Chances are, you’ve heard the word pixel thrown around when it comes to monitoring advertising performance, but what exactly is a pixel and how does the tracking process work? It’s important to understand these essentials to pixel strategy when you’re thinking about campaign planning, targeting, and optimization. To help you on your way, we’ve broken down the basics of pixel tracking for you below.

What Is a Pixel? 

Pixels are a piece of code that is placed within a website’s code that allows the demand side platform (DSP) to collect information about actions that are taken on that website. Essentially this means that the pixel allows the DSP to understand what’s happening on the site so that we can segment our audiences, track conversions, and really understand more about that conversion lifecycle for your customers. The information that’s included in that code is going to tell the DSP how it should categorize the users that have taken the action that triggers the pixel. 

Here is an example of a pixel:  

In the above example, the green section tells you which DSP you are sending the information to. In blue, you have the advertiser ID, which tells the DSP that this is the advertiser that you need to tie this pixel to so they know which client or campaign or group this user is attached to. The pink is the Pixel ID, so the DSP knows which pixel fired. The pixel is the actual event that we’re looking to track, whether that’s a page view, or a conversion, or any kind of other event that we’ll define shortly. 

How Does a Pixel Work?  

The basic pixel process functions kind of like a string of dominoes. 

It starts with the trigger, which is the event that we’re looking to track. This trigger can be a page view, a button click, or whatever we’re looking to keep track of. Once that action has been taken, it triggers the pixel code, which triggers the network call, and the network call passes that information to the DSP in that process.

The network call is really like a digital phone call. Think of it as the marketer’s website picking up the phone, calling The Trade Desk (DSP) and saying, “Hey TTD, just so you know, Person X looked at that landing page that we’re looking to track so add a cookie on her so that we can track her and keep that information tied to the campaign.”

If we had a parameterized conversion pixel on the site, we would also say, “Oh by the way, Person X made an e-commerce purchase, and her purchase was $29.”

Then the DSP classifies all that information in its system which allows us to retarget and to report on all of that activity. This helps the marketer identify which of their marketing tactics helped influence the purchase decision. 

What Can You Track with Pixels? 

Pixels are used to identify groups of users based on actions that they take on a website. Those actions could really be anything, as long as it’s a unique event on the site. That could be a site visit, or a purchase, a form fill, viewing a specific landing page, etc. Any kind of action that you want to track to understand how people are interacting with your brand and with your website can be something you pixel. 

Can Pixels be Used Outside of Retargeting Tactics?

Yes, pixels are what we use to track conversion events too. So if you have a lead form that you’re looking to track, or a purchase, we would place a pixel and then the DSP would report on the number of conversions that take place that are tied to the campaign. That’s the other main function of pixels. 

We can also use them to create lookalike audiences in certain platforms because that pixel essentially is just storing a group of people and information about them.

How many actions can you track in one pixel?

One pixel tracks one action. So you would need to place a separate pixel on any individual action that you plan to track.

What is a Parameterized Pixel? 

Parameterized Pixels offer the ability to include additional information with the pixel. 

There are dozens and dozens and dozens of different parameters that we could include but the most common are tied to the e-commerce space and include: 

  • Revenue order ID
  • Product SKU 
  • Order quantity 

Here is an example of a parameterized pixel: 

In addition to the network call telling the DSP that a conversion happened, the parameters included will also tell the DSP that a purchase happened, and that purchase was for $20.99,  and the Order ID was A81DHF5.

The DSP captures all of this information and will understand it as revenue, which is how they’ll calculate return on ad spend (ROAS).  

Does having a pixel risk slowing the load time of a webpage? What about having multiple pixels?

The short answer is no. All of this happens in a nanosecond. If you had 37 pixels on a page, it might slow the page down slightly. But since a pixel is only triggered when a certain event  happens, they probably don’t all load on the same page or on the same action. If you had 37 network calls happening at once, it may slow your page down, but only by a few seconds. 

What Should I Pixel for a New Campaign?

 When you’re thinking about planning for a new campaign, and what to pixel, the short answer is to pixel everything! 

Any action that might tell you more about your consumer, your conversion life cycle, or your product relevance should be tracked. You don’t necessarily have to use it, but once the pixel is placed, that data is there for you to analyze. This gives you the opportunity to be strategic and try new things consistently throughout a campaign.

Some essential things to track are: 

  • Key landing pages: If you have a ‘learn more’ page, it’s really crucial to your conversion lifecycle, so you want to make sure to add a pixel to that landing page. 
  • Home page: Include a pixel here as an upper funnel event, to understand who is coming to your website.
  • View product: A pixel here will help to determine who is looking at certain products but not purchasing. Retargeting might help them finish the purchase later on. 
  • Add to cart: This is a really high intent audience, so you just need to nudge them a bit further down the funnel. 
  • Lead forms: Adding a pixel to the start button of the lead form and also to the confirmation page will help you to not only understand how many final form submissions you’re receiving, but also what kind of upper funnel interaction you’re seeing from the campaign or from different tactics. 

In some of your prospecting tactics you might see that you’re not getting many conversions because this is a first touchpoint for these users. If you have additional pixels placed, you might see that your third-party audience is getting a 25% add-to-cart rate, which is fairly high. In that case, you know that your prospecting has created strong upper funnel awareness, even if it’s not delivering purchases. You can use that audience to feed into retargeting groups, in order to bring them into the funnel and continue to engage with them. Having information like this about your consumers is really helpful when you’re thinking about how to optimize at all stages of the conversion life cycle.

How Do I Know Where to Start Pixeling? 

You want to place a pixel on every milestone within your conversion process, or anything that you think could tell you something different about your audience. If you are an e-commerce retailer and you have both a men’s and a women’s purchase category, you might want to pixel those two sections separately in order to determine which people take action on ads for male vs female clothing. 

If you’re not sure where to begin, it’s a good idea to determine the ultimate goal of your campaign first which could be e-commerce purchases, lead form submissions, etc. Whatever that goal looks like, work your way backwards. What steps does a consumer need to take in order to reach that final conversion action that you were looking to push them to?

If you’re still unsure where to start, take a look at the data that you have in Google Analytics or in your CRM system see which landing pages have the highest conversion rate.  

Any goals you have configured in Google Analytics should also be mirrored with pixels so that you’re giving the campaign the same information that you’re reviewing daily when you asses your brand and your site functionality.

As an example, let’s look at Madewell, a large e-commerce retailer. 

If we were running a campaign for them, we would start by pixeling the following: 

  • Site visits
  • Content categories  (denim, clothing, shoes, accessories, dresses, etc)
  • Products pages
  • Add products to cart 
  • Initiate checkout payment
  • Complete purchase
  • Email signups 

If you’re trying to figure out where exactly you want to benchmark your pixels, go through the process and think about which events will reveal how interested a user is in your  goods and services.

Are there pixel strategies outside of web-based uses, meaning can pixels be used in other media outside of the web? How well do pixels work on mobile in-app versus mobile web or desktop?

Pixels are based within the browser, so you can’t track anything that happens offline. For example, if you were trying to track a phone call, you could track the button click of the phone number that’s on the website, but you couldn’t track if someone decided to actually pick up their phone and type in the number manually. There’s no code trigger that would tell the pixel that that event happened. You can place pixels on mobile apps, but they are placed differently. You typically need what’s called an SDK, or a software developer kit in order to place a pixel in a mobile app. 

How do pixels from various media partners such as a DSP relate to Google tag manager?

Google tag manager, as well as Tealium are known as container tags or tag management solutions. When you are working with one of these, you place one piece of code across the entire site in the universal header. This allows you to bucket all of the pixels together in the container. Container tags are a lot less work for developers. It’s a lot of work for them to place 75 pixels individually across your site, but if you utilize a tag management solution, the pixels are placed one time and the tag management platforms allow you to set triggers for all of them. 

Using Google Tag Manager, you can set a pixel to fire on the homepage and once that pixel is configured you can use the same logic for all other placements. If you were placing DSP pixels, Facebook and Adwords, but you’re tracking all the same events, you can place those all together. Tag management systems and container tags are easier than placing individual pixels on a site. Container tags always seem to simplify things for advertisers who have a lot to manage anyway, especially when it involves the development team.

From an agency perspective, we can go back and forth with clients for weeks in order to place pixels correctly. If they just place a container tag and give us access to it, we’re able to see everything that’s happening and troubleshoot more effectively. It takes the extra work off of the agency because container tags are more like a one-stop solution. Once they’re implemented correctly, there’s less risk of pixels accidentally being removed when they make creative updates to the site.

When would you recommend tagging a page load versus a button click?

If you’re looking to track the “add to cart” action, you should track the button click rather than the page load. There might be some degree of drop off in people who are adding to cart versus people who are actually being redirected to that next page, so you want to track the event that is closest to the action that you’re looking to understand. 

A button click and a page view are different actions. A button click will typically lead to a redirect, so if that’s what you’re tracking, it will give a more accurate understanding of the number of events that have occurred. If there isn’t a button click, a page view is fine.

How many pixels would you typically recommend to an e-commerce client to place?

The number of pixels that are placed, depends on the size of the site. For example, if your site doesn’t have much product differentiation and only has 100 transactions per month, you don’t need to place 35 pixels. If you’re a huge ecommerce site, it would make sense to have pixels placed on every key header. From there, you may want to pixel several different actions on each page. 

A good rule of thumb is that any key milestone or any action that the user has to take to purchase something should be pixeled.

Outside of Google Analytics, can you set up a pixel to time on site?

With container tags, such as Google Tag Manager (GTM) or Tealium, you can set up certain types of logic that will have the pixel fire on a time delay. This would require a highly skilled developer for implementation. 

What is the Process for Creating Lookalike Audiences with Pixels? 

The process for building lookalike audiences depends on which site you’re on. Using Facebook for example, you would build your lookalike audience directly on the social platform. Any event that’s pixeled in Facebook can have a lookalike audience built based on that event. 

For the DSPs, it depends on the particular platform. In some instances, we would have to work with a data onboarding partner to help onboard segments that would create the lookalike. In other instances, you can create the audience within the platform directly. Regardless of whether you are inside or outside of the platform, the system (either the DSP or the additional partner) uses the information about that audience and a level of machine learning to build out the lookalikes. 

If every customer who purchased a Madewell t-shirt online also loves chocolate milk, there’s no way that we as humans could ever figure that out. But the DSP can figure out that information from the pixel and cookie data for those users. This would allow them to say that since we know that the users who purchased Madewell shirts also like chocolate milk, we can find people who may not have purchased from Madewell but do like chocolate milk in order to expand our audience. The DSP will use various systems of logic to identify who is the most likely to be interested in your product, based on your previous purchases. 

5 Key Takeaways About Pixels

  1.  Pixels allow advertisers to track key events on their websites and segment their audience based on the actions that each user takes on the website.
  2. The process for a pixel to fire functions like a chain of dominoes, so that when one event occurs, it triggers a series of events that will pass details about that initial event back to the DSP.
  3. Aim to pixel as many milestone events in the customer journey as possible. If you’re ever unsure whether or not you should pixel it, do it. 
  4. Place pixels on pages/events that signify differences in consumers in order to segment an audience and create custom messaging.
  5. The sooner pixels can be placed, the better – the more data the merrier!

Ready to Get Started with Pixel Tracking?

AUDIENCEX runs campaigns across every vertical and channel, and we are pixel experts. Contact us to talk to one of our campaign strategists about how we can help you to set up, plan, execute, or optimize your campaigns for you.