Same question. I've only found 1 article on Google and the company wants you to pay them for the answer. Looks like it started on 9/2/20. Hope Shopify fixes it, soon. It's almost 15% of our revenue.
Hello and thanks for your message.
Just an update, I spoke to the Shopify support chat. They said that shop.app / referral is probably the Shopify Shop App, though they still advised me to check with the Community.
The shop_app source is still unclear.
We cross checked shop.app usage with realtime referrals in GAnalytics. No activity was reported during the test. We're wondering if this could be incorrect attribution from further in the payment process or one of the following:
1. Is Shopify pushing certain users to adopt the app mid purchase, are they being prompted to complete the order in the app? (doubt it)
2. Could the merging of Shopify Pay and the Shop app be causing improper attribution? If certain customers are using the Shopify Pay feature, they might be included in that shop.app referral traffic.
So, it definitely looks like it's the new Shop app. I downloaded it and referral traffic would be the correct way to categorize it. Take a look at the app. There's a new home screen that recommends products to users from Shopify stores that they've visited in the past. We actually got a ton of sales from it since it launched on 9/2, and it's free traffic.
Since shop.app started to appear as a source it's accounting for around 18% of a client's sales - all appearing as last touch.
It's possible (given Labor Day) but I think unlikely that all of this is incremental revenue rather than a last touch in the purchase journey.
If so, this is a problem, because it's mis-attributing - what's important is how the customer got to the app.
Any ideas? Would adding shop.app in the referral exclusion list in GA solve?
All of our shop.app referrals came with a pathlength of 2+ (literally 0 a single touch).My best guess is that this is shopify payments, though our GA numbers and pay numbers inside shopify do not match so I'm not 100%, but as the CVR is about 68% on this referrer, I'd say it's definitely something to do with checkout.
I added it to my domain referral filter about 3 hours ago. It looks like the solved the problem, but I won't know for sure until tomorrow morning when I have a bit more data
This is an accepted solution.
This is new, but we have done analytics work for a Shopify client who had this come up and solved it. Conversion rates were in the 90% from the traffic source which typically happens only from self-referrals or payment gateways.
There are two things going on. Firstly, the app itself will send traffic from the shop.app utm_source. These are people who have the Shop app installed. This isn't best practice for Google Analytics because utm_medium is not set which will lead to such warnings in analytics:
Secondly, the app is a quick checkout option for users at checkout like, say Google Pay. This is likely to be the source of all most traffic from the app for sellers, at least for now until the app grows.
Unfortunately, traffic between the two sources are not differentiated so you'll need to treat them the same by dealing with what's most valuable to you, and that is not having the app at checkout mess with your purchase data. You fix it by adding shop.app to your referral exclusions.
I've updated our ultimate Google Analytics guide for Shopify that explains the need to add shop.app to the referral exclusion list.
The shop.app / referral issue started on September 2, 2020 on clients with the Shopify Quick Checkout enabled.
When visitors to your website utilize this Shopify feature refer gets reset to shop.app. While trying to be helpful and demonstrating the value of the quick check out options this will disrupt traditional referral traffic numbers. It is especially damaging when comparing transactions or revenue generated. On one client's analytics, we saw 100% of over $8,000 in sales attributed to this referral. Not knowing the true traffic source, could lead to investing resources into traffic sources that never performed, or ignoring traffic sources that were performing well.
While this may seem like a one-off bug this previously appeared as shopify.com. PayPal's check out options will do the same.
For this reason, we recommend to all of our clients to block these domains utilizing the Google Analytics Referral Exclusion list. Here are Google Analytic's instructions around this feature: https://support.google.com/analytics/answer/2795830?hl=en