We are running ads for a Shopify store that is doing hundreds of thousands a month.
In Facebook, it says our ads are crushing it with an average ROAS of 15 and brought in an extra $50k for my client.
Our Cold Traffic campaign says it brought in 72 Purchases.
However Shopify is reporting only 1 conversion:
I understand that the attribution models are different but this seems too different. We are worried that we are over reporting our generated revenue to our customer.
What is the easiest way to 100% guarantee the Pixel / Shopify is working and reporting correctly?
We are also having this same issue on Klaviyo. Shopify is dramatically under reporting what Klaviyo is reporting. We are using UTM tracking.
Any and all help is greatly appreciated. Thank you.
Solved! Go to the solution
There is no way to get this 100% perfect. It's better to use something like Google Analytics for you reporting because it will look at each channel the same way. The issue with using SHopify for one thing, FB for another and Klavioy for another is that they all report on data differently for a campaign.
Facebook will take credit for any user that was served an impression of an ad, then got back to the website and made a purchase, regardless if they clicked through on the ad. This is called view-through (or post impressions) conversions.
Klaviyo will report in a similar way, but instead of post-impressions on an ad, they will count conversions after a user opened an email, even if they do not click through from the email.
Shopify is only going to give credit if a users clicked through on an ad/email, then made a purchase - so the difference between these three sources will be very different.
In FB you can run a report looking at just the through (or post-click) conversions to get a better sense of how users that clicked through converted, which will get you closer to the numbers Shopify reports.
For the FB pixel, did you just implement through the Shopify preferences? You cannot implement in any other way or it will not track properly.
To test the pixel firing, add the Facebook Pixel Helper to Chrome and make a test purchase to make sure the pixel is firing correctly.
Hope that helps!
Thank you for the reply.
I forgot a couple of important notes that were pointed out to me. Our client is on Shopify Plus and they already have a pixel setup in their preferences that we are not associated with. We had to use GTM in order to setup our pixel - so the store now has 2 pixels on it. Both seem to firing correctly although our pixel gets a warning that it is not setup with a product catalog.
Does this additional information make any difference to your assessment?
Hi Jack -
Thanks for the additional info. Yes, this makes a big difference. Are both FB pixels writing back to the same FB account? You don't want the FB pixels from the same account firing twice (and will double count conversions among other things), which it will if it is setup in GTM and in Shopify.
This is an accepted solution.
Ok got it - this is not an implementation issue. You need to associate the pixel with FB in the catalog manager : https://www.facebook.com/business/help/1044262445604547
So, back to your original question, it looks like the pixels are setup properly and this is an platform attribution issue. To give you an example of the discrepancies between FB and Google Analytics (which reports similarly to Shopify (with some nuance)),
I have a client where FB is taking credit for 21,968 transactions, but GA is only showing 1,976 that came from FB during this same time period. This is because FB is taking credit for every user that saw an ad and made a purchase up to 30 days after seeing an ad, regardless if they clicked.
There are many different ways to look at attribution. At the minimum for FB, I would only give credit to 15% of the post-impression conversions and 100% of the post-click conversions (which is pretty standard in the industry). That way it normalizes the numbers a bit, but still gives FB credit for influencing a conversions based on the users seeing ads served on FB, not just clicks to the website.
Does that help?