ROI is one of the most important metrics in marketing. No matter which marketing strategy you choose to implement, you expect it to bring you a return on investment. 

But when it comes to content marketing, it isn’t that simple. The ROI of content marketing is harder to measure than with other marketing strategies. 

With ads, attribution is direct. You can see how many people bought the product out of all those who saw or clicked on the ad. But it’s not the same with content. You can’t really know who read what or which piece of content pushed the customer over the fence. 

Content marketing isn’t just about lead generation and conversions that can be measured in a few clicks - it’s an ecosystem. Each piece of content works in unison with other content pieces across all platforms, creating an interconnected system that grows over time. 

It’s very rare for someone to convert after reading just one blog article or LinkedIn post from you. They’ll only convert after being exposed to your content ecosystem for a while. 

And that’s why attribution can be pretty tricky when it comes to content marketing. 

I’m here to show you how to do it the right way. Or, at least, how to get the most out of the data you’ve got. 

How Does Content Marketing Attribution Usually Work?

A brief definition before we go on: content attribution is the act of attributing specific conversions to content marketing. 

In other words, content attribution is the process of determining how much credit a particular piece of content should get for driving a particular conversion or sale.

When it comes to content marketing attribution, it’s hard to know where the conversion really comes from, as the reader usually goes through various content touchpoints before converting. 

And the last piece they read (the page they convert on) may not be what actually got them to convert!

You can use a couple of attribution models to measure that. Let’s go over them. 

Single-touch attribution 

Single-touch attribution is the easiest way to measure a conversion. It’s easy to set up because you’re only looking at one specific interaction (say, a click on an ad). 

For example, single-touch attribution may mean measuring the number of people who made a purchase out of all the people who looked at comparison pages. 

But if those people come back 30 days later, we wouldn’t be able to attribute conversions to content, since you’d only be looking at single-touch.

It’ll be measured as a separate conversion rather than the continuation of the customer journey that led them there. 

This is because the single-touch attribution model credits the conversion to a single touchpoint rather than distributing the credit across multiple touchpoints in the customer journey. 

This model isn’t very accurate because it doesn’t tell you what happens beyond this single interaction. It’s a good start, but it’s best to move to a multi-touch attribution model as soon as possible. 

Multi-touch attribution 

Multi-touch attribution is a more complex method of attributing results to content marketing. 

Rather than giving full credit to a single touchpoint, multi-touch attribution looks at all the interactions a prospect had with a business across multiple channels and throughout a specific time period. 

For example, a prospect may find a business on Google, see a couple of ads on LinkedIn, wait 30 days, run another Google search, and finally make a purchase.

This approach is indeed much more complex, and requires more complex modeling, but it’s much closer to what happens in reality.

Attribution is a rigged game

It’s nearly impossible to accurately measure which piece of content had the most impact on the customer’s decision to convert. 

For instance, your blog content may be discovered through various channels - such as search, Google Discover, social media, newsletters, Slack, or email - and keeping track of them all is almost impossible. 

If you try to approach content marketing attribution from a purely analytical point of view, you’ll lose hours cherry-picking data that won’t give you 100% precise results. By definition, content marketing helps with conversions by building trust, authority, and loyalty - and these things can’t really be measured. 

Plus, web analytics aren’t 100% trustworthy, either. Just think of all the tracker-blocking software, web browsers that block trackers by default, deleted cookies, and privacy policies. 

So, when measuring content marketing attribution, you have to kiss perfection goodbye and focus on looking for trends and patterns rather than mathematically precise results.

Just keep that in mind 🙂  

Content Attribution 101: The Landing Pages Report in Google Analytics

The easiest way to get started measuring your content attribution is through the landing pages report in Google Analytics (or Universal Analytics). 

To do that, go to Behavior > Site Content > Landing Pages. 

If you set up your conversions as goals in Google Analytics, they will show up in this report:

You can select the specific conversion goal you want to see in the top-right corner of this report. The results include all traffic, so if you want to only see organic traffic, you need to select the “Organic Traffic” segment. 

This attribution method is very basic. It only includes last-click attribution, so it’s not very precise. But it’s very easy to access, so it’s a good way to start. 

Another way is to go to Acquisition > All Traffic > Source/Medium:

Here, you’ll get a very straightforward attribution report per channel (very handy to pull up in a meeting in a few seconds): 

Don’t forget that in all GA reports, you can compare different timeframes to get a better view of the data across different time periods. And if you want to get a more granular report, you can use the advanced filtering function. 

In Google Analytics, the conversion is typically attributed to the last non-direct traffic source on the customer conversion path. 

Both of these reports are easy to pull up, but they only count conversions based on the last interaction (this is called last-touch attribution - more on that below).

This is problematic because this attribution model fails to paint the full picture of the customer's conversion path. 

Readers rarely convert directly from content. Their journey is longer and more complex. Last-touch attribution won’t show you that. It’s like seeing the last scene of a movie and trying to piece together the plot. There can be a million possibilities that you can’t see. 

How To Measure Content Attribution Correctly in Google Analytics

Multi-touch models are the best way to measure content attribution in Google Analytics. As I previously explained, these models assign credit to each measurable interaction in the customer’s path to conversion. 

To set it up, go to Conversion > Multi-channel funnels > Model Comparison Tool. 

With this model, you’ll be able to perform advanced content marketing attribution analysis that looks like this: 

The Model Comparison Tool allows you to set up three metrics: 

• Conversion: Here, you can choose the specific goal you want to analyze. Pick just one. It doesn’t make sense to analyze all goals at once. 
• Lookback window: This determines when Google Analytics will start looking at the data. There’s no reason not to set this at 90 days for content marketing.
• Model: Now, pick your attribution model - or models. This is why this tool is called a “model comparison tool” - it lets you compare 2, or even 3, attribution models.  

Let’s explain these 3 main attribution models for content marketing.

Last-Click Attribution

Last-click attribution is the most intuitive content marketing attribution model. When you think about what caused the customer to convert, you’d think about the last interaction they had with your website. That’s last-click attribution. 

This attribution model gives 100% of the conversion credit to the last channel the prospect has interacted with. For instance, if someone clicks your Facebook ad, then finds your website from a Google search, and then converts, 100% of the credit goes to the search.  

This model overlooks everything that happened before that final interaction. This can inaccurately represent the weight each interaction carried in the conversion.

This is why last-click attribution isn’t ideal for content marketing. Content is usually at the start or middle of a buyer's journey. People usually convert from an ad or direct visit rather than from the content itself. 

To get an accurate picture of what actually leads to conversion, you need to look at the whole journey your customers go through, not just the last click.

First Click Attribution

First-click attribution is a full 180 from last-click attribution. This model assigns 100% of the conversion credit to the first channel the prospect interacted with. 

So, if someone first discovered you via search, then saw a few Facebook ads, and lastly read a couple of blog posts before converting, the conversion will be attributed to that initial search. 

This model does an excellent job of accurately measuring which interaction gets the customers through the door. And since the goal of content marketing is often to get you discovered, it's a really effective attribution model to use to measure content effectiveness. 

Linear Attribution Models

Linear attribution assigns equal credit to each interaction in the customer’s path to conversion. 

For example, if a prospect discovers you through a Facebook ad, then finds your blog post via organic search, then comes to your website directly, and finally converts from another ad, each of these 4 channels will get 25% of the credit. 

So, while this model considers the entire customer’s journey, it doesn’t really tell you which of these interactions had the biggest impact. It doesn’t tell you what really tipped the scale toward conversion. 

👉 If you want to dig deeper into content marketing analytics, read about our three frameworks

How to Use the Model Comparison Tool To Measure Content Attribution The Right Way

The Model Comparison Tool is the best way to measure content attribution in Google Analytics. It lets us compare different attribution models based on three key metrics: conversion, lookback window, and model. 

To get started, open the Model Comparison Tool. 

Start by setting up the lookback window to 90 days instead of 30. Since we’re looking at overall content performance, we want to get as much data as possible.

Next, select the Landing Page URL as the Primary Dimension by clicking on "Other," searching for "URL," and clicking on "Landing Page URL." 

You should see a list of URLs with their respective data displayed. It should look something like this: 

The last step is to select all 3 content attribution models. Et voilà! You can now compare everything: 

Your data will vary big time depending on the attribution model you choose. As you can see in the screenshot above, getting first-interaction data can give you a better understanding of the role of content in the funnel.

Using Model Comparison Tool Segments 

Here’s a super helpful hack almost no one knows about in the Model Comparison tool. 

When you click “conversion segments” at the top of the screen, a list of segments will pop up for you to play with. 

You can even create your own. It will look like this:  

Using these segments will allow you to get more detailed and precise data. 

For instance, let's say you want to find out if organic search has an impact on people who came to your site via ads. 

To do that, start by selecting "First Interaction is Paid Advertising" in the segments and set the modules to Linear and Last Touch. Since it’s all PPC, you don’t need First Touch. 

Then, click on "MCF Channel Grouping" as the primary dimension, et voilá:

On this screenshot, you can see that out of the 71 conversion that happened when people first came to the site via ads, 11 of them (so around 15%) last interacted with the site via organic search. 

How about this for a slick move when leadership claims that content and advertising don’t work well together? Pretty slick, if you ask me!

You can also do the exact opposite: measure channel performance when the first interaction is organic search. 

Now on this screen, you can see that a lot of people come back to the site directly, even though their first interaction is organic search.

Here, we can also see that this client isn’t doing a great job at retargeting. They're not leveraging paid search enough to make the most of their huge organic traffic (or maybe it's just not working out for them).

This is just a glimpse into what you can do with this tool. Since you can create custom segments for the MCP, your creativity is the only limit.

The “Modern” Buyer’s Journey and the Limits of Content Marketing Attribution

I’ve already said it, and I’ll say it again: no attribution mode will ever be perfect. 

You may have heard people say that "the buyer's journey is changing," and while I'm not completely sold on that idea, the truth is that our habits and technology are constantly evolving. 

With new channels like TikTok popping up left and right, the way we interact with each other is changing too. 

All of this means that there are several limits to content attribution that you should definitely keep in mind.

Limit #1: Anti-tracking technology

Content marketing attribution will never be spot-on because the buyer’s journey data is tainted, to begin with. 

Anti-tracking technology is now pretty much everywhere. 

People are using tools like Privacy Badger to protect themselves from being tracked online. Some browsers block trackers by default. 

Firefox took it even further by blocking Google Analytics altogether. And let's not forget about those who regularly delete cookies to stay off the tracking radar.

What this means for content marketing attribution is that you’re losing part of the data before you even begin measuring anything. And if the data isn’t precise, the results won’t be either. 

Online privacy is becoming a big deal, and the major tech players are adapting to this new reality. 

But this doesn’t mean that attribution is going extinct anytime soon! There’s still server-side, cookieless tracking.   

Limit #2: Lookback window

The lookback window for content marketing attribution is usually 90 days. 

That’s the maximum number of days that Google Analytics will look back on to track your prospect’s interactions with your content. 

Most of the time, that’s enough. 

But if your sales cycle is longer than that or if people need to be exposed to your content for extended periods of time, it can become a problem. 

This is the case for high-ticket B2B SaaS, where sales cycles can last for more than 6 months.

Enterprise companies will usually do some due diligence even before requesting a demo or getting in touch with you.

Limit #3: People use lots of devices

We interact with content on multiple devices. 

How many times have you read a blog post on your laptop and then clicked on a Facebook ad from that same brand on your smartphone? 

We do this all the time. 

This makes content attribution significantly harder. If your prospect starts their interaction journey with your content on their laptop and ends up converting by clicking on your ad on their iPhone, you have no way of tracking this. 

Limit #4: People are … well, people

People are like wild cards - they interact with each other, and with your content, in all sorts of unpredictable ways. 

This means that your amazing blog post can be shared over email, in a Slack community, or via Whatsapp. 

It can be shared in a million different ways between people who work at the same company, and you’ll never know. 

Sure, there are some advanced attribution tools like LeadFeeder or HockeyStack (they actually use servier-side technology if I’m not mistaken - don’t ask me how they do it) that you could use to figure it out. 

But these tools are pretty pricey. Plus, we cannot vouch for how trustworthy or reliable they really are. 

Limit #5: Content isn’t always meant to be thoroughly read

I know this may sound a bit unconventional, especially coming from someone who owns a content marketing agency. But not all content is meant to be thoroughly read. That’s totally okay.

Sometimes, your content can serve as a validation tool. Having great content will help people form a positive opinion of you without needing to read every word you’ve written. It’ll make the right impression from a distance. 

But no attribution model will ever tell you that because it simply can’t be measured at all..  

Limit #6: Content Marketing is supposed to be part of a bigger ecosystem

SEO is a great way to get your content seen by the right people, but let’s be real - people don’t buy from you based on a few pieces of content. 

There are many other factors at play that carry a bigger weight in the buying process than your content. 

For people to buy from you, all your marketing needs to work together. Your positioning and messaging need to be on point, your social proof needs to be convincing, your content needs to be unique and helpful, your ad creatives need to be attention-grabbing, and so on. 

Content marketing is just one piece of a bigger marketing puzzle. 

Not being able to attribute the conversion down to one specific blog post is ok because doing so would miss the point of content marketing as a whole.

Don’t Get Caught Up In The Intricacies of Content Marketing Attribution

Content marketing attribution can be a tough nut to crack, and it’s easy to fall down the rabbit hole of data scrutinizing. 

With unreliable data to begin with, content attribution will never be as precise as we may like. And because of this lack of precision, we have to make the most out of the data we have at our disposal to prove how our content impacts our revenue and other KPIs. 

But as much as I love data, I won’t tell you to analyze everything down to the URL level.

It’s enough to analyze trends and patterns to figure out how your content is performing as part of a bigger marketing ecosystem. 

Plus, some things - like brand awareness or building authority and customer loyalty - cannot be accurately measured in numbers. 

You don’t have to have all the answers.