In this article, our goal is to illuminate the technical details involved in tracking users as they move through your lead funnel. We'll see where data is collected, where it is lost, and where to update your technology to maximize results. After reading this article, you should have a clear understanding of what's possible for your digital marketing attribution.
Let's start by defining our funnel.
A Typical Lead Generation Funnel
Our focus is on a typical lead generation funnel for a B2C or B2B service business that has the following stages:
- Prospect is shown an ad for your service (either via PPC or social).
- Prospect clicks the ad and is directed to your landing page or website.
- Prospect fills out a lead form or calls your business directly. Prospect is now a lead.
- You make the sale.
- Lead becomes a customer and pays you.
We're focusing on this type of funnel because it's a very common one for service businesses. However, the principles in this article are extendable to other types of marketing funnels and strategies.
Now that we have a funnel to track, let's look at how we measure its performance.
Data Collection
Critically, data is collected at various points in the funnel with the goal of monitoring and optimizing ad performance. Ideally, optimization should happen both by marketers and by ad platforms.
It is this data feedback loop that makes digital marketing so powerful. It lets marketers make changes based on the broader context we have around our industry and customers, and lets the ad platforms make changes based on what they know about their users' behaviors and interests.
Below we use our funnel from above and show the standard conversion events (circles) that are used to measure the performance of each funnel stage. Conversion events are recorded whenever a user takes an action in the funnel. Each conversion event is recorded by a single data source, represented by the surrounding boxes.
Purchase conversion events contain a dollar amount ("value") associated with each event. This is standard and reflects that we usually want to optimize for total revenue or profit, not total purchases.
First-party Data
A common theme below will be our emphasis on first-party data. First-party data is data that you collect and control. It provides accurate, reliable, and privacy-compliant insights into your customers behavior. It's also flexible, and can be integrated across data sources to provide a complete picture of your funnel performance. These integrated results can be used by ad platforms to optimize your campaigns, and by your team to make better marketing decisions.
Next, we'll review the collection details of each data source. We're going to go out of order from our funnel and discuss the straight forward data sources first. Then we'll dive into your website, which requires a deeper discussion.
Data Source: Ad Platforms
Ad platforms collect engagement information directly on the platform. They monitor how many times your ad is viewed (impressions) and how many times a user engages with it (typically: clicks). This top of funnel data is used to automatically to optimize who within your audience is shown your ad - with the goal of only showing your ad to users who are likely to engage. Ad platforms also report aggregate conversion data back to you, so you can monitor engagement.
Data Source: Call Tracking Service
Call Tracking Services track inbound phone calls generated by ads. For most service businesses, this represents a sizable chunk of their inbound business, and so visibility into this channel is vital.
Call Tracking Services work by registering new phone numbers and assigning each to an ad, ad group, campaign or ad platform. These tracked phone numbers are only displayed in ads, and so calls that are made to the phone number can be assumed to be from prospects who viewed an ad.
When a user calls one of these numbers, the call is forwarded to the businesses' main line. The Call Tracking Service also records information about the call, including the calling number and time of call. Typically, this information is first-party data.
Data Source: CRM / Accounting Software
There are a variety of CRM and Accounting Systems. They have a wide range of features tailored to a wide range of industries. However they all share one common characteristic critical to your marketing success, they record the first-party data that matters most: who your customers are, and how much they spend. This is the most important information you own: it gives ad platforms the best signal possible to optimize your ads and let's you calculate your true marketing ROI.
Data Source: Websites
Your website offers the broadest range of options, and challenges, for data collection. For our funnel, we'd like to collect page views and lead form submissions. To be useful, this data needs to be matchable back to a user on the ad platform.
We'll start by looking at how websites are able to match conversions to users on ad platforms. We'll then look at the options for where we collect data, and what consequences that has to ad platform optimizations and our visibility. Finally, we'll review landing pages, their deployment options, and how they can be used to ameliorate attribution data loss.
Matching Conversions to Users
In addition to "what" happened, conversions must include information about "who". This is called “match data”. There are two types of match data: click IDs and personal identifying information.
Click IDs
When a user clicks on your ad, they are redirected to your website. Each user is shown a unique link.
All the links point to the same page, but they have unique tracking information embedded. These are called “Click IDs”. They let the ad platform link downstream conversions to the upstream ad click that generated them.
For example, here is what a Meta link w/ a click ID might look like:
https://www.example.com/some-landing-page?fbclid=uFCrBkUgEvKgoaoAOALkxnksaoiaFALKhqwiecn The portion of the URL after the ? is called a query string. It does not change
the page that is displayed (here the user will see the content for /some-landing-page), but
provides information to the website to be used later. In this case, it tells the website that
the parameter fbclid (Google's equivalent is gclid) is equal to
uFCrBkUgEvKgoaoAOALkxnksaoiaFALKhqwiecn, which is a unique
value that identifies the user's click.
The website then sends this value back with any conversions it sends to the ad platform, letting the ad platform match the conversion back to a user.
Click IDs are the best way to match off-platform conversions with ad platform users. However, Click IDs uniquely identify user behavior and, accordingly, are a major target of privacy regulations and software.
Cookies
If a user visits multiple pages on your website, any click IDs must be remembered, so that conversions on other pages can be properly attributed. Typically, trackers use cookies for this purpose (Google calls this a “Conversion Linker”). An example of where cookies are needed would be if a user lands on your home page, but must navigate to the "contact us" page to fill out a lead form. The website asks the browser to store the click ID in a cookie, which is later read when the user submits the form.
Problems with Cookies
There are a couple reasons why cookies may not be available to track users across your site:
- Depending on the laws governing your website, you may be required to get user consent before before saving any such tracking information.
- Browsers let users disable cookies.
Apple's Link Tracking Protection (LTP)
Apple has also made a big push for privacy in recent years. One of the steps they've taken to increase user privacy is to strip URL query parameters that are uniquely identifying, like fbclid and gclid - this is called "link cleaning". Apple's not alone on this front. Privacy first browsers like Brave are also becoming more prevalent, and overall, it's expected that this trend will continue.
Personal Identifying Information
The other way platforms can match conversions to users is via personal identifying information (PII), e.g. email, phone, first name, last name, address, etc. Google refers to these as “enhanced” conversions.
PII does not work as well at matching users as Click IDs, for a couple reasons:
- Users may not enter any identifying information. E.g. if they just visit your website, but don't convert further.
- Users may enter different identifying information across sites. E.g. they might use their work email on your lead form, but use their personal email on Facebook.
Additionally, depending on your location and your audience's location, privacy regulations may prohibit the sharing of PII without consent.
Now that we've examined how ad platforms match users to actions on your website, let's look at the architecture of your website, and the implications that has on your tracking.
Collection Location
Websites are broken up into two key technical components: the front end (“client side”) and the back end (“server side”). Let's review those components and how data can be collected in each location.
Client Side
The client side of the a website is the part that runs in a visitor's browser. It is responsible for displaying all the text and graphics you can see. It can also run code, Javascript, which enables it to display animations, allows for user interaction, and tracks user behaviors.
Ad platforms provide their own user tracker code, which you've probably heard of: Google has “tags” while Meta has “pixels”. Trackers are relatively easy to install, and simply require adding a couple lines of copy & paste code to your site, which loads the tracker.
Both tags and pixels perform the same task: track user behavior on a website. Both also suffer from the same weaknesses.
Ad Blockers
Since client side code runs on a user's machine, the user has full control over what is, or isn't, run. Ad blockers are an example of this control. Ad blockers are built to detect code, like Google's and Meta's trackers, and disable them from running. It is estimated that ad blockers are installed on around 1/3 of all browsers, which means that around 1/3 of all web interactions cannot be tracked by client side trackers.
Third Party Data
Google's and Meta's trackers send data directly back to Google and Meta - you don't get full access. Both ad platforms provide reporting dashboards that show your website's aggregate results, but you aren't able to see the exact, informative, details. You're also not able to combine tracker data with your own data, like purchase information, to see a complete marketing picture. It's not your data, it's a third-party's.
Server Side
The server side of a website is the part of the website that is “serving” the content on a “server” (hence the name). Servers run 24/7/365 just waiting for a client to make a request. When a client requests a webpage, the server responds with the webpage's content: text, graphics, and code to run. When a client fills out a lead form, they also send a request to the server to process the form information.
Unaffected by Ad Blockers
Since your server is controlled by you, it is not affected by client side ad blockers. This gives your server better visibility into your website's conversions. Monitoring conversions on the server is called “server side tracking”.
Generates First Party Data
And since you own your server, you control the data it collects. I.e. you are collecting first-party data. Like with third-party trackers, this data can be forwarded to ad platforms so they can optimize your ad delivery via an API.
Technical Challenges
Though server side tracking is more reliable than client side tracking, it is also more complicated to set up and to integrate correctly, which is outside the scope of this article. There are variety of server side tracking options, but they all have higher technical requirements than copy and pasting a tag or pixel.
Landing Pages
Landing pages are a special type of web page designed to be the target of an ad click - they're where the prospect "lands". In this section we're not going to review how to design a good landing page, but rather how to deploy them to maximize your data collection. There are two ways to deploy landing pages: as a regular page on your website, or as a standalone website via a subdomain.
Landing pages suffer the same tracking weaknesses as the rest of your website. They won't help ad platforms track your conversions any better than your website does. However, they can help you track your conversions better.
By structuring your landing pages correctly, you can collect valuable first-party attribution data about your leads - up to which ad each came from. Let's look at what a landing page needs to be useful for tracking, and how you can deploy them to maximize your data collection.
Requirements for Landing Pages
To be useful for tracking, landing pages must:
- Display a unique, tracked telephone number for leads to call, so we can track which calls came from the page.
- Uniquely identify lead form submissions with some sort of meta data, that identifies the form's origin as the page.
- Have a 1-to-1 relationship with some ad source - i.e. an ad, ad group, ad campaign, or ad platform. In other words: only one ad source should should link to any one landing page, so that we know all visitors to the page came from that source.
- Not be linked to from your main website - we want to isolate the traffic to the page so we know all conversions came from the ad source.
- Be configured as "noindex, nofollow" so Google won't index them - we don't want organic traffic to muddle our results
By meeting these requirements, you can ensure that your landing pages are collecting the data you need to best attribute each conversion. Now, let's look at our landing page deployment options.
Internal Landing Pages
Internal landing pages are pages that are part of your website. They are typically created by your web team, and are hosted on your website's server. They are easy to create and deploy, and can be updated quickly. They are also easy to track, as they are part of your website's normal tracking setup.
However, for complete tracking, internal landing pages require cookies, which may not be available for the reasons previously discussed.
Cookie Dependence
Impact on Form TrackingConsider the example below, where cookies are not available. We have two internal landing pages configured. One for Meta ads, one for Google. We also have the normal organic traffic that hits our home page. All of these pages link to our contact us page, where a user converts. Unfortunately, without cookies we can't tell which page the user came from, so we can't attribute the conversion back to the source.
Additionally, we want to ensure that if a user decides to call us, they call the tracked phone number associated with the ad source. Obviously, we can put that tracked number on the landing page itself, but we also need to handle the case illustrated above: a user navigates elsewhere on our site and then calls. If they head to our home or "contact us" page, and then call the main business line listed there, we won't be able to attribute the call back to the landing page and the ad source.
Ideally, we'd like to show a different tracked number to each user, depending on the landing page they first visited. That way, no matter where they call from, we'll be able to attribute the call back to the ad source. This is an advanced feature, and requires advanced web development skills to implement.
External Landing Pages
The other option is to deploy landing pages as external, standalone websites. These are assigned a special subdomain, e.g. "landingpage.example.com" and are hosted on a separate server.
Setting up an external landing page is more involved and requires more expertise than an internal landing page:
- A new server must be provisioned.
- Brand assets and themes must be loaded on the new server.
- IT must map a subdomain to the new server.
- Tracking must be set up (correctly) on the new server.
- Lead form submissions must be forwarded to your CRM, email, etc.
However, external landing pages have a major benefit - their traffic is isolated. Unless you explicitly link to your main website, users will not have a direct way to navigate away from the landing page. This makes them ideal from an attribution standpoint. Any conversions that happen on the landing page can be attributed back to the ad source, even without cookies.
It's important to note that if cookies are available, external landing pages may not be the best option. For one, cookies cannot be shared between your main website and the external landing page. This means that if a user navigates from the landing page to your main website, you will lose the ability to attribute the conversion back to the ad source. However, if cookies are not available, external landing pages offer a nice way to attribute leads back to an ad source.
Websites - Summary
Websites are the most important part of your marketing machine. They are where the majority of your data is collected, and where the majority of your conversions happen. They are also the most complex component, and the most likely to have tracking issues. By making the right technical choices, you can ensure that your website is collecting the data you need to accurately attribute your conversions, and maximize your marketing ROI.
Regardless of your situation, server side tracking offers you the best visibility into your marketing results. It gives you first-party data, that you can integrate with your CRM and other marketing data to get a complete revenue attribution picture.
Then, depending on your situation, you can deploy internal or external landing pages as ad destinations (being sure to adhere to the tracking requirements). If cookies are an option, and have the development skills to show tracked phone numbers dynamically, internal landing pages will give you the best visibility, particularly for longer sales cycles where a prospect may come back to your site multiple times before converting. Otherwise, external landing pages offer a way to attribute leads back to an ad source without cookies or advanced web development skills, though they have a more involved setup.
Conclusion
We've reviewed the technical details around tracking users as they move through your lead funnel. We've seen how data is collected at various points in the funnel, and importantly how first-party data can be collected and why it is important.
We've seen to optimally track our funnel, we need the following three data collecting components, each generating first-party data:
- Call Tracking Service to track inbound phone call based leads.
- Landing pages, with server side tracking, each dedicated to a specific ad platform or campaign (however granular you want to be).
- CRM / Accounting Software with purchase information.
We can integrate this first-party lead data across our marketing stack, and get a complete picture of our marketing ROI.