December 27th, 2007

What Is The ROI Of Requiring User Registration To Access Online Content?

by

I somehow got logged out of my NYTimes.com registration and just hit the registration wall when I tried to read an article — I almost forgot it was there. Which made me wonder, now that the TimesSelect pay wall is gone, what the real ROI of this registration wall is for the New York Times and others sites, many of them newspapers, that require registration.

new-york-times-registration.jpg

The theory goes that personal data collected from registered users enables sites to better target ads and charge premium rates. But I wonder whether the lost traffic from users who choose not to jump through the registration hoop — which I bet is particularly true of NYTimes’ large volume of visitors from search engines — outweighs the gain of higher ads rates (assuming NYTimes’com is consistently able to charge higher rates).

When NYTimes.com ran the numbers on TimesSelect, they decided that the lost traffic, particularly from search, and the associated lost ad revenue outweighed the revenue from paid subscriptions. I wonder whether they’ve run the same analysis for the registration wall.

Of course, the NYTimes.com registration wall is not costing them search traffic, since they have enabled search engines to crawl the same content that users can see after they register (you can find some information on this practice here.)

But the registration wall likely prevents some non-trivial number users from accessing the site. Many frequent NYTimes.com readers, like me, have passwords saved, but NYTimes.com is clearly focused on growing their readership beyond the current regular readers — and that’s where they likely run into a lower tolerance for jumping through registration hoops.

NYTimes.com claims that registration only takes “a minute,” but it takes more than a minute to decide whether to share all of this required personal information.

new-york-times-registration-form.jpg

You have to wonder in the age of behavioral targeting networks (I believe NYTimes.com participates in Tacoda’s BT network), which derive user data based on user actions, e.g. what types of content they view, whether creating a content access barrier to collect data about users is really necessary.

Or perhaps it’s more timely to wonder whether in the age of Facebook Beacon it’s necessary to collect user registration data, since Facebook aims to be a marketplace for buying and selling user data. Interestingly, New York Times was one of the first to sign up for Facebook Beacon, although their press release characterize Beacon specifically as an “opt-in” program, and we all know how that turned out (which may have left NYT feeling like they were sold a bill of goods):

Beacon is an opt-in feature that enables users to decide if they want to share their activities from other sites back to Facebook for distribution to their friends. NYTimes.com and some of the Times Company’s Regional Media Group properties (beginning with Gainesville.com, Gainesvillesun.com and Gatorsports.com) will participate in the Beacon program. Boston.com will also participate in the coming months. Beacon gives Facebook members, for example, the option to share their ratings of travel venues featured on NYTimes.com with their Facebook friends.

You could say that Facebook and other social networking sites represent the next generation of requiring user registration — they are basing their entire business model on monetizing user data, since selling ads next to user profiles has generated little profit or sense. Facebook Beacon would have allowed Facebook to do all of the user data collection “dirty work” — perhaps it still might. Certainly the $15 billion valuation depends on it.

That said, there’s clearly a tension right now in the online media business between removing friction from user access and content discovery, and maximizing content monetization. The trend seems to be towards removing friction with more unfettered access, but the monetization models have yet to catch up.

Google, of course, still has the perfect model, whereby users willingly hand over their “data” in the form of search keywords, which power Google’s AdWords money making machine — no registration required.

It may be that NYTimes.com and other publishers that require registration have determined that the registration data is still paying off — it would just be interesting to see the numbers.

Update

Howard Owens supplies an answer to the question — I particularly agree with his statement that “the future for newspapers is in tying registration to participation”

Comments (23 Responses so far)

  1. Dont forget revenues from direct marketing (emails etc…). Pure margin..
    This is what we called free access. In fact you pay with your personal datas, nothing new here,

  2. Faulty assumption that registration costs traffic.

    I’ve run two registration sites, and have spoken with other newspaper.com site managers who have run their own registration-required sites, and two things I found to be true based on empirical evidence:

    1) There is no drop off in traffic past the first 60 days of registration (after 60 days, traffic exceeds pre-registration numbers and continues to grow).

    2) Most local people — the only thing a local news site should care about — are quite honest in the information they share with local newspaper.com (like 96 percent match rate against independent data).

    Both of these verifiable facts fly in the face of the two main arguments netheads use against registration schemes.

    I’ve heard from enough online executives that have both a robust registration scheme and aggressive opt-in e-mail marketing program that revenue directly tied to registration is quite healthy.

    That said, I think the future for newspapers is in tying registration to participation.

  3. The worst part is, because you so rarely need to login, a lot of the time, you have NO IDEA wtf your password is. Since they never cover anything that’s not in a million other places, I usually just go find the same story elsewhere if I get logged out.

  4. Scott, you’d have to confirm/check, but I believe in various scenarios, you can view one article before registering, so the “wall” doesn’t smack you in the face right away every time.

  5. Emmanuel, ah, yes, the old fashioned (pre-Beacon) way of selling people’s personal data.

    John, you may be right, but I believe the wall hit me as soon as I clicked on the first article without being logged in. One “free” article would soften it, but it’s still going to kick in for anyone you hope will browse the site with eye towards coming back

  6. Registration at this point seems to be a pointless and annoying blocker. Why not display ads based on a user’s implicit actions? User enters the travel section to read a story about the south of France, display an ad for airfare to Nice.

    There are so many chances for users to opt in to all of the NYTimes e-newsletters, they’ve got plenty of opportunity to collect user data without requiring a login.

  7. Am I the only person who doesn’t mind logging in to websites? It really doesn’t bother me most of the time. And in theory, any site that knows who I am should better be able to personalize my experience, which more than makes up for the registration time.

  8. [...] like to offer my two-cents on an article featured on Techmeme but displayed on Publishing 2.0 in regards to what’s the point of having user registration on a site? Does it ultimately lead [...]

  9. It amazes me how NYT wants to look and feel all bloggy but they’re still hanging on to web 1.0 conventions like registration walls. They ought to learn from owners of community blogs –you require registration for commenting and posting content.

    There must be some VP hanging on to dear life to that registration ruse (as if I ever answer those in truth or even with my real name!) You can just tell it is someone’s pet project and they’ll justify it to the end.

  10. i cannot count the number of times i have closed the window and moved on whenever a registration form shows up… example, i have never read the LA Times (a good paper) online, and never will…

    life is too short, junkmail is too much, you want me to register, i feel used from the gate, so, fu and i go elsewhere

    bet i am not alone

  11. Hi,

    — Begin disclaimer
    I work for the Times, but only tangentially with the website. I’m speaking for myself, and not for the company. I’m not divulging anything you couldn’t find out yourself.
    — End disclaimer

    OK. To answer one question: There is a built-in “soft” limit before you actually are prompted to register or log in. I believe it’s five stories, or something on that order. If you find Times content on a search engine (or any other way), you should never be prompted to log in right away. I don’t know why you were in this case.

    I think Howard’s points are about right, but I also think there are huge problems with the registration wall as well, just from a user experience point of view. I also wonder just how accurate the information gleaned about users through registration actually is. I have a…um… let’s say… friend who never, never gives accurate information about himself when asked to register. I don’t think he’s the only one, either.

    So, I wonder if the actually information is valuable, or the appearance of having actual information is good enough? I honestly don’t know the answer to that.

    The real solution won’t come until there is some kind of open, ubiquitous way to ID users without registration/login. I think if there were some way to carry your credentials around with you site-to-site, users would probably be OK with giving up some very basic demographic information about themselves (age, zip) in return for never having to login/register again. I know I would.

    But open ID, or whatever that standard ends up being is a ways away. So, in the meantime, I think we’re all stuck with the registration wall for now.

  12. Wow, this is a terrific piece! This is something I’ve been thinking about for a while. I think the ROI of requiring registration from casual users who visit a site, is highly questionable. As search engine users become increasingly important to content publishers, it’s a question of time before the registration and payment walls start crumbling.

    I’ve expanded on this theme on my blog:
    How long before the walls around content come crashing down?

    The numbers observed by Howard Owens are both interesting and troubling. Intuitively, it makes sense that at least some users will get discouraged and drop off when confronted with a “registration required” notice, so there’s bound to be some negative impact. If there isn’t, I suspect that it might be due to some other things that were happening at the same time (say, SEO enhancements). However, in a comment on my post, he says that there were no other changes beyond requiring registration. Hmm – I guess more thought is required. :-)

  13. Good piece, good comments.

    Nitink’s and Howard’s comments and expansion point up the difference between local web sites that want to drive traffic to local advertisers and global ones.

    However, if I were a big local site, I’d be looking to become global, in the way that the New Yorker and New York magazine ( hardly equivalent but …) and indeed the NYT are widely read outside the city.

    I live in Santa Barbara and our used-to-be pretty good local paper is imploding under weird private ownership, leading to two new paper/websites and a greatly expanded weekly. Its a laboratory for the transition from paper to bits, I think. Or maybe just our little soap opera. If my point is valid, then these sites should be looking to provide content that is interesting beyond our cozy little enclave. Maybe thats naive.

  14. [...] What Is The ROI Of Requiring User Registration To Access Online Content? Scott Karp wonders if there really is any value to the annoying requirement to register to read at otherwise free newspaper sites. Mark Pitts has some thoughts on the issue, too. [...]

  15. Jawfish sort of touched on something I want to clarify.

    I believe registration behavior is different for people visiting their local newspaper site than for any other site.

    Based on my own observations, I believe that people have a high trust level with their own local newspaper.

    In other words, the person who might gladly register for his hometown Ventura County paper won’t register, or give false information, to a Chicago paper. They are also more likely to opt-in for e-mail provisioned by the paper from local advertisers.

    And again, I’m not advocating for registration-to-read schemes. I think there are better and smarter options newspapers should pursue. My only point to weigh in with some points I believe to be factually true. It just bugs me to see people arguing against registration with assumptions I know not to be true. Be against registration, ok, but not because it hurts local newspaper.com traffic or because most people lie or use sites like BugMeNot. I just know those assumptions are not true, not for the local newspaper.com.

  16. good questions raised here, Scott. I’m also interested in how the ROI plays out when publishers participate in networked behavioral targeting ad programs, as you mentoned. seems to me there should be an incentive for a publisher to contribute data to a network that can be used elsewhere in addition to benefitting from what the network pays to the publisher for inventory. I suspect reg walls may then be justified, but there are many other more integrated and relevant ways to build data-rich relationships with users than a reg wall. reg walls are lazy.

  17. [...] answer is inherently unknowable, of course, but my friend Scott Karp of Publishing 2.0 had a great post recently about the ROI (return on investment) of registration systems — something he only thought of [...]

  18. [...] answer is inherently unknowable, of course, but my friend Scott Karp of Publishing 2.0 had a great post recently about the ROI (return on investment) of registration systems — something he only thought of [...]

  19. I had been toying with the ideas of bringing registration onto our site for a while.

    I have always beleived that we needed to keep the website open to anyone who wants to read a story. The challenge has been though we have lots of casual readers – we have a much smaller number who subscribe to the weekly email newsletter.

    Having read this article has confirmed my view that registration does create a wall and I will hold off and only use registration if people want to participate – e.g for putting in comments, customise the delivery of information or a bookmarking service.

  20. I wonder how accurate the data is? In my case, I always use a real email address but everything else could be whatever gets me through the form fastest, such as the first options in the drop-down menus.

    Just yesterday, I was an 18 year old graphic designer who was registering his first iPod with iTunes. :-)

  21. There’s always BugMeNot, which I use whenever possible. There’s a Firefox extension that automatically pulls up BugMeNot passwords. Easier than registering every time and no worry about what information to disclose.

  22. [...] even though the title is a bit over the top, there seems to be a new push to look at the real benefits of forcing users to register before they are able to see any content on newspaper sites. Mathew [...]

  23. [...] What Is The ROI Of Requiring User Registration To Access Online Content? – Publishing 2.0 The theory goes that personal data collected from registered users enables sites to better target ads and charge premium rates. But I wonder whether the lost traffic from users who choose not to jump through the registration hoop — which I bet is particularly true of NYTimes’ large volume of visitors from search engines — outweighs the gain of higher ads rates (assuming NYTimes’com is consistently able to charge higher rates). [...]

Add Your Comment

Subscribe

Receive new posts by email