Top Story

Young Journalists Use Facebook Ads to Reach Prospective Employers

1.	X-Ray Reading Helps Interpret Author's Intentions 12:05 AM Feb. 10, 2010
2.	USDA Drops National Animal ID Plan 12:01 AM Feb. 10, 2010
3.	Live Chat Today: What Position Should I Apply for When Company Has Multiple Openings? 12:01 AM Feb. 10, 2010
4.	AP stories are back on Google News again 7:21 PM Feb. 9, 2010
5.	Foursquare partners with New York Times for Olympics 4:39 PM Feb. 9, 2010
More Recent Articles

Most E-mailed

Article Feedback

View all E-Media Tidbits feedback

Forced Registration: The Black & the White of It
(Read the Article)
Post Feedback | Feedback Guidelines | Report Feedback Abuse

Sorting options:	most recent	\| author
Display options:	expand all	\| collapse all

Page 1 of 1

Re: How "forced" registration works
Posted by Greg Linden 7/17/2004 11:41:31 PM

Thanks, John, for the lengthy comment. This is an interesting discussion.

It sounds like you did a careful rollout of the registration requirements to maximize ease of use and minimize customer complaints. It's excellent that your advertising revenue has grown year to year and that your registered user base is large.

However, low customer complaints and growing advertising revenue do not prove that registration is better than no registration. You don't know how well the site could have done had you not required registration. To determine that, you need to test both under the same conditions. Running a test is relatively straightforward. For a random sample of visitors to your website (by IP address or a cookie identifier), don't require registration. Compare the logs for this sample vs. the normal group (of users who are required to register, including those that leave immediately at the registration page) to see if advertising revenue (probably based on advertising clickthough) is higher.

If it is higher, you know that the benefit (in additional revenue) of not requiring registration. You can extrapolate that to your entire user base and determine whether the benefit of registration exceeds the cost.

You may have already done this test but, if not, it's worth considering. It would give you exactly the information you need to make the business decision about the costs and benefits of registration.

Registration Questions
Posted by Terry Steichen 7/2/2004 4:48:10 PM

John,

Your lack of interest in blogger-generated traffic surprises me. Given that 50% of your pageviews come from from 5% of your audience, it would seem logical to entice as many viewers as possible to join that elite audience segment. Allowing external referrals to get a few "free" pageviews before you demand registration, would seem conducive to that objective.

You are also silent on the idea of augmenting ordinary news with additional metadata that can support new value-added news services. (This seems to me the only way to lift news out of the "commodity" category.)

With respect to the quality of registration data, if users can get access by withholding personal details (providing bogus data instead) why would they not do so? Do you validate the e-mail addresses (and remove registration for non-validated users)? Do you validate the ages and salaries that people provide?

Publishers argue that advertisers who adopt their targeted advertising campaigns see greater results (though few appear to be willing to share verifiable data about that). Yet, how much of such benefit actually comes from behaviorial and contextual targeting that often go on concurrently with the demographic targeting supported by registration data?

In spite of my questions above, I found your comment thoughtful and informative, and your passion about having an audience-centered focus admirable

A damn good reason for registration...
Posted by Graham Strouse 7/1/2004 6:23:45 AM

Print is dying. Let there be no mistake there. Local papers, medium-sized papers will live and the major institutions will survive for a time, but my own reading habits suggest to me that even the Grey Lady may one day pass largely into the virtual ether.

I've worked online and I can tell you this: it's a bugger trying to sell anything online. People want the free lunch. Early dot-coms were killed off by dint of the fact that back when just about the only thing people were willing to pay for online was porn. Targetted advertising was the answer but no one really knew how to do it properly.

Now they're getting a clue, and registration, accretions of metadata, and new advertising techniques are the only way to make this work.

Here endeth the lesson.

How "forced" registration works
Posted by John Granatino 6/30/2004 11:22:20 PM

--part 3 of 3--

I do acknowledge that there is another group of customers that speaks to us all the time about our policies. These are people within our own profession (loosely described as online media), who seek unfettered access to our work product, without an exchange of information about themselves in return. Bloggers, for one, do not like our policies, because the value of a blog is diminished if the blogger cannot freely present information someone else has spent money putting together. If the blogger's customer has to stop and register for a link that a blogger has provided, well, that blogger's site is perceived as less valuable. I am sorry about that but it just is not a part of my focus. Addressing the needs of these kinds of customers is an expense and resource drain for me, and meeting the needs of these customers competes with meeting the needs of customers who are my everyday visitors, willing to provide registration information.

Local online media will continue to grow and thrive into the future as long as they keep a laser-like focus on the needs of their regular customers.

All the best,
John Granatino

How "forced" registration works
Posted by John Granatino 6/30/2004 11:19:58 PM

--part 2 of 3--

Once we had settled on the model that was best received, we rolled it out slowly, section-by-section, to our various web sites, all the while keeping a close eye on customer reaction. Complaints were well below expectations (around 1 per 100 registrations). There were individual stumbling blocks, which we tried to address one-by-one. But today, six-plus million registrations later, we are able to provide our regular customers with information and services that they desire, at no monetary cost to them.

We actively seek and continue to develop new, positive incentives for registration -- unifying elements of our overall customer experience that can be enhanced by stored profiles, such as personalization, e-mail newsletters, message boards, contests, archives, online classified and e-commerce stores. But incentives alone would not yield the addressable audience we need. They are simply part of the value proposition that encourages individual registrants to provide accurate information.

Forced registration has turned out to be the most efficient, least intrusive method of achieving our business goals. The best evidence we have that this strategy is working is that our advertising sales have increased substantially in the last three years, and many of our Web sites regularly win national and regional awards for journalism and public service each year.

--more--

How "forced" registration works
Posted by John Granatino 6/30/2004 11:16:29 PM

Steve,

There's a reason why so many media companies use registration: It's to understand their customers better so the companies can grow and improve their businesses. One of the great promises of the Internet's addressability is that if I know something about you, I can make available directly to you products or services of interest. The mass media lack this capability, and as a result are experiencing an outflow of advertising dollars to targeted media (direct mail, cable, the Internet, and many others). There are many examples of how advertisers will pay more if they are able to target their sales message to customers on the basis of address, interest, likeliness to purchase, etc. A media business needs to employ some kind of registration to even swim in that pond.

There just is no business model for general-interest media companies to give away information and services to unknown visitors. Even the Googles and Yahoos of the world make their money by trading on the information you tell them about yourself (e.g., you provide an address and they provide a Local Search result; you indicate a topic of interest and they provide contextural advertising -- which, incidentally, your pop-up blocker can't screen; you write an email and GMail matches advertising to concepts in your email; etc.)

So your question is: Isn't there a better way than forced registration?

Our answer has the feel of history now, but the question keeps coming up, so it's worth repeating: Three years ago, when we at Belo Interactive started asking customers to register in return for access to the information we provide, we were hugely concerned about their reaction. Many of our Web site customers were already customers of our newspapers and TV stations, so it was very important that we get the "ask" right. We started out slowly, with several models of registration, which we tested qualitatively with focus groups.

--more--

Yes'nYes
Posted by Terry Steichen 6/30/2004 2:25:19 PM

Jay,

I gather that "yes and yes" means you have solid data demonstrating (1) that most registration data is not bogus, and (2) even at its current quality, using registration data does significantly improve ad performance.

I also gather from your 'nuff said' comment that you aren't comfortable summarizing, let alone sharing, that data with us.

Unfortunately, given other data/studies that appear to suggest the opposite conclusions, that puts us in a kind of awkward position to make good use of your own insights and findings.

Oh, well.

Yes, and yes
Posted by Jay Small 6/29/2004 8:13:25 PM

'Nuff said.

The Value of Registration Data
Posted by Terry Steichen 6/29/2004 9:28:33 AM

Jay,

I think you are getting to the nub of the issue. You say that Belo Interactive has verified that the proportion of bogus registrations is much smaller than generally reported.

Do you have any reliable data on that? I mean, it's very hard and costly to verify such personal details, so I'm wondering if you actually did a scientific verification of a decent sample of the data? (I ask because it seems somewhat illogical that people would provide - potentially valuable and intrusive - accurate personal details when they don't actually have to.)

You also mentioned that advertising which uses this data for targeting is substantially more effective. Again, is this a "gut feel" or is there actual data (and it would be very nice if it could be shared) that shows this improvement? (I ask because it seems to me that a publisher has a natural inclination to assume more improvement than may actually be there - and thus justify higher rates.)

It isn't just to restrict to local
Posted by Jay Small 6/28/2004 1:24:39 PM

The purpose of registration isn't just to turn away "grazers" who happen to be outside a Web site's primary market.

It is to understand and address that site's audience with greater precision, for both commercial and noncommercial messages.

I am committed to the idea of providing every reasonable incentive to register -- but I also know that incented registration without access-control registration will secure only a tiny fraction of the number of profiles.

And adding metadata and tuning up behavioral targeting based on traffic to those metadata-rich files is a good idea. But targeting based on that method is still more of a guess than targeting based on expressed audience profiles and interests. The results won't be as good.

Reports of widespread registration fraud are, as my friend Steve Yelvington has also reported, apparently quite exaggerated. At Belo Interactive, we do see a few bogus profiles, but not anywhere near the order of magnitude research companies seem happy to report.

More importantly, when we target advertising using our registration database, the results for the advertisers are better. MUCH BETTER. And I would dare say that probably means the customer experience of advertising on our sites -- though some would argue it's never good because consumers don't want to see ANY "interruption" ads -- is better than it would be on any site that serves ads on a run-of-site, anonymous basis. The ads simply fit the users better, so more of them respond.

Culture-of-the-Internet Utopians will rail against access-control registration until the Web is replaced by that skull-cap, retina-authenticated, holographic interface we're all waiting for. But I know registration works, from experience, so all this philosophizing from outsiders-looking-in will never persuade me otherwise.

Local Focus = Success?
Posted by Terry Steichen 6/27/2004 1:10:55 PM

Greg Linden is absolutely correct that it makes little sense to arbitrarily restrict a global medium to local relevance.

However, if the sponsoring newspaper *does* have a local focus (which most do), then a local web focus makes more sense.

For example, if the sponsoring newspaper has newsgathering capabilities concentrated in a local area, the articles it produces will be of interest mostly to people in that local area (regardless of the site's global reach).

And, the sponsoring newspaper may not have access to regional or more global advertisers. Unless/until it builds up relationships beyond local advertisers, it may not have much of a choice.

In theory, it might appear to make more sense to expand the scope of the site's content and advertisers to take better advantage of the web's global reach. However, doing that often puts the site head-to-head with other sites competing for the same audience.

Perhaps a better strategy would be to deepen the coverage of local events (where competition is often non-existent).

Clipping your own wings
Posted by Greg Linden 6/25/2004 11:47:48 PM

I have a hard time with the argument for registration. The argument rests on the claim that there are only costs and no benefits to attracting non-local readers.

Once an article is written, it seems that the cost of each additional reader is very low. While "hits cost money," each additional hit costs very little. The sunk costs of buliding your website or producing content are irrelevant to the decision of whether to show content to a global audience. Only the incremental costs matter, and page serves are cheap.

As for the benefits, why would a website that has global reach only show local ads? The key is to show advertising that is most effective for the content and the reader. For some articles, especially local news, that may be local advertising. For other articles, it may not. The critical point is that online advertising is different than the advertising in the print newspaper. You can't just show the same ads you show in print and expect that to be optimal. Unlike a print newspaper, you can show different ads for each article and even for each user, dynamically optimizing and finding which ads work best. Focusing exclusively on local advertising is missing an opportunity.

At a time when so many newspapers in the US are struggling, you shouldn't clip your own wings. The large online market is an opportunity. It is an important new source of revenue at a time when print subscription bases are shrinking.

Page 1 of 1

View all E-Media Tidbits feedback

See All Jobs

Add Your Resume

Post Your Job

Become a Member

More media jobs

Article Feedback

Latest Poynter Blogs (See All Blogs)