Rise in Direct Traffic & New Users in Google Analytics? It Might Be ITP…

Have you noticed a significant increase in Direct traffic in your Google Analytics account recently? Maybe it’s coincided with a drop in Organic traffic? This is something we’ve seen happen quite significantly for a number of our clients – in particular over the last couple of months – so we decided to investigate further.

Below you’ll see a proportionate visualisation of traffic by channel over time – prior to Feb 2021, Direct traffic accounted for roughly 10% of the total each month. Since then, it now accounts for just over 30% of the total.

As you can see in the example above, there was a sudden and sustained increase in Direct traffic around March 2021. Nothing changed with the tracking implementation on the site around this time, so why has this happened?

On further analysis, we saw that in March/April 85% of Direct traffic came from a Mobile device, 90% were using the Safari browser and more specifically, 87% were using Safari version 14.0.3!

What is Intelligent Tracking Prevention?

Intelligent Tracking Prevention (ITP) is a privacy feature developed by WebKit and used by Apple’s Safari browser since 2017, however, it has ramped up protection over the years and with the most recent versions of iOS and Safari, all of the Webkit Intelligent Tracking Prevention mechanisms are turned on for users by default.

PRO TIP: the best place to keep up-to-date with the most recent tracking protection mechanisms implemented by major browsers is cookiestatus.com

Why does ITP cause an increase in Direct traffic in Google Analytics?

One of the key points is that WebKit restricts the lifetime of JavaScript cookies to a maximum of 7 days – and in some instances, it can be as short as 24 hours.

If we also consider the attribution model that Google Analytics uses – Last non-Direct Click – prior to ITP, someone could find your website via an Organic Google search for the first time, come back within TWO YEARS by typing the address into their browser directly and that second session would still be credited to Organic due to the attribution model (Last non-Direct, remember?)

Now, thanks to ITP, if they come back more than one week later, it’ll be considered Direct traffic because there is no longer a cookie linking that session back to the original non-Direct session – it’ll have expired after a maximum of 7 days.

Why does ITP cause an increase in New Users in Google Analytics?

It’s the same story here I’m afraid… now that those cookies only last for up to 7 days – if someone comes back again more than a week later, they’ll be seen as a totally new user even if they’re using the same Safari browser and device they were previously.

So what you’re likely to see now in Google Analytics is a big increase in perceived New Users and much less transparency over the original source of that traffic, with much more lumped into Direct because we no longer have the cookies to tell us which channel that user really came from in the first place, specifically if they’ve returned to the site again Directly a second (or third or fourth…) time.

This will be particularly noticeable if you run an ecommerce website selling high value items with a lengthy time lag from first visit to purchase i.e. if your users visit your site, go off and think about it for a week, compare prices elsewhere then finally come back to buy three weeks later, you may find it much harder now to properly attribute sales to the original acquisition channel.

Have you noticed a similar increase in New Users & Direct traffic in recent months? If so, let us know in the comments below!

7 responses to “Rise in Direct Traffic & New Users in Google Analytics? It Might Be ITP…”

1. Rajiv says:

   July 12, 2021 at 1:54 pm

   We have seen this recently on multiple projects. One of the projects however shows a sharp rise in pageviews on the homepage from both direct and organic traffic. This project receives 90%+ traffic from organic search.

   Any thoughts on what could be causing this? It’s only the iOS visitors that show a rise.

   Reply
   • Michael Wilson says:

     July 16, 2021 at 1:40 pm

     Hi Rajiv,

     Not really sure why you’d be seeing an increase in pageviews on the homepage specifically – how big an increase are we talking? Is there any kind of cookie consent banner on the site? If so, maybe it refreshes the page and you’re getting duplicate pageviews? Are there any redirects?

     Hard to answer without knowing more details – but if you want to follow up via email, I’m happy to take a look.

     Thanks,

     Michael

     Reply
   • Roshan says:

     April 19, 2022 at 9:42 pm

     The exact scenario Rajiv described happened to us on ~August 18, 2021, as well as around ~June 4, 2021 and ~April 27, 2021. The combined increase has been staggering and multiplied our traffic many many times.

     Did you ever look at this into more detail Michael?

     Again, iOS traffic only, both organic and direct to the homepage. Typically we didn’t receive much traffic to the homepage based on the nature of the site and no other variables can justify these increases.

     Reply
2. SEODan says:

   August 24, 2021 at 10:55 am

   This is insightful, seeing this issue on my website too, is there any setting on GA that can be done to showcase that organic/SEO is still working very well?

   Reply
   • Michael Wilson says:

     August 24, 2021 at 2:43 pm

     I would recommend applying a Segment for Organic traffic – and using that to look at your organic traffic performance over time.

     You can find out more about using Segments here:
     https://support.google.com/analytics/answer/3123951

     Reply
3. Will S says:

   August 25, 2021 at 12:27 pm

   I have seen the same trend, an increase in users and traffic lumped into Direct. Also the same timing being March 2021.

   Do you have any insight into how ITP affect UTM parameter urls?

   Reply
   • Michael Wilson says:

     August 25, 2021 at 12:44 pm

     UTM tagged links will still function as usual but you might still see the same thing I suppose – if a user lands on the site via a UTM tagged link and then comes back ‘Direct’ a second time 7+ days later, they’ll be seen as a new Direct user because the cookie used to identify them as someone who visited previously from a campaign has expired.

     Reply