22

We're currently discussing to add pop-unders to one of our platforms to get users to use one of our notification features. As you can probably imagine, as a UX person I'm really against the use of pop-unders. But I need some data or at least very strong arguments (because on one of our sister platforms, they do convert really well).

Are there any studies on pop-unders? (didn't find anything relevant so far...) Any other thoughts?


Update: Found one study (daipratt.co.uk/popunders/), but it's about ad revenue (so not exactly my case).

Update II: To clarify, this is what we would do: After using the search on our classified website the pop-under opens. The content would be something like "You were looking for XY, if you are interested in getting new results as soon as they appear please enter you email address here"). So it's not third party advertising and might even be useful for the user. But to be honest I still hate the idea - I just don't like when my browser opens stuff I didn't ask for.

Update III: Great answers, thanks everyone! I just accepted the one with the most votes, but the other answers are very helpful as well.

Update IV: So it turns out that the conversion rate of the pop-under on our sister site was actually dwindling over the past years - probably because of built-in pop-up blockers. Good news for me: No pop-under on our page, yey! :)

5
  • 25
    Yes. Yes they are.
    – Ben Brocka
    Dec 13, 2011 at 14:33
  • 4
    Maybe try a feature such as one on this site. (Scroll to the bottom of the post.) It doesn't interfere with the current task, yet still provides an attention-grab. Dec 13, 2011 at 16:01
  • 1
    @jberger: I see those things quite often, would be nice to know how good they work. Could be an option for us too (IMO nearly everything is better than a pop-under...)
    – Phil
    Dec 13, 2011 at 16:14
  • @Phil: They worked on me. I think I clicked on 2 of them within 15 minutes on the site. As long as it doesn't impede current work... We've been conditioned to hate pop-u_s (even if they're helpful to us). Dec 13, 2011 at 17:31
  • 1
    That technique linked by @jberger is used on the New York Times site all the time. Far more preferable to a pop-under.
    – ghoppe
    Dec 13, 2011 at 22:10

7 Answers 7

28

Pop-unders suck. The Internet has been awash with hate for them since their inception. But you know that, you need data.

Research has shown that pop-under ads are almost indistinguishable from pop-up ads (pdf source), and are actually worse than pop-up ads in terms of intrusiveness in tests vs in-line ads.
Emphasis mine:

Pop-under ads were indistinguishable from pop-up ads on every dependent variable. It is perhaps worth mentioning that intrusiveness of pop-under ads was reported to be 33.1% higher than the intrusiveness of in-line ads (t=4.728; 264 d.f.; p=.000), while the intrusiveness of pop-up ads was only 24.1% higher than the intrusiveness of in-line ads (t=3.478; 277.96 d.f.; p<.001).

Bottom line, if you accept that pop-ups are bad, pop-unders are worse. In-line is the way to go.

1
  • 6
    It doesn't sound like your client considers them "ads" (they still are) but this at least shows they're as annoying or worse than one of the most hated parts of the Internet experience, so hopefully that's good enough evidence.
    – Ben Brocka
    Dec 13, 2011 at 15:48
12

Every time I have seen them, they have been poorly done and looked suspicious to me. I just assume that there is something sinister going on like key logging or something like that (not even sure if that's possible but I err on the side of caution). I often think that the reason they are behind the window is so that they can do whatever they need to do without hoping that I do not see them until it's too late. I could just be ignorant to the capabilities of malware etc, but I always try to take the "better safe than sorry" approach.

That being said, I came across this article discussing why they can be effective:

The effectiveness of pop-under windows most likely stems from the timing. When your visitor first enters your site, their mind is on what lead them to your site to begin with. Pop-up windows will most likely cause a higher percentage of your visitors to instantly close your window before viewing its content. That's what makes pop-under windows more effective -- your visitor is through visiting your site and their mind is clearer. They will be much more likely to take the time to look at your offer when they don't have another objective on their mind.

This is a good point. The only thing I am not sure of is if they can overcome the bad reputation and disgust that has been created by pop up advertisements. Most people seem to be conditioned into hating any sort of unwanted/extra window that they weren't expecting. But again, the above article has some good points.

If the site was a reputable one, and the pop under was directly related to the site's content, I would be much less suspicious. For example, if Amazon had a pop under that offered me a discount subscription to Amazon Prime, I would be likely to check it out.

These are major things to consider when making this decision. It seems as though no matter how hard you try and how well you do it, it will annoy the user.

6
  • Thanks, interesting thoughts and link. I'll update the question to clarify our case.
    – Phil
    Dec 13, 2011 at 13:08
  • Pop unders also created a whole new swarm of buzz and hatred when they first popped under the scene: informit.com/articles/article.aspx?p=26535
    – Ben Brocka
    Dec 13, 2011 at 15:34
  • 4
    The problem is that the user isn't necessarily just done with your site, they could have been done for hours. Or they could have spent all that time browsing your site and others that use pop-unders and when they're finally done using the internet they now have fifty windows to close. The idea that "the timing" makes them more effective is ridiculous because by making a pop-under you surrender all control over the timing, it could be hours or even days before the user sees it.
    – Random832
    Dec 13, 2011 at 16:50
  • @Random832 That's good point about tab browsing and timing. Dec 13, 2011 at 17:12
  • I didn't mention tab browsing (though that makes the issue much worse, since the user may have had a dozen tabs open before they went to your site, and will work through all of them before seeing your pop-under) - simply going back to google after you're done on one site (and continuing to do so when they're done with their next task), rather than closing the browser window and reopening it, has the same effect, as would opening multiple conventional browsing windows (unless the pop-under goes in front of all the other windows)
    – Random832
    Dec 13, 2011 at 17:14
10

Pop up notifications, like the ones used in Gmail and here on StackExchange might be a good way to get your message across.

enter image description here

enter image description here

The method used on StackExchange is especially relevant because it offers information that is site specific, and often how to make ones experience on the StackExchange better. Since your message:

"You were looking for XY, if you are interested in getting new results as soon as they appear please enter you email address here"

is a way of optimizing the user's experience on your site, this would be a perfect use the pop up/down notification.

This solution isn't sneaky. It's upfront, appropriately timed, and relevant. All of which pop-unders are not.

5
  • Could be a good alternative but we're already using this space for other messages (similar to gmail).
    – Phil
    Dec 14, 2011 at 8:20
  • Perfect! So why not use it for this message also? If for some reason you need it to be distinguished from the others, change the color or make it larger. Dec 14, 2011 at 11:13
  • Two reasons: 1. I don't want to use the same system for very different purposes because I'm afraid that once a user learns that it's used for advertising he might ignore it when it'd be helpful. 2. There's the possibility that at one point the two things could collide (i.e. both would be displayed at the same time).
    – Phil
    Dec 14, 2011 at 11:26
  • What about at the bottom right corner like @Jberger suggested? Dec 14, 2011 at 11:34
  • Yep, that's an option we'll look at. I'll try anything to avoid the pop-unders :)
    – Phil
    Dec 14, 2011 at 11:43
9

This study seems to address your concerns exactly:

http://interruptions.net/literature/McCoy-HCIRMIS04.pdf

Here's the abstract:

Pop-up, pop-under, and in-line ads have been said to be intrusive, and previous findings suggest that they could have important effects on user perception and cognition. Using a 2x2 factorial design, this experimental study examines the effects of those ads. Besides a control group without ads, factors included ad placement (pop-up vs inline) and ad congruence (with the site's content or not). Results indicated that intention to return was impaired by ads; retention of website information was higher when ads were inline or when ads were not congruent with website content; and retention of ad content was higher for inline ads and those that were not congruent to the content of the website. However, contrary to expectations, intentions to return were not affected by ad placement, retention of site content was not affected by the existence of ads, and intrusiveness of ads was not affected by ad congruence.

8

So I have Firefox open with twenty tabs, many of them containing articles I want to read later. I then close Firefox, secure in the knowledge that the session is saved and I'll see the same twenty tabs again the next time I open the browser. And now I discover that there's another Firefox window open with your lousy ad in it. And that will be the last window closed. So next time I open Firefox I don't get the interesting articles in my saved session; I get your lousy ad. (Firefox saves the tabs of the last window closed.)

1
  • I came here to say this. In addition, the proposed implementation makes a lot of big assumptions, like that users close their browser windows and they use a platform that pop unders work on.
    – Malachi
    Dec 14, 2011 at 1:34
6

I can't link to any data or research; I don't see them used very often, probably for good reason. They might not interrupt workflow in the same way a pop-up might, but they're invisible at the time when the notification is most relevant! I would explore other ways of notifying the user. I would recommend some kind of inline validation rule that displays a <div> to the right of where the user is focussed, perhaps.

1
  • 1
    Good idea... we could A/B test a on-site fly-in vs. a pop-under. See you in London! :)
    – Phil
    Dec 13, 2011 at 13:16
2

Your example "You were looking for XY, if you are interested in getting new results as soon as they appear please enter you email address here" is conditional (only if they are interested in getting new results) and most pop-unders are modal which means that they need to be acted upon by the user, so you are forcing them to make a decision that they have not asked for.

The pattern that you are implementing is 'saved search' where the user 'follows' or adds a search as an 'interest'. Either way, saving a search is something that I'd expect to do easily if it exists but I wouldn't like it to be forced upon me.

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.