Notes from a news-site paywall attempt

New Zealand's National Business Review stuck a paywall in front of its website back in mid-July, betting that enough readers would stick around and pay for "quality" that it would make up for the stupendous drop in readership. Looks like they bet wrong. Their traffic plummeted, and traffic to their free competitors skyrocketed. On every metric, NBR is failing: pageviews, session duration, unique readers, and total time on site. NBR has a high paywall price, so maybe they've got enough money from corporate subscribers to make up for the advertising losses -- but how long will they keep them for, with all the links, visits, and attention going to their competitors?

NBR's performance since the subscription wall was built (via O'Reilly Radar)


  1. Sorry, but I do not see “Their traffic plummeted, and traffic to their free competitors skyrocketed”. I see a steady decrease, and no spectacular change in the tendancy from the “week of the wall” date.

    If I were the manager, I would be quite satisfied: I had 100% of non-paying readers the 6st of April, I have 80% of those now being paying readers, which is probably way beyond the initial goal. What manager would expect the loss of such a radical change being inferior to 20/25 %?

    The question is really is, for a manager: we see a downward tendancy; what is the “paywall part” of it, and how can we inverse the “non-paywall part” of it. This might as well be redactional insufficiency.

  2. Hi Cory, thanks for the post, it’s a really interesting debate. Only one thing is missing – and that’s the *real* key metric: the cash. They must have been in dire straits to consider this approach and the advertising revenues you mention were, quite possibly, dropping through the floor. I don’t know the publication well enough to know if it’s got content that’s valuable enough – but maybe they believe they’ve got good enough penetration of the audience sector and that it will grow over time. It’ll be interesting to see where they’re at in a year. Richard

  3. Hey, I’ll happily admit that it’s not easy to monetize fame. But it’s a lot harder to monetize obscurity.

  4. Sorry, I just don’t see from the graph how are they failing.

    A drop in visits is of course to be expected with introduction of paywall, so it is hardly surprising. Do they get more or less cashflow now? Those figures don’t tell, but the clearly biased presentation sure suggests you have an axe to grind with the “old press”.

  5. Unique browsers looks to be at over 90% of their reference date. Their worst metric is average session time, and it is only a little less than %80. Of course they are going to loose some readers when they block their site with a pay wall. Many even. Given that you say theirs is a high priced pay wall, I’d call this a raging success.

    Too call this a stupendous drop is pure hyperbole Cory. The sort that I expect more from the MAFIAA and friends. What in your mind would have been the expected drop in order to consider this a success?

  6. Hmmm, it’s really impossible to say from this whether or not the pay wall was successful.

    The blurb says that NBR was “betting that enough readers would stick around and pay for “quality” that it would make up for the stupendous drop in readership.” It then shows that the readership dropped (this would have been obvious before the experiment), but doesn’t show whether “enough readers stuck around to pay for “quality” that it made up for the stupendous drop in readership.”

    Frankly, at first glance, that drop doesn’t seem that huge. If NBR has converted all those views into paying customers, it looks like they could be doing fairly well. But it’s impossible to tell from this graph without the other half of the data.

  7. Cory, I clicked through and read the link. It does seem that their traffic has dropped, but no “plummeted,”, and there isn’t support for the claim that the competitors sites soared. The article says increases were likely due to site redesigns, and points out that the NBR site had very little traffic compared to the two big free NZ sites prior and after the wall. Your point may be valid, but sensationalizing and exaggerating the story does not add to your credibility.

  8. I don’t know where you got your statistics training, but this tiny set of data certainly doesn’t indicate a failure at all. Actually it looks pretty good from a business perspective. I’d rather have 70% of my original customer base paying me something than 100% paying me nothing. Hurrah for paywalls! You should try to be a little more objective while on your soapbox.

  9. I’m sorry but I have to agree with the other comments.
    Looking at the data I’d be pretty happy if this was my site – the drop is not significant.

  10. I don’t find the data particularly convincing of the thesis. First, there seems to be a developing downward trend before the pay wall. Second, if pay was the issue, then I’d expect a strong step function, with everything dropping off immediately after the pay wall was instituted. But things continued to slope down. Third, you’re not looking at this from a business view. You’re assuming that the additional traffic somehow more than makes up for the revenue gained by the pay wall. This is a big fallacy on the part of many who criticize the idea of charging for content – I’ve written about it at BNET: Most media companies may simply have little to lose in the trade, though it may be that neither free nor paid scenarios will generate enough revenue to cover the costs of doing business.

  11. If I retained ~80% of my websites’ readership after implementing a paywall, I’d consider it a stunning success. The downward trend in the stats is troublesome however and cause for concern.

  12. Cory, I don’t think paywalls are a bad strategy, I just think they need to get the pricing structure correct. I hate it when people demand goods and services for free, as if they have a right to it. Why shouldn’t NBR charge for a service they produce? Agree?

  13. I always find it interesting when people try to second guess a business plan that they have no interest in ever contributing to the bottom line, but feel cheated in some way, or have to comment negatively about it because they no longer get their free service.

    I run a few sites…run isn’t the right word as I’ve been a little hands off over the last year…but the one I’m most active with went entirely behind a paywall. Invite only, and if one of the people you invited acts up, your reputation may be affected negatively. Site dropped from 50k registered users to less than 1k. Some have looked at that metric and said it was a failure. I look at it as these users are now paying, I moved from a several hundred $$$ a month dedicated server to a cloud based system that expands as I need, no spammers, no advertising, and my users can step out behind their usernames and actually say who they are (the industry I work with really doesn’t need idiot fanboys signing up ruining every conversation just because they figure out someones nom de plume).

    Personally, I am happy to pay for the content I want to read…I like the fact some sites are free, and some would be worth a subscription if they weren’t…and some wouldn’t. BB would be one of the sites I’d probably pay for…

    1. Clif, I would argue that your site works best when it is exclusive. That means that a paywall is probably a good policy. I’m not arguing that you’re wrong, only that you’re applying your valid experience to a different issue, which is a news site.

      I have to agree with the other readers that the overall effect of the paywall seems to be positive (money instead of zero). However, without access to the raw data to compute correlation, I’d have to guess that before the paywall readership was steady, while after the paywall readership is declining. That would seem to indicate that the money made by the paywall is declining.

      What would be far more interesting is seeing what happens if all the major news outlets erect paywalls. Then we’ll see if people can live without corporate news or not.

  14. Dude, Cory– we know you’re passionate about this sort of thing, but let the stats speak for themselves. The only thing “stupendous” here is that they’ve retained so many readers post-paywall. This sort of customer retention is what businesses usually dream of!

  15. Did we read the same post on NBR?

    While there is a visible decline it is a far cry from a “stupendous drop in readership”. Their traffic did NOT “plummeted” and traffic to their free competitors has NOT “skyrocketed”.

    You even say:
    On every metric, NBR is failing: pageviews, session duration, unique readers, and total time on site.

    Yes, those are all down, but thats not a fail.

    In some cases it could be, like if the downward spikes continues a lot further. But its not a fail (yet).

    I’m not a advocate of pay-walls, but lets but our bias aside and be fair and honest about this. The original post does not draw this conclusion either (though suggests its too early to tell).

    With any pay wall there will be a drop in traffic, pageviews, uniques, etc. The question is whether the new revenue stream a pay wall provides makes up for this or not.

  16. I’m not so certain I agree with Cory. Academic Journals have been involved in this sort of practice for years.

  17. I agree with those above – by no means does that graph imply that traffic has ‘plummeted’. Personally, I think that putting everything behind a paywall is a bad idea (your site can’t be spidered and consequently accessed through google) but it’s understandable if some companies consider other strategies to make money, especially in the current economic climate.

    … and I’m sorry to say that Cory’s writing style has been lapsing into the hysterical recently, especially if it relates to a business model that he doesn’t agree with.

  18. I’m so glad to know from the comments that I wasn’t the only one who felt that the metrics for NBR far from plummeted with the addition of the paywall. Down yes, but far less than I would have suspected.
    I’d love to see followup stories on NBR’s traffic over time as well as a thorough case study on the impact of the paywall on their readership and bottom line.

  19. Speaking as someone who’s job is as an internet market analyst I wholeheartedly agree with everyone’s previous posts.

    If I presented this data and made such judgments I would be fired. The correct way to look at these is not a recent trend but current trend compared to last year’s trend. Site traffic is going down? Well, we’re also entering New Zealand’s summer months, traditionally a down time for the internet. I suspect you’d see similar traffic patterns in 2008 and 2007. The author doesn’t even explain how he’s calculated the index. And the reason he’s indexed is a mistake; because the other NZ sites are so large? Put them on a secondary axis and give me the actual data.

  20. Like everybody else has said, these data really prove the opposite of what you’re saying. Since I came on at VQR, I’ve pushed hard to open things up. But looking at these numbers, I’m rethinking that. We’re both a magazine and funded by a university—the financial situation is not good. Neither fame nor page views will pay my rent. If we lost 20% of our traffic and (say) doubled our revenue, that would be a huge success.

  21. Cash is king.

    Please call us back when you know what are the cash flows pre and post pay wall.

    To describe this as a failure without knowing the real numbers that matter is disingenuous to say the least.

  22. Speaking as someone who does development for a financial firm, business people WILL PAY if youi ask them to for content. They honestly don’t mind and expect to pay; this is where they get that concept that content should be paid for because they honestly can’t believe they can get it for free and think it can’t be good if it’s free. If you charge them, they think it must be good.

    The rest of the world is different however and realize you can get things nearly just as good for free (or at least cheaper). So if this stat were to be applied to something not business related, you would see far lower stats.

    Take for example Murdoch’s sites. Most of them have nothing to do with business and the content is mirrored in a million other news locations and blogs. If they were to put up a pay wall (or delist their content from Google), people would just go elsewhere for the content.

    Early on in web development at Amazon, we realized that the AVERAGE consumer will spend 30 seconds searching and then give up. Nowadays that is lower. Most people do not go beyond the top three pages in a search and if you do not show up on those top three pages, you can guarantee that you will be missing 80% of that search traffic.

  23. I thought I needed a new set of bifocals when I saw the word ‘plummeted’ in relation to the article referenced.

    As others have said already this looks like a successful launch of a pay site – the question is ho long will it last. We’ve seen huge increases in the amount of traffic coming to our site through search and search ONLY. Put it behind a pay wall and who will come?

  24. The only number that really counts isn’t on that graph at all: Income/Profit.

    If I charged $1 a month for my site and in doing so lost 99% of my traffic, I would still make more money than just by ads. Admittedly I only place a single small ad area as a tip jar. If I tried to exploit ads I could still turn away 90% of my current traffic and be in the green. Well… light green anyway.

  25. It may just be me, but this post is getting hard to find! I’ve been checking back throughout the day to see if there were any viewpoints that weren’t along the lines of “that’s not how you define plummet” or “what is the cash flow trend” but so far nothing — not much anyway. And it doesn’t seem to be in the list of posts anymore. I’m commenting primarily to bookmark it.

  26. I think there’s some interesting points the article makes, but as some have suggested, at least initially, only losing about 20% of your unique visitors for paid ones isn’t that bad.

    However, it also makes the point that total time spent on the site dropped by 38%, and the trend may indicate that new users are going to be less likely to pay for what they can get elsewhere, so while this initial drop isn’t all that bad, if the trend continues, they may have very few readers in the future.

  27. in what is becoming an increasingly common BoingBoing phenomenon, I post something politely critical of Cory, and my comment magically disappears. No personal attack, just honest criticism. *Poof*.

    I’d like to know which part of my comment refuting Cory’s conclusions was in violation of the moderation policy.

    1. Zieroh Tardy,

      Your idea of politely critical doesn’t meet mine. And you complain a lot. I suggest that you read the Mod Policy more thoroughly.

      If the bulk of your comments are whining, hostile, insulting or otherwise annoying, you may be banned.

  28. I agree with the other comments. I have to add that the very slight drop for a business publication in a shift to paid content is not really a surprise.

    When all this is over, there will still be only one main way to maintain newtorks of quality, trained journalists: paid subscriptions. It’s no surprise that organisations like Reuters, Bloomberg and Down Jones can keep the world’s last big reporter networks (2,000 journalists or more each) afloat. These organisations work almost exclusively behind firewalls, providing news to clients who pay for accurate, first hand information.

    What will happen with entertainment and lifestyle news? I’m not sure. It doesn’t look good, though. Paid journalists and first-hand reporting are absolutely crucial to transparency and solid information, and without a subscription model, it’s hard to see how that will be possible.

Comments are closed.