people who didn’t want to tell me they weren’t interested in the piece. They could politely say that they downloaded it from SSRN or could always say that they were going to download it from SSRN.
4. One of the values of sending reprints is that they announce the existence of your article. Even if people don’t read it, they’re at least aware of its existence, and when they’re writing something in the area, they’ll be more inclined to think of your work. So the reprint might also serve as a kind of advertisement. To this extent, the email approach seems just as effective as the regular mail approach. And perhaps more effective, since it enables me to expand my list.
5. Another nice benefit of the email approach was the fact that I caught up with a number of friends and colleagues. It was a more social way to distribute reprints, since some of those who responded struck up a conversation.
6. Will I keep distributing reprints this way? I think so. I might do a mix between the more traditional distribution and the new email distribution, sending out reprints to a much smaller list and emailing the rest with the menu of options. Or I might just leave the traditional distribution approach in the dust. I haven’t figured it all out yet. But I have decided that there’s a lot to like about the e-print approach. I like knowing the response rate, despite the fact that the ego takes a small hit when only 1 out of every 5 on my list responds. I also like being able to have a larger list — and I like knowing I can expand my list as much as I want. In fact, if you’re reading this, are not on my list, and want to be on it, just let me know and I’ll add you.