I’m a huge proponent of campaign tracking for marketing campaigns. But used the wrong way, they can totally trash your Google Analytics data. Like to the point of making the data in your profile completely unusable. And since I’ve run into this in the past six analytics audits I’ve done, I thought I’d aim to raise the level of awareness on this topic.
What Is Campaign Tagging?
If you’re unfamiliar with campaign tagging, check out my comprehensive guide on campaign tagging, where I talk about marketers not taking enough credit for their work. It is beyond critical if you’re doing email marketing. If you’re doing email marketing and you’re not tagging your links, you’re junking up your direct and referral traffic numbers. (Can you say bogo special?)
A Cautionary Tale
But this post isn’t about the efficacies of campaign tagging; it’s a cautionary tale about what can go wrong if you use campaign tagging improperly. Why? Campaign tagging is meant for external links that point back to your site.
Ergo, the best way to trash your data is to use campaign parameters in internal links.
Deep In The Weeds
The reason for this is when you tag a link with campaign parameters using Google’s URL Builder or my Google Doc that enables you to just paste a list of URLs and have the parameters added automatically (wooOOoop!), you are reassigning the source and medium data. So let’s say you tagged a link to look something like this:
And someone clicks on this link from from hootsuite.com. Because I assigned the source to be twitter.com (via utm_source=twitter.com), that referral source will now show up as twitter.com, not hootsuite.com. And instead of the visit showing up as a referral, it will show up as social because I assigned the medium to social (via utm_medium=social).
You following me?
So what I’ve seen sites do is add campaign parameters to their navbar and footer links or even internal banners. The problem with this is let’s say you have a navigation link tagged, and it looks something like this:
http://www.yoursite.com/about/?utm_medium=top+navbar&utm_source=homepage&utm_campaign=nonsense
Now let’s say someone comes to your site via Pinterest and then clicks this tagged link to get to your about page. That visitor no longer shows up as coming from Pinterest. Your referral is your own site.
Worst Case Scenario
In auditing a company’s Google Analytics profile this week (a service I offer), they had 1.4 million sessions in the past month that had been reassigned. My heart stopped. I’ve seen campaign tagging cause shenanigans in data but never anything this pandemic. It was the first time I had to tell someone that their analytics data is completely useless. As in they can’t use their analytics data to measure any of their marketing efforts — organic, paid search, campaigns — nothing. Because they had tagged links in their navigation and all through the site, those reassigned visits were just too pervasive.
So if you’ve been using campaign tagging on internal links you need to see what percentage of your visits have been reassigned. If it’s a significant amount of data, I would recommend consulting with an experience analyst to make sure your data is trustworthy at all. Here are a couple preliminary checks I use to check for evidence of tagging internal links.
Learn More
If you want to see if you’ve been tagging your campaigns correctly, and how the rest of your Google Analytics is doing, check out my DIY Analytics Audit Template.
You can also check out my Definitive Guide to Campaign Tagging in Google Analytics for a more in-depth look at how you should be tagging your campaigns.
I’m really glad you wrote this post Annie – It is something I have seen clients do far too often and it will be nice to have a resource to refer them to in future. Thanks.
Thanks. I kept thinking it was a one off when I’d see it, and I put off writing the post because I never want a client think I’m writing about them. But then I’d forget. So this time I just decided that this is probably pretty prevalent among enterprise-level organizations.
Interesting article – how would you track internal links then?
With events. Campaign tagging for external; event tagging for internal.
Thanks for this very simple summary. Is this still valid for the “Universal Analytics” stuff being rolled out? Or will we need to use “custom dimensions” now?
Custom dimensions replace custom variables, not campaign tagging. So, yeah, this is still very relevant with UA.
Great article Annie that made a sometimes complex subject to some straightforward. Crazy how dangerous this could be to future data.
Yeah, it’s scary how easy it is to annihilate your data. Oh, and good to see you here, Joel! 🙂
Thanks Annie, always keeping an eye on the cool stuff you’re doing. I appreciate the breakdown and I’m sure my team does too. 😛 Have a great week, keep cranking out great stuff.
Thanks, Joel! You too! 🙂
ork ork ork I just came across a giant client competitor site that had done this on Friday.
You just want to email them and say, “Good job, guys! You got this.”
Is there any value in using campaign tracking for secondary domains that are 301 redirected to the primary domain name? Example: domain.org redirected to domain.com (with campaign tracking). Or would this be making the same mistake?
As long as it’s external, it’s totally fine. I recommend a similar strategy for billboards. Put up a vanity URL, redirect it to a page on your site with campaign parameters. Totally copacetic.
Wow, can’t believe you’ve been seeing it that often lately Annie. That’s a shame. In your very first example, (Hootsuite vs. Twitter), there’s not really a way around that, right? I mean, traditionally, I’m saying that even though the click actually came in from Hootsuite, it makes sense to me to still give the credit to Twitter (assuming the original act was me tweeting that link) since the url started there. Agree? Or am I missing something.
No. That’s a totally acceptable use of campaign tagging. I’d rather have Twitter traffic be consolidated into twitter.com. I just used that as an example of how campaign tagging used properly works.
Very interesting… Didn’t know it could be that huge !
How did you discover the problem ? In analytics ? With a crawl (screaming frog like) ?
These visits show up as suspicious sources, mediums, and campaign names in their campaign reports (e.g.,anything with “internal” in it). Then I verify by running Screaming Frog and using the line item filter to search for utm_.
I’m assuming if you still want this level of visibility “event” tagging would be more appropriate.
Yes, event tagging is for internal links; campaign tagging is for external.
I agree completely. I always prefer event tracking for the internal links rather than using UTM builder. That’s the best way to decrapify my Google Analytics data. Events are a good indicator of user behavior and interaction with my site and also offer valuable insight as to what content/link should be highlighted and what to demote and remove.
onClick=”_gaq.push([‘_trackEvent’, ‘top_navbar’, ‘Click’, ‘awesome_insight’]); is better anyday over http://www.yoursite.com/about/?utm_medium=top+navbar&utm_source=homepage&utm_campaign=nonsense
What do you think?
Yes. I’d make my category broader than top_navbar, but that’s an issue of preference.
We ran into this problem as an unintended consequence of good intentions. We use a “short URL” redirect script for use in print (postcards, etc.) The developer set it up to automatically append campaign tags, which was great until we realized that content authors were also using the short URLs in their web pages. Any visitor clicking one of these links would get their true source info overwritten.
It’s reassuring to know that we weren’t the only ones to fall into this trap! It makes me wonder, why doesn’t Google preserve the http referrer as well as the campaign source?
When used properly, you want this data to be overwritten. That way you can segment by it in more than just your campaign reports. It just has to be used with extreme caution.
Very interesting Annie, thank you.
You’re welcome!
I think it could called the “omniture syndrome”: they need to tag-tag-tag-tag because it’s the “only way” they know of tracking the user interactions in a website and in a conversion funnel. Thankfully analytics has events tracking.
Heh. Omniture Syndrome. 🙂
What about campaign tagging links that don’t come back to your site? For example, should I bother campaign tagging links for our newsletter that go to other peoples’ blogs?
Great question. No, you should never tag links that point to someone else’s site b/c it routes their traffic into reports they might not be monitoring. I liken it to painting your neighbor’s house.
Do campaign parameters impact your pageview counts or does GA filter that out when aggregating results? Is that another reason to be concerned about campaign tagging?
No more than any other referral. I think you might be thinking of virtual pageviews. They bloat pageview counts.
Yes, this has been a common issue with some of our clients too. In addition to the attribution issue, doesn’t this overstate the visitor count? Since there are two source/medium pairs for the same visitor, aren’t both counted in visitors reports as two different visitors?
Re:Custom variables/Custom dimentions (UA). In one of Justin Cutroni’s articles he has suggested custom variables as an alternative for internal campaign tracking. Appreciate if you can shed some more light on this preferably using GTM
A visitor is only recorded once b/c the source and medium are overwritten so no double dipping.
Custom variables or event tracking can be used to track clicks on internal links.
Interesting.. In this article Lunametrics says GA records two visits if a visitor clicks an internal URL which is tagged with Google campaign tracking variables.
http://www.lunametrics.com/blog/2012/01/26/track-conversions-internal-external-campaigns/#sr=g&m=o&cp=or&ct=-tmc&st=%28opu%20qspwjefe%29&ts=1393463537
Campaign tagging, in and of itself, doesn’t overstate visitor count. Using campaign tagging on internal links will overwrite referrer data mid-session. If that’s what you were saying, then yes. But that was addressed in the post, so I thought you were addressing a new issue.
Hi Annie! Thank you for this useful example of how NOT to use campaign tagging 🙂
I was curious about Suranga’s question, so I checked Google’s documentation on campaign tagging (https://support.google.com/analytics/answer/6205762) and found this sentence:
“[…] a change in value for any of the following campaign URL parameters triggers a new session: utm_source, utm_medium, utm_term, utm_content, utm_id, utm_campaign, gclid.”
This seems to confirm that changing any of the UTM parameters does indeed produce a double counting of the visitors. Another reason to avoid internal campaign tagging 🙂
Writing is an art that everyone does it. Congratulate you for having this art.
کرکره برقی
With onClick= is there anyway to track revenue? I should mention, for an ecommerce site. Not just setting a value. I am running a test and all I can see is click data.
No. Even if there were, it would – in most cases – be highly unreliable data.
I have been surfing online greater than three hours as of
late, yet I never discovered any attention-grabbing article like yours. It¡¦s
lovely worth enough for me. Personally, if all site owners and bloggers made
excellent content as you did, the web will probably be much more useful than
ever before.Check here
Hi Annie, Do you have a guide for setting up events for internal tracking or any pointers on that. I’d like to know how I track a URL visited internally if for example it came from a homepage banner vs. another placement on another page. I tend to look at previous page path, but am not always clear that’s from the banner if there were 2 placements on a page, etc.
Sorry for the horrifically late response! I didn’t realize I wasn’t receiving comment alerts. :/
This is a great starter resource for event tracking.
Hi Annie,
Great article! I recently encountered a website using UTM tracking on internal links. If they aren’t using UTM parameters to track things externally (because they have other methods), is this still messing up their analytics data? My thought process being– if there is no tracking to override, their data wouldn’t be skewed.
Am I completely misunderstanding?
Hi Molly, if they’re using campaign parameters on any internal links they’re trashing they’re data. The reason is that they’re overwriting the referral data. So if a visitor comes from Facebook and clicks on a link that sets the source (via utm_source) to topnav, that visitor won’t show up in social reports b/c s/he no longer came from Facebook; topnav was the referer as far as GA is concerned. Hope that clears things up!
Thanks, Annie! So this issue could potentially impact ALL data within Google Analytics? Organic traffic, visits, pageviews, goal completions, etc.? I apologize if the answer seems obvious, I just want to make sure that my recommendations are accurate.
Also, is there any way to remedy this issue? My initial thought is to remove UTM parameters from all internal links and use event tracking or Google Tag manager for future behavior tracking.
Thanks again! I’ve had a really hard time finding information on this subject.
I can’t believe I used the wrong their. Sigh.
Anyway, campaign tagging will indeed impact sessions, pageviews, and conversions. Let’s say you overwrite some of your organic traffic as internal_campaign, your pageviews will also show up as coming from internal_campaign. And if visitors convert, the conversions will be attributed to internal_campaign.
Thanks so much! This is all very helpful. One last question. (I promise it’s the last one !!) Could internal UTM tracking be the cause of self-referrals?
If you set the source to your own site, yes. So if I set utm_source=annielytics.com, I’m going to have self-referrals, even if I’ve excluded my own domain under Admin > Property > Tracking Info > Referral Exclusion List.