How to Optimize Content Marketing Conversions Using Channel Source Data

How to Optimize Content Marketing Conversions Using Channel Source Data

by Liam MoroneyMarch 12, 2018

Driving conversions is critical to content marketing success.

Effective calls to action (CTAs) can make the difference between a sustainable, measurable, positive ROI program or a short-lived experiment. Quality content is the most important component, for sure, but a considered conversion strategy for that content is what turns reader engagement into business action

However, many content programs place their CTAs – if they even have CTAs – without much thought, often settling for default modules that come with their webpage templates, rather than taking a strategic approach. The inevitable result is poor results. In this new era of performance content marketing, demand generation strategies must be core components of content marketing programs. It’s for this reason that NewsCred built our Advisory Services team with these skill sets to help our clients develop conversion strategies.

In this post, I’m going to share real data from a customer program over the course of several months. The data highlights the importance of having a CTA strategy and illustrates how different user journeys can result in very different CTA conversion performance. We’ll look at each situation and demonstrate how you can implement this data to optimize your own program.

Our Client Overview

Here’s a little more information about our anonymous program so you can better understand the data. Our example is a B2B SaaS business that developed a content marketing program to build brand awareness in its rapidly developing industry. Its primary goal is to generate new leads for the demand generation team. To accomplish this goal, NewsCred crafted the client’s content hub around two CTAs:

  1. Newsletter Subscriptions: Because much of the site traffic would be a new audience unfamiliar with the brand, the company wanted to offer a low commitment way to get people to regularly engage. It implemented prominent opt-in forms for its newsletter, such as a pop-up and right rail CTAs.
  2. High-value Content: This client already had plenty of lower funnel content, such as whitepapers, webinars, and research papers. It wanted to link top-of-funnel audiences to those lower funnel assets.

Engagement before Action

Before we dive into conversion rate data, it’s important to validate that our audience is showing the right level of engagement with the content. This information will be critical to understanding if we have a strategy gap. If we only looked at conversion rate data, we wouldn’t get a true sense of how well a channel is performing. For example, we could be led to believe that a certain channel is low-converting, when in reality, there’s a CTA mismatch.

Let’s start by looking at how each of the main traffic channels performed from an engagement standpoint. NewsCred’s analytics has a custom-built Engagement Rate that measures what percentage of traffic actively engages with content, so we’ll be using that measure here. You can utilize your site analytics too, but for this experiment, we’re using metrics designed specifically for content marketing.

Screen Shot 2018-03-01 at 3.16.38 PM.png

What you’ll notice above is how we have some clear champions, as far as engagement goes. Twitter and LinkedIn are driving the most content engagement, with Organic Search and traffic from the main Domain following closely. This is a great sign that we’re reaching the right audiences with content. But are we driving the right conversions from each?

Conversion Data: Newsletter Subscriptions

To best compare, we’re going to add a new bar group into the same chart as above. Keeping Engagement Rate data, let’s add the Conversion Rate for each of these channels against the first of our CTAs: Newsletter Subscriptions. We’re going to look at two versions of the same CTA, the sidebar CTA and the pop-up, since one is static and the other is timed to appear at a certain page depth. Typically, we’ll expect the pop-up to heavily outperform the static option.

Screen Shot 2018-03-01 at 4.14.44 PM.png

Let’s unpack this data a little bit. What you’ll notice right away is that our highest-performing channel – Twitter – is not generating newsletter subscriptions at all. Given the high engagement rate, something seems amiss. Typically, we’d expect to see newsletter subscriptions as the better performer for paid social, particularly since they likely haven’t had much exposure to content and are therefore more likely to convert against lower-commitment CTAs like this, rather than higher-value conversions like gated content CTAs. 

On the opposite side of the spectrum, the traffic from the domain is performing way above the other sources, which is often common, considering that this traffic just came from exploring the most product-centric information on the main site.

Email, too, is doing well, which seems counterintuitive. Why would an email source opt in for emails? It could be that this email traffic is from other nurturing sources, but could also be that we’re accidentally serving a redundant CTA to email subscribers. Using clear UTMs on your email traffic and digging into this data will help answer these questions.

Ultimately, though, we’re seeing a mismatch between some of our best performers here and the conversions they are each taking. Before making any major judgments, however, let’s look at some more conversion data.

Next up is the conversion rate for the static sidebar.

Screen Shot 2018-03-01 at 4.14.53 PM.png

Suddenly, we see a complete role reversal from Twitter, even though the goal outcome is exactly the same. What’s causing the sidebar to so heavily outperform the pop-up for our most engaged channel?

In our experience, this is likely a badly optimized pop-up. If you set the CTA to appear too early, you can serve an annoying CTA before a reader has had enough content exposure to even want to be subscribed to more. Time it too late, and you could miss them before they’ve read enough and moved onto something else. A/B testing is your best friend to find this sweet spot, but bear in mind that you may need to have various versions for different content lengths or sources.

In this case, because the static CTA is converting better, it’s highly likely that the pop-up is appearing too late and not reaching the captive audience, while the static option is. There’s a great opportunity here to test this theory and drive those conversions up on Twitter.

But what about the other channels? Organic is a very valuable and engaged channel, but failed to convert on any subscriber CTAs. Similar outcomes for LinkedIn and Facebook. So let’s look at a higher-value CTA: signing up to receive Gated Content

Conversion Data: Gated Content CTA

Screen Shot 2018-03-01 at 4.15.00 PM.png

Here, we’re seeing a similar pattern to our first pop-up results. Domain traffic is driving very powerful results again for lower funnel content too. This shows just how valuable traffic that comes from your main site can be if targeted correctly.

However, our other champions are all performing low. For this CTA, it’s not surprising, since paid social is much less likely to convert on high-value actions. It also means that we’re not driving any successful conversions from our most expensive channels. 

More worrying though, is the Organic channel. For the work it takes to rank in search results, we’re clearly not serving this channel with any effective or appropriate CTA. This could be because we’re simply not recording the right action. But this data is a critical piece to help identify just what that action should be. The advice here would be to dig into the pages that drove that organic traffic and see if we can add logical CTAs for the questions that drove them to find that content. Perhaps, it’s simply more of that topic, or specific gated content related to that.

As we build and refine our CTA strategy, we’ll start to move closer and closer to CTA personalization when we have the right data.

Conversion Data: Returning Audiences

Before we start drawing conclusions, let’s look at one more view of this data. How did traffic that was coming back to content compare to brand new visitors? This view will help validate a lot of the data that we’ve already collected. In our experience, returning visitors can be much more likely to convert against higher-value actions, since they’re coming back to deepen their content knowledge. So making sure you’re optimizing for that can be crucial to this primed audience. Let’s look at each of our conversions comparing new and returning visitors:

Screen Shot 2018-03-01 at 3.51.21 PM.png 

Exactly as expected, we see that new visitors are more likely to convert against our higher-funnel goal, while not converting against the lower funnel goal at all. On the other hand, returning visitors do convert against each of our goals, but they’re nearly 4x more likely to convert towards gated content.

Before we start assuming if all returning visitors should be served this goal, let’s break down our returning visitors in the gated content goal to see just where they came from:

Screen Shot 2018-03-12 at 9.32.38 AM.png

Right on cue, we see the most controllable sources of returning visitors topping the list: Email and Domain traffic. Unsurprisingly, these sources have had the most exposure to your brand and content, so lower funnel conversions are logical next steps for their journey. The paid sources and Organic are clearly not at that place yet, and so this helps us identify which CTAs we should be serving these return visitors.

How to Optimize CTAs Based on the Conversion Data 

We’ve just looked a lot of data, and found a host of scenarios and outcomes. Where do we go next? Let’s recap each of the major data points here, and build some strategic next steps for our client’s program.

1. Optimize the Pop-Up

We saw how our pop-up worked well on the domain traffic, but didn’t move the needle on the engaged Twitter audience. Yet the sidebar CTA showed a different story. We need to start running A/B tests on our pop-up for each major channels to find the right moment for each. Most pop-up platforms, such as OptinMonster, which NewsCred uses, allow for controllable triggering for different channels with A/B testing options, so that you can run these tests easily and with confidence.

2. Rethink the Organic Traffic CTA strategy

None of the CTAs performed well with Organic visitors. This is not surprising when you think about it. Traffic sources like paid social were likely prompted to read a piece of content based on a boosted or sponsored post. But Organic traffic found its way there based on a search query. They were looking for answers to a question, and the engagement rate suggested that they found it. However, it doesn’t look like we gave them a logical next step. Sometimes, just getting them to read more content is that next step, and serving them that pop-up may not be appropriate until they’ve gotten close to fulfilling what their search query was.

 3. Treat Returning Visitors with Consideration

As we’ve now seen, returning visitors are in a different stage in their journeys. But even that is quite different based on their traffic source. There’s clearly a time and a place to target them with higher-value CTAs, but let the data guide you on that path. In some cases, like for Organic and Social return visitors, building a pop-up for these visitors may be a great test. Many pop-ups are cookie based, so they often only show up once for a visitor’s first session. This could mean that returning visitors from these sources aren’t seeing a pop-up when it’s most appropriate. Again, a triggered pop-up just for return visitors may be just the ticket.


The examples shown in this experiment were specific to a B2B company, but the same data rigor applies to all client types. As content marketers, it’s imperative to consider what the logical next step for our audiences should and could be, and understand that there may be several for a single piece of content because of how many simultaneous journey types may lead to that post. To truly have an optimized program, we need to think about all the angles, journeys, users, and motives that lead a visitor to read content.


Liam Moroney is NewsCred’s Director of Analytics.