Pixel Based Tracking versus 1st Party Data Attribution

December 20, 2019

The more sources of marketing attribution and tracking data you use, the more confused you are likely to become, but why is that?

As online marketers increase ad spend, marketing channels, and complexity of campaigns, the need for accurate marketing performance becomes critical.  The increased ad spend adds more pressure to perform. 

The dreams of scaling your ad spend in lockstep with your business growth creates dreams of freedom to the business owner/client, and when those dreams suddenly are dashed, there is a need for answers.  Why isn’t the marketing scaling with the ad spend?

The quest for answers leads people to look more deeply into their Facebook tracking, Google Ad tracking, and Google Analytics reports.

The numbers don’t usually add up.  The problem is that you are looking for answers in data that is not designed to answer your question.

Here is a recent email from a Wicked Reports prospect that prompted this blog post:

“We’re seeing Revenue of 156K total in our profit and loss statement – that is without tax.

FB PPC is reporting 94K rev and google PPC is reporting 141K rev (total incl tax 235k) take the tax off and that leaves: 188k.

Analytics is reporting Total revenue of $227k (including tax).

I’m getting insanely confused as no numbers stack up, I know you don’t know the business, but we’ve increased marketing spend, generated good ROAS from FB, but it doesn’t seem to show anywhere?”

First of all, hats off to the person behind this email. They are taking the time to do the work to dig into why their numbers don’t add up. The fruits of that labor are plentiful:

  1. It is possible to reconcile your sales back to marketing dollar spend and activities.
  2. This reconciliation allows you to scale marketing that works and stop marketing that doesn’t.
  3. Marketing campaigns are more likely to be most effective at the top, middle, OR bottom of your funnel, NOT all 3 funnel locations.
  4. When you know where in the funnel a marketing campaign works, you can spend more on that audience in that campaign at that point in your funnel.
  5. When you know where in the funnel a marketing campaign does not work, you can stop spending on that audience in that campaign at that point in your funnel.

Going back to the email above, how come the profit & loss statement does not match the combination of Google Ad PPC and Facebook Ad PPC pixel tracking results?  And why is Google Analytics so far off?

The answer lies in the pixel-based tracking’s limitations and/or bad setup.  Here are the top problems and limitations we see at Wicked Reports:

  1. Pixels don’t account for your most valuable sales – subscription rebills
  2. Pixels sometimes don’t fire
  3. Pixels sometimes fire 2 or 3 times (some specific situations in this list) for unknown reasons
  4. Data used by a pixel can be blocked by Safari cookie rules, Mozilla browsers, and other ad blocking tech
  5. You are driving traffic from all marketing channels to the same conversion page, and multiple marketing channel pixels are firing at the same time
  6. Your marketing tech stack uses multiple domains and iframes and the pixel + cookie info is not being relayed between all of them
  7. There is a 3rd party add-on in your WordPress or Shopify site that some well-meaning person installed that is firing pixel events
  8. You have a click listener on a page so it overcounts because most click listeners do
  9. Your page reloads for some other reason which causes the pixels to again fire

Some or all of these 9 top pixel risks can lead to your Facebook Ad, Google Ad, and Google Analytics conversion revenue being different than your actual business revenue.

Let’s suppose you have your pixels correctly firing once on each page.  Will your numbers add up?

Not if you are a multi-channel marketer using email, Facebook PPC, and Google Ads, plus with some organic SEO traffic added to the marketing mix.

Both Facebook and Google ads will sometimes claim credit for a sale, when the actual credit belongs to email, organic, or the other ad platform!

Let’s look at this scenario that could apply to us at Wicked Reports:

  1. Day 1 – Prospect searches “marketing attribution” and clicks on an ad from our Google Campaign “Marketing Attribution Search”.  Intrigued, they opting to our email list.
  2. Days 2 to 14 – we send emails, retarget on Google and Facebook, beg, plead, but the Prospect ignores us (this happens every once in a while, sadly).
  3. Day 15 – Prospect sees an ad from our Facebook retargeting campaign “On our email list get them to watch a demo”, clicks, opts back in for demo of the Wicked Reports platform
  4. Days 16 to 25 – Our sales rep, emails, and retarget on Google and Facebook alternatively ask, beg, plead, etc., but no sale.
  5. Day 26 – Prospect can’t make any sense of their attribution and conversion data (such as the email above), and contacts sales rep via an individual email response, asking a few pointed questions.
  6. Day 27 – Sales rep sends emails with answers and link to the order form.  Prospect clicks on properly tracked email from our CRM campaign “Hot lead send order form link” and makes the purchase (a wise decision).
  7. The purchase page fires the Facebook, Google, and CRM conversion pixels.

This one sale can show up as follows:

  1. Google ad campaign “Marketing Attribution Search” shows 1 sale.  This is because Google tracked a click within 28 days, and then the Google conversion pixel fired indicating a sale.
  2. Facebook ad campaign “On our email list get them to watch a demo” shows 1 sale.  This is because Facebook tracked a click within 28 days, and then the Facebook conversion pixel fired indicating a sale.
  3. CRM email campaign “Hot lead send order form link” shows 1 sale. This is because the CRM tracked a click to its order form and then the CRM conversion pixel fired indicating a sale.

In this one simple scenario, (and yes, this only has a few clicks and campaigns to look at, generally there is a lot more going on!), if you are trying to reconcile your Facebook, Google Ad, and CRM sales tracking, you will see 3 sales, when you actually only had 1 sale.

This can lead you down a rabbit hole of Google Tag Manager sophisticated installations (could work if you get a top dollar expert who installs it flawlessly, and you pay them to maintain it), attribution click time windows, what the hell to do with view-through conversions vs clicks, and the agony of still never having your tracked sales equal your actual sales.

Wicked Reports solves all these challenges using 1st party data tied to real business outcomes with true ROI.

First, we start with your 1st party data within your CRM and shopping cart.  1st party data is the ONLY way to know a lead or sale is real.

All of our patents and algorithms are based on the premise that we need to start with real, verifiable data first. Otherwise, you are guessing or acting on data that is only semi-accurate. That’s like building a house where the foundation is 60% solid – how can you sleep at night without worrying the house is going to implode on you?

2nd, we match attribution models to specific marketing goals.  You need to measure marketing results based on the intent of the marketing.  I can’t stress this enough.  What good is looking at some complex mathematician’s attribution modeling mix if you can’t understand it, and worse, if the attribution model doesn’t understand you or what you are trying to achieve?

If you are doing cold traffic lead gen, your goal is to find leads that become high value customers over time.  We have a specific marketing attribution model designed precisely for this situation. When your marketing finds a new lead, and that new lead ends up buying over and over again, either with one-time sales or subscription re-bills, that is literally GOLD for your business. 

3rd, all of our attribution data is verifiable.  You shouldn’t trust a black box with your job security!  We show the data behind the ROI numbers of your marketing campaigns.  Since we start from real sales and leads, it makes it a lot easier to give you data you can verify.

4th, we filter out the noise and give you only the strongest signals. Sure, it’s an interesting read over at thinkwithgoogle.com about how different ecommerce shoppers have 142 touch points before they buy.  What the hell are you going to do with 142 touchpoints of data when you can’t even reconcile your shopping cart sales against your ad platform tracking and ad spend?

Wicked’s premise is that a sale is the strongest signal, a lead email submission is the next strongest signal, and a click is the 3rd strongest signal.  We only report ROI, revenue, leads, and clicks against attribution models tied to verifiable 1st party data.

This approach limits the time to understanding and acting on the data.  It also takes into account the reality that you are already a busy marketing professional who does not likely have 30 additional hours a week to try to comprehend why we took your $100 sale and spread the credit among 142 touchpoints.

5th, we partner with Google and Facebook to safely leverage your 1st party data so that their machine learning algorithms give you the best chance to win the digital marketing game.

Wicked Reports is a Premier Platform API partner with Google and an Ad Tech and Offline Conversion partner with Facebook.

There are massive revenue gains and time savings with our Google conversion manager.  Wicked Reports worked with Google for 9 months to develop the 1st Google conversion import manager at scale.  We use a machine learning optimized attribution algorithm to help Google advertisers see their real leads, sales, and attribution value inside of Google Ad manager. 

Further, this data is sent to Google in a way that optimizes campaign ROI for sales funnels where leads take time to buy. 

Our Facebook offline conversion import allows advertisers to push sales into Facebook if desired with a one-time authorization.  Wicked can then use Facebook view attribution when it makes sense to do so.

These integration partnerships with Google and Facebook are unique to Wicked Reports and not available with any other attribution platform.

These 5 unique advantages of Wicked Reports exist to give our customers an edge over their competition.

Bringing this full circle back to the original email, Wicked Reports would start from the shopping cart revenue.  Let’s assume this was the basis of the profit and loss statement. 

  • We will report $156k for the month for attribution models that are designed to reconcile against your actual shopping cart sales.
  • We then match this up against new leads generated, existing lead reactivated, clicks tracked, and ad spend for the month.
  • The data is aggregated, and the revenue attributed based on different marketing attribution models.
  • These models are aligned with common marketing strategies (new lead generation, existing lead reactivation, sales, awareness, etc.).
  • ROI and attributed revenue are computed based on the attribution model selected.
  • In some attribution models, only one campaign will get credit.
  • All leads and sales attributed to a campaign can be verified with one-click.

We won’t be able to tell you which orders led to the over-counting in the other platforms.  They are great platforms for advertising, but they are not built for multi-channel attribution that is verifiable by 1st party data.