Facebook Conversion Tracking: Why Aren’t the Numbers Adding Up?

513 Shares Facebook 142 Twitter 202 Buffer 71 Google+ 18 LinkedIn 80 Pin It Share 0 513 Shares ×

facebook conversions featured Facebook Conversion Tracking: Why Arent the Numbers Adding Up?

[AUDIO VERSION: I also recorded an audio version of this blog post. Click below to listen. Let me know if this is something you find helpful!]

It’s a question I get almost daily. You have a Facebook conversion pixel installed on your site. But the numbers aren’t adding up.

Today I’m going to explore the problem itself before diving into the likely explanations.

The Problem

The problem tends to be when advertisers compare conversion numbers reported from Facebook to the Facebook referred conversions reported by Google. In most cases, Facebook is reporting more conversions than is Google.

facebook conversions reported Facebook Conversion Tracking: Why Arent the Numbers Adding Up?

Many incorrectly assume this means Facebook is over reporting conversions. That’s usually not the case.

The Wrong Page is Set as a Conversion

The first potential explanation for this problem is that you placed the conversion pixel on the wrong page, so Facebook is reporting inflated conversion numbers.

When you create a conversion pixel, Facebook instructs you to paste it between the HEAD tags “in the webpage where you want to track conversions.”

facebook conversion pixel code 700x366 Facebook Conversion Tracking: Why Arent the Numbers Adding Up?

When a user who was shown your ad on Facebook eventually visits that page that fires the conversion pixel, Facebook is notified. This is how they are able to report the number of conversions resulting from your ad.

Understand that Facebook doesn’t naturally know what a conversion means. You have to define it for them. And if you define it incorrectly, Facebook will report inaccurate results.

The mistake many advertisers make is that they place the pixel on the sales landing page. This is the page a user will visit immediately upon clicking on your ad. If a user visits this page, however, they don’t necessarily convert.

You need to put the pixel on the page a user will see immediately upon completing the conversion. It’s usually a “Thank you for registering” or “Thank you for your purchase” page.

It is a page that only new customers will see, and it notifies Facebook — correctly — that the sale is complete.

Facebook Referrals

The second likely issue is that advertisers are using Google Analytics and incorrectly assume that the number of conversions reported by Facebook should be the same as the number of conversions referred by Facebook, as reported by Google.

google analytics facebook conversion referrals 700x366 Facebook Conversion Tracking: Why Arent the Numbers Adding Up?

These will not add up due to the different ways that Facebook and Google are reporting these numbers.

Facebook Conversions Defined

Your Google Analytics are only reporting conversions that resulted from a direct referral. Even if you created a campaign using URL parameters to track the link used in your ad, Google only reports those conversions that happened in a straight line:

User Clicked Ad > User Converted

Facebook, on the other hand, is much more liberal in their conversion reporting. By default, they count either of the following as a conversion:

  • User Viewed Ad and Converted Within 1 Day
  • User Clicked Ad and Converted Within 28 Days

Even if you eliminated “view through” conversions where users never clicked your ad, the numbers won’t add up since Facebook will track conversions that weren’t a straight line.

Like this:

User Clicked Ad > User Left Site > User Returned to Site > User Converted

Neither method is necessarily “right” or “wrong.” They’re simply different.

There’s value in knowing whether your ad led to “view through” conversions. And there’s also value in knowing that someone who clicked your ad later converted, even if it wasn’t immediate.

Google ignores these things in their conversion referral reporting.

Changing a Defined Conversion

As mentioned above, Facebook reports a conversion when a user has viewed your ad and converts within a day or clicks your ad and converts within 28 days.

However, this may be casting too wide a net. Maybe you want to isolate only those who clicked your ad. And maybe you want to change the number of days to conversion.

First, click “Reports” on the left within your Ads Manager. That will bring up a pretty worthless default report.

facebook ad reports edit columns Facebook Conversion Tracking: Why Arent the Numbers Adding Up?

You can make the report useful by clicking the “Edit Columns” button. This is also how you can change the definition of a conversion.

When you click the Edit Columns button, you’ll get a dialog like this…

facebook edit columns actions 700x366 Facebook Conversion Tracking: Why Arent the Numbers Adding Up?

Click on the “Actions” link in the middle menu. At the top you’ll see that the definition of a conversion is set to 1 day after viewing and 28 days after clicking.

If you click the “Change Attribution Window” link, you’ll get this…

facebook edit columns attribution window settings 700x366 Facebook Conversion Tracking: Why Arent the Numbers Adding Up?

There, you can choose any of the following to define a conversion:

  • 1 day after viewing ad
  • 7 days after viewing ad
  • 28 days after viewing ad
  • 1 day after clicking ad
  • 7 days after clicking ad
  • 28 days after clicking ad

Your Turn

What’s been your experience with tracking conversions between Facebook and Google?

Let me know in the comments below!3

513 Shares Facebook 142 Twitter 202 Buffer 71 Google+ 18 LinkedIn 80 Pin It Share 0 513 Shares ×
About Jon Loomer

Jon Loomer is a digital marketing consultant with a unique perspective on social media. He was introduced to Facebook in 2007 while with the NBA (back before Pages) and has been using Facebook for business ever since. Stay in touch by liking his Facebook Page (Jon Loomer Digital).

  • Eric Tipton

    Hey Jon, great article as usual. I have already passed this on to other members of my team who were bringing up the question about Facebook and Analytics. I have a question of my own though, which is not discussed in your article. If I add a conversion pixel to a custom audience ad and the pixel is placed on a Thank You page how should the conversions be treated?

    • http://jonloomer.com/blog Jon Loomer

      Hey, Eric. I’m not clear what you’re asking? Facebook will report the number of times those in your custom audience viewed and converted within a day or clicked and converted within 28 days (which, as noted above, can be adjusted).

      • Eric Tipton

        My apologies. I am running a custom audience ad for a contest right now. If you leave the page and go to Facebook then you will see an ad much like yours stating why they are seeing it and asking them to like the business page. If a person goes back after seeing my ad and converts does the the conversion count as a real conversion? The reason I’m questioning it is because I’ve reached 5,896 people and the website conversions are at 3,762 currently and the pixel is only on the Thank You page. These may not be unique as you can enter once a day.

        • http://jonloomer.com/blog Jon Loomer

          Something sounds off. I’m not clear what the conversion is though.

          • Eric Tipton

            The conversion is registering to participate in a contest. Where it gets complex is that a person can register once a day to win a daily prize while being entered for the grand prize. This is where I need to know if the conversion pixel tracks uniques or it just simply fires when the form is filled out. Either is an acceptable outcome just as long as I can relay the message.

          • http://jonloomer.com/blog Jon Loomer

            They are tracking every time it’s fired by those who have seen your ad. Just like someone can buy your product multiple times — and it should measure each sale — Facebook is reporting each time they enter.

          • Eric Tipton

            Awesome! Thank you for your time in answering my scenario.

  • Blake Jamieson

    Very helpful post, Jon! Thanks for the clarification! :)

  • Heather Salovin

    Just launched a test campaign with a tracking pixel and will be analyzing the results on Wednesday. I think I may need to listen to this a few more times before I fully understand what the heck just happened.

  • James Deck

    The Assisted Conversion reporting in Google Analytics can be very useful for attributing non-straigh-through conversions. Another tip for conversion tracking is to use Google Tag Manager so you don’t need to modify the code on pages, you can just set up triggers in GTM.

  • http://www.jkrule.com Juliette Rule

    Nice explanation, Jon! Thanks for writing about this!

  • David

    John, I’ve been wanting to change the attribution window for a long time but couldn’t figure out how. Thanks!

    This brings up a question about advertising to existing fans with paid ads. Although I’ve done a lot of paid advertising on FB (using custom audiences), I’ve always been hesitant to do paid ads to existing fans it because the reporting would count as conversions any organic traffic received from the click for up to 28 days. So I figured a large part of the conversions reported would have happened organically anyway. You just gave me a tool to measure this and I will definitely experiment. But what are your thoughts on this?

  • David

    Does FB only attribute a conversion to the last ad that touched the customer? For example, it a customer click on a paid ad that links to your website, but doesn’t buy. And then clicks on a FB remarketing ad (set up with custom audiences) and at that point buys, do both ads get the conversion? Or just the last ad the customer clicks?

    • Olga Kanzychakova

      I had the same question and I’ve got the answer from facebook that the last viewed/clicked ad will get the conversion so that the conversions don’t overlap if the same user sees ads from different campaigns.

  • PhillipaKiripatea

    This is something I’ve been having a problem with.
    I’ve been using LeadPages and MailChimp for my sales funnel and I’ve found that the the conversion tracking in Facebook has been inflated.
    The numbers Facebook is showing me aren’t adding up to LeadPages or to the optins in MailChimp.
    Of course, the page I have the tracking pixel on is the Thank You page, and MailChimp is set up as a single optin so it’s not that people aren’t confirming their subscription.

    It’s been a bit annoying as I can’t accurately track conversion from Facebook so I’ve had to create separate landing pages and Facebook ads to check which ads are working best for conversions.

    • Michelle Pescosolido

      That’s because I believe you need to insert the tracking pixel on the next page after the lead opts in. I use lead pages also. Feel free to correct me if I am wrong, but if you are tracking leads you don’t put it on the landing page but on the thank you page.

      • Frances

        Maybe read the comment again

  • Patrick Alban

    Jon,
    Can you select 2 pixels for a given ad?
    I’d like to get stats on when 1) someone clicks my FB ad to go to my product landing page and 2) someone ends up on my product thank you page. I have a different pixel on each of these pages to capture stats on 1) a view of the product landing page and 2) a product sale.
    If I select both pixels for a given ad, my thought is that I will be capturing the results of this sales funnel. Does it work this way in FB?

    • http://www.networkmarketing-local.com/ Bertie van Greunen

      Patrick I’ve just completed the experiment with one of my clients. I had a pixel for each page, landing and thank you. My client used Lead Pages and there was a big difference between FB & LP so I’m in a bit of doubt if FB calculations was 100%. I trust LP much more as it is their page, but their was also a difference Go To Webinar and Lead Pages on the actual registrations.

      FB showed 152 conversions
      LP showed 196 conversions
      GoToW showed 190 conversions

      So between LP & FB their is a 44 conversion difference. Now I must also mention that I’ve also setup an ad using an event which meant that if you clicked on the ad you would go to the event page and from their clicked on the link that will take you to the landing page which in this case was the registration page.

      I’m not sure if FB will track this. I could not get any indication if that was the case, but 33 people said yes they would go. Now this is no indication if they have registered or not. If i make the assumption that they did register then their is still a difference of 10 to be accounted for.

      This was my experience and would love to hear Jon’s feedback. I would also recommend to test and let me know as I’m very curious what your results would be.

      • Patrick Alban

        Bertie,
        Thanks for the comment. My feeling is that having 2 pixels on one ad will not be necessarily a bad thing. Mostly I wonder if FB will track from one conversion to the next. If they would do so accurately, that would be great. But I have my doubts.

      • Robert Grant

        Big question is whether you are looking at unique hits versus total hits when you are comparing the stats

  • http://www.carissacoles.com/ Carissa Coles

    What if you put a tracking pixel in the landing page and on the thank you page, would that be okay?

    • Chris B

      Hi Carissa,

      If the 2 tracking pixels you use are the same then you will double count any true conversions from visitors since they will hit that pixel on 2 separate pages. You could use a different tracking pixel on the landing page if you want to consider that a conversion on its own, but it’s likely that won’t be very useful because a conversion is best used to measure some sort of action, plus you will already have referral and click through data from your analytics solution and Facebook.

      Hope that helps.

      • http://www.carissacoles.com/ Carissa Coles

        Thanks Chris. It helps :)

  • Alex

    Thanks for the audio version!

  • Chelsea Hejny

    For people using Improvely and a Facebook tracking pixel for their ads, this info might help:

    I create all my Facebook ads using an Improvely tracking link, this way I’m able to track conversions in both Facebook’s Ads Manager and my Improvely Dashboard. After running an ads campaign on Facebook, I noticed that my ad’s Website Conversion Value reported by Facebook was drastically higher than the conversion value I saw in Improvely. When I reached out to Facebook about why my conversion values were not the same, their response was extremely roundabout. When I reached out to Dan over at Improvely, he was able to clarify things for me. Here’s what he said:

    “Facebook is going to attribute 100% of people that clicked your ad then reach your conversion page to their ad. Improvely is only going to attribute those visitors who first found your site through the Facebook ad to that ad as well. The other people that originally found you through some other source, Improvely attributes the conversion to that source instead, according to the conversion attribution settings you set for your project. That’s why you use a tool like Improvely rather than the tracking offered by each individual site you advertise. We know more about your traffic (all their previous visits, and all the ads you’re running) and can give you a more accurate picture of what sources are bringing in new signups or sales for your company, and which are just bringing back repeat visitors.”

    • Dino

      thanks, I will try this

  • Carole@RusticArtistry.com

    I’m curious as to why the number of Website Clicks that Facebook reports for my ad is 3X greater than the number of visitors to my site from Facebook that Google Analytics reports for the same 3 days. Your thoughts?

  • http://www.shop.graciousstore.com/ Gracious Store

    I am listening with interest to discussion on FB advertising because I am seriously looking into it as a marketing platform, I have been very skeptical about using FB as a marketing platform

  • Roger J. Stevens

    Amazing work Jon! Thanks for giving genuine practical advice. Also, do check the blogs at FuturOn, where I work. Id like to draw some inspiration from your writing ! http://thefuturon.com/blog/

  • Thomas

    great stuff jon thanks :)

  • Xabier Izaguirre

    Hey John. A bit unfair on Google Analytics to neglect the existence of Multichannel analysis and the various attribution models available. Great post otherwise,

  • VeloNomad

    Super useful. I was creating a lead pixel and putting on landing page. I was then placing a conversion pixel and placing on thank you page. So, in the power editor, 2 pixels tracking conversion.

    I am using landing pages for lead magnets/mailing list adds and also straight sales pages.

    This helps explain the conversions but no mailing list additions.

  • Tiaan Van Zyl

    This is exactly my problem too. I used google tag manager to implement the facebook pixels through a custom HTML tag on conversion pages. After the first month of converting, we saw that facebook had a lot more conversions than reported on analytics. How come the amount of facebook conversions is more than the amount of pageviews the conversion page received? It is def firing on the right conversion page, as AdWords are showing the correct figures when comparing.

  • Barry Paul Price

    HI Jon. Thanks, but in my case the disparity is between Facebook’s reported number of conversions and the actual number of new subscribers who show up on my list. The pixel is on the Thank You For Registering page after they opt-in for a webinar. The list numbers are inside Infusionsoft, tagged specifically for those who come from the FB ad registration/thank-you page.

    I’ve consistently been noticing FB’s conversion numbers are around 10-15% higher than the number of ‘true’ new users on the list.

    My only theory is that someone may register on one device, then go back and visit the same registration page link from a different device (laptop, iPad, phone) or different browser, thus fooling the cookie-system and firing the pixel again as if another unique user has converted.

    Moving forward, I’ll be calculating my cost-per-conversion strictly based on my FB ad spend divided by my new members on my list(using the Infusionsoft numbers). The conversion pixel will, however, be a nice approximate, fast way to check where I am at in case I want to split test, or adjust my ad spend early in a campaign.

  • Kenny Cullen

    Why would I have 4 conversions (thank you pages) even tho I stopped the Facebook advert?

  • Dino

    Hi, im new to this blog and its helping me ALOT as a student. Im having a similar problem – I placed the tracking pixel on the ‘Thank You’ Page, in the correct place.

    When prospects fill out and submit the data capture form it takes them to this thank you page. The ad is only a few hours old but claims I have 5 conversions…but when Iog into wordpress and see the actual form entries, it is 0.

    This happened to my last ad which had 500 claimed conversions, but it was actually just 395.

    Any ideas why it would be so inaccurate?

    Apologies in advance if im doing something obviously wrong, im still learning..

    • Dino

      I should also point out that you can only see this Thank You page if you click the “submit” button on the data capture, so no one is wandering onto it. If anyone can give me some pointers to potential issues for me to investigate, I would really appreciate it.

513 Shares Facebook 142 Twitter 202 Buffer 71 Google+ 18 LinkedIn 80 Pin It Share 0 513 Shares ×