{{ searchResult.published_at | date:'d MMMM yyyy' }}

Loading ...
Loading ...

Enter a search term such as “mobile analytics” or browse our content using the filters above.

No_results

That’s not only a poor Scrabble score but we also couldn’t find any results matching “”.
Check your spelling or try broadening your search.

Logo_distressed

Sorry about this, there is a problem with our search at the moment.
Please try again later.

With marketers spending billions of dollars annually on paid search campaigns, accurately measuring conversions is a top priority. After all, conversion data provides important signals that marketers can use to manage budget and refine campaigns.

Unfortunately, conversions are often more difficult to measure than it seems they should be. According to a study by Marin Software, an ad management solutions provider, the proliferation of iOS devices, including the iPhone and iPad, is making accurate conversion tracking even more difficult.

The reason: the default browser on iOS devices, Safari, blocks third-party cookies unless users take steps to change settings most probably don't know even exist.

Because many marketers rely on third-party solutions that use redirects in the measurement of conversions, "the growth of iOS represents a threat to some advertisers’ ability to accurately measure and optimize the performance of their paid search marketing campaigns".

This threat isn't theoretical either. In its study, Marin Software found that a whopping 80% of conversions on iOS devices were "missed" when third-party cookies were used.

This, of course, means that marketers may be undercounting the vast majority of their iOS-based conversions. As the study notes, "most advertisers have assumed that the degree of undercounting is uniform across all browsers", but if this is truly not the case, the implication is far reaching: the underreporting of conversions from iOS devices has the potential to create a false impression of these devices' worth.

So Marin Software looked at actual conversions, and it found that "the actual conversion rates for iOS (i.e. adjusted for the third-party cookie based undercounting) were on average 23% higher than on Windows".

In other words, search traffic on iOS devices may actually be productive at driving conversions, something most search marketers would probably want to know.

Right now, of course, iOS devices are generally driving a small percentage of paid search traffic (in some cases up to 5% according to the study), so even though conversions may be underestimated by as much as 80%, it's not the end of the world for search marketers.

But given the rise of the iPad and the growth of mobile search, one has to assume that iOS devices will grow in importance to search marketers. As that happens, being able to accurately measure conversions on these devices will be crucial. 

Unfortunately, marketers probably shouldn't rely on Apple to change its default privacy settings on iOS devices. That may provide give marketers a reason to look at solutions that avoid the use third-party cookies.

At the very least, however, even those who can't get away from third-party cookies should be aware of the limitations of their data collection techniques and take them into account when managing campaigns.

Patricio Robles

Published 21 March, 2011 by Patricio Robles

Patricio Robles is a tech reporter at Econsultancy. Follow him on Twitter.

2403 more posts from this author

Comments (12)

Comment
No-profile-pic
Save or Cancel
Avatar-blank-50x50

Email Marketing Solutions

The iOS conversion rate is pretty interesting. Undercounted by 80%? It seems like there is huge potential to tap into for search marketers if they can make it more accurate.

over 5 years ago

Ben Sidebottom

Ben Sidebottom, Media Systems Manager at Essence

This is not strictly true. Whilst Safari does, by default, block the SETTING of third party cookies, it doesn't block the READING of them. This means if you use a redirect on your marketing like PPC and Display, the redirect can set a First Party cookie (which is accepted) and then read by the 3rd party conversion pixels later. So the "3rd Party cookie redirect based trackers" graph I don't really understand? All redirects should be setting First Party cookies.

over 5 years ago

John Braithwaite

John Braithwaite, Managing Director at Ergo Digital

Not only that, but using Google Analytics avoids Third Party Cookies altogether... so as long as you resolve Analytics with your ad data you will get a true reflection.

Post sponsored by Apple?

over 5 years ago

Ben Sidebottom

Ben Sidebottom, Media Systems Manager at Essence

True (and that applies to most analytics tools, not just GA) but you have to take into account loading times over 3g/edge/gprs on mobile devices so if you load a large JS file (and the GA one is pretty beastly and GA doesn't do a noscript tag) you could get a pretty large discrepancy in GA clicks and actual clicks. There are many ways to track mobile perfectly reliably with redirects and JS tags all using First party cookies. I think there has been a misunderstanding with 3rd party cookies and how browsers handle the setting and reading of them.

over 5 years ago

Simon Whittick

Simon Whittick, Marketing Manager at Marin Software

Hey Ben,

Thanks for picking up on this discussion. Good to see you're still a Don of the Web Analytics industry over here ;-)

As we've now highlighted elsewhere, in the conclusions of our original research brief, we occasionally referred to tracking solutions that rely on redirects and third party cookies interchangeably, and we’ve now made a slight correction to that. Many advertising solutions that rely on third-party cookies also rely on redirects, however, redirects do not require the use of third party cookies. Some tracking solutions, for example, use redirects to deploy a first-party cookie from the vendor’s domain which is later read on the confirmation page of the advertiser.

As such, you can download the updated research brief here, providing greater clarity:
http://marinsoftware.com/downloads/marin_ipad_research_brief.pdf

Again, thanks for highlighting, it's all healthy debate in an area where there isn't enough public discussion.

Catch up soon,
Simon

over 5 years ago

Avatar-blank-50x50

malleylaw@gmail.com

Is the "Problem" the technology that circumvents the iOS default settings and user's security preferences, or the use of a cookie from a third party domain coded as a first party domain, a domain other than the domain and parties of the original page requested by the user?

over 5 years ago

Ben Sidebottom

Ben Sidebottom, Media Systems Manager at Essence

Hey Simon,

No worries. You know I cannot turn down a good discussion when it comes to the geeky side of analytics and data collection!

@Malleylaw - There is no "circumventing" of settings. It is the way the browsers are developed. Cookies are managed by the browsers, and how they are dropped or read is actually up to the browser. It just so happens that Safari, IE and Chrome ALL allow reading of 3rd party cookies, regardless of the setting. Only Firefox (AFAIAW) blocks 3rd party cookie reading when 3rd party cookies are disabled.

Also there is no "coding". A domain/site cannot dictate what type of cookie it is setting. It just tries to set "a cookie". The browser then decides what type of cookie it is based on the URL the browser is on and the URL dropping the cookie. As a redirect IS the URL the browser is on for a split second, the browser accepts it as 1st party.

Hope this helps.

over 5 years ago

Avatar-blank-50x50

joe malley

Is the "problem" the technology that circumvents the iOS default settings and user's security preferences, or the use of a cookie from a third party domain coded as a first party domain, a domain other than the domain and parties of the original page requested by the user?

attorney joe malley
dallas,texas
malleylaw@gmail.com

over 5 years ago

Avatar-blank-50x50

malleylaw@gmail.com

Please excuse the double posting. It initially appeared that my posting had been deleted. For some reason blogs seem to delete inquiries from the "Privacy Crusader".

Ben: Thank you for your information and interpretaion of my inquiry. Please allow me though to clarify the reference to "circumvention" of settings referred to any method to bypass the user's iOS default settings and user's security preferences, without notice or consent, as such, as to technology existing, or being developed, to bypass the user's security preferences, although a "problem", is never in doubt.

Since you are in the business, and the concensus by some appears to be that this process is legal,ethical, and is w3c compliant, can you please email lists of all websites, companies involved in redirection,ad networks and web analtic vendors using this technique, with server logs so I can view this process. Anyone else that reads this post can also email me such information for my review.

As for the "coding" inquiry, I was not interested in coding as it relates to setting of browser cookies, but the coding, used by the URL in the original landing page, as it relates to an internal rewrite to the subdomain url, prior to the external re-direct and cookie SET.

As for w3c compliant:
1) would the URL located in the original location response header be required to provide notice within it's compact policies of the existence and activities of the subdomain/redirect, within it's domain,involved in the "read only" and a soft redirect?

2) would the compact policy of the subdomain/redirect entity even be read, at the read only phase, if the original url did not mention the existence of the subdomain?

over 5 years ago

Ben Sidebottom

Ben Sidebottom, Media Systems Manager at Essence

Hi Malleylaw,

Sorry but I don't get your comment about the circumventing. There is no circumventing of settings by using redirects. It is just the way the browsers work.

If I was to send you a list, it would be very long. Lots of company do this and some research can show some of these.

Compact privacy policies, as far as I am aware, are only used by IE? Don't think any other browser uses them.. *I think*. Also CPP has nothing to do with the redirects of a URL so not sure what your points are for 1 and 2? They are to do with how the browser accepts the cookie.

The rule is, if it works, it works. If it doesn't, it doesn't. Anything that does work is within w3c's compliance and does not "bypass" any settings. If a user is worried about this, then they can change their settings within the browser or simply deactivate all cookies (though 80% of sites would then refuse to serve you if you disabled *all* cookies).

over 5 years ago

Avatar-blank-50x50

malleylaw@gmail.com

The rule is, if it works, it works. If it doesn't, it doesn't. Anything that does work is within w3c's compliance and does not "bypass" any settings.
Ben:
I note the following:
1) In lieu of using the word, "circumventing", we can use bypass;
2) I do not agree with your thoughts:"The rule is, if it works, it works. If it doesn't, it doesn't. Anything that does work is within w3c's compliance and does not "bypass" any settings."
So viruses,malware,spyware, worms "work". What about the use of flash cookies, history sniffing, obtaining UDIDs or GPS? Man-in-the -middle attacks,that WORKS!
w3c requires a domain with a computer-readable/ full p3p only to provide notice that it has a subdomain that obtains Pii, and for the subdomain to have a CP,and for the CP to be linked from the domain. It "works" though if that w3c obligation is not meet, and BYPASSES the user's securtiy settings and iOS default settings.
3) "If a user is worried about this, then they can change their settings within the browser"
The user relies on the iOS default settings and their own security settings limiting third party cookies. The "PROBLEM" is that technology "WORKS" to bypass these settings.
4) "If I was to send you a list, it would be very long. Lots of company do this and some research can show some of these."
Don't worry, I like long lists, but let's keep it to the top 20 largest companies using this redirect method. I would to need to map the server paths so I will need:
a)20 of the largest websites using this technique. Review the list of most visited websites,
b) redirect companies associated with these websites. I will need logs to show the association btw website and redirect company;
c) Web analytic vendor associated if different than the redirect company;
d) ad networks associated with redirect and/or web analytic vendors.
I would also suggest you write a study in this area detailing this activity with documented support. Please provide me the data as you develop your study.
Thanks,
malleylaw@gmail.com

over 5 years ago

Ben Sidebottom

Ben Sidebottom, Media Systems Manager at Essence

Hi Malleylaw,

I think you are going off the topic a bit. I am strictly talking about 1st and 3rd party cookies and they way they are handled. Not viruses or worms.

Points 2 and 3... when you block 3rd party cookies, all browsers stop the setting of them, which is the compliance part. Not the reading of them as a cookie isn't set as 1st or 3rd party - a cookie is just a cookie! The 1st and 3rd party part of the cookie is how the browser sets them from which domains.

Point 4, sorry, but I don't have time to produce a list for you. I am not developing any study, I was just passing on my comments on a post.

Good luck with your own studies but I think this topic is done for me.

All the best,

Ben

over 5 years ago

Comment
No-profile-pic
Save or Cancel
Daily_pulse_signup_wide

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Daily Pulse newsletter. Each weekday, you ll receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.