6 months after my CFO requested I provide evidence that our content marketing efforts were worthy of the $43,000 per year in content operations investments he made, I still couldn’t provide an answer.
It’s a fair request. He could demonstrate ROI (return on investment) in all areas of his organization except content. While we produced thousands of articles each month, there was no way to connect any of them directly to closed revenue. Traffic and engagement were good indicators of interest; however, they were not indicators of influence as it related to driving new clients.
In order to address this issue, I had to completely rebuild how we track and measure the performance of our content.
Prior to this effort, we measured the wrong side of the funnel. We tracked page views, time on site, social shares, etc. While these are great indicators of consumer behavior toward your company's content, none of them demonstrated whether or not your content was contributing to revenue or simply being a drain on your budget.
The distance between “Someone read our blog” and “Someone signed up to be a paying client” is incredibly large. There are dozens of touches that occur between when consumers consume your content and sign their name on a contract. In the absence of tracking those interactions, you are essentially making assumptions about how effective your content has been. We made assumptions for 3 years. Published consistent amounts of content, were able to increase website traffic, and assumed that our traffic increases would eventually help drive business development efforts somewhere down the line. But we had absolutely no proof.
My CFO needed proof. As such, I was required to build a system that did not previously exist.
The System/Stack That Allows Us To Track Revenue-Generating Content
First, I used Hubspot for both CRM and marketing automation. It is very expensive, approximately $800 per month. However, it is also necessary if you wish to implement an attribution stack that will work correctly.
Each piece of content is tagged using UTM parameters so that we may determine where the person originally came from (source), what type of advertising vehicle they used (medium), and the campaign name. Once they click on a link from a LinkedIn post to the next level (blog article) and then ultimately fill out a contact form, every interaction in between is captured by Hubspot. This allows us to identify which content items were consumed by the individual prior to becoming a lead.
Using Google Analytics, I am able to pass additional behavioral information into HubSpot, including:
- How much time they spent reading the article(s)
- Which other articles were viewed?
Was this a single visit conversion, or were there multiple visits before they converted?
All of this behavioral data is important. For example: A person who reads one article and immediately completes a contact form is different from a person who reads 12 articles over a period of six weeks before calling. Both are valuable; however, they demonstrate different levels of content effectiveness.
To further understand user experience and ultimately identify potential friction points within our website, I use Hotjar to record actual video sessions of users viewing our content pages. By doing so, I can see where they pause, what links they click, and where they leave. This provides additional insight into where users are experiencing friction during the customer journey that I wouldn't be able to capture solely using analytics.
For instance, using Hotjar, I discovered that individuals were abandoning some of our most successful blog articles due to CTA placement at the bottom of the page. Since many of the users weren't scrolling far enough to view the CTA button, it was abandoned altogether. Ultimately, moving the CTA button above the fold resulted in a 41% increase in conversions on those same articles.
Had I relied solely on Google Analytics data alone, I would have missed identifying that exact issue and opportunity.
Finally, using Salesforce.com integration with Hubspot, I am able to pass lead data to our sales teams along with historical data regarding which content was consumed by the lead prior to their first conversation. This changes how sales reps engage in conversations with leads as they now enter conversations already having knowledge of the specific content items the lead consumed prior to contacting the organization.
When the data really speaks for itself
We conducted a complete attribution study when we had the appropriate tracking system in place for 6 months. And then we found out things we hadn’t expected.
23% of the revenue we closed involved content somewhere in the buying process. All of our deals did not involve content, but a significant amount of them did. The $217,000 in annual contract value (acv) of those deals directly attributed to content marketing equated to 23% of the total number of closed deals ($943,500).
Our spend on content marketing was $43,000. Our return on that spend was $217,000. The CFO finally quit asking me how much money I wasted on content budgeting.
The best part of this story was figuring out which specific content items drove attribution. Only 84% of the revenue attributed to content came from 11% of the pieces we published.
A lot of content we created generated traffic and engagement but in no way affected purchasing decisions. A small percentage of the content produced the majority of the revenue. We now knew exactly which pieces were driving revenue, and therefore we could increase production of the good ones and eliminate waste on content formats that looked successful based on vanity metrics but ultimately never drove business outcomes.
Case studies converted 3.2 times higher than thought leadership posts; implementation guides generated two times as many qualified leads as trend analysis pieces; founders' perspective content built relationships and rarely influenced deals.
Using the data we obtained from this exercise, we changed the type of content we produce. We have started creating more case studies and implementation guides. We are producing less trend analysis-type content. We still create founder's-perspective content, but we recognize that it builds trust and does little to nothing in terms of influencing deals directly.
Six months later, after adjusting the type of content we produce, the percentage of closed deals attributed to content went up to 31%. The ACV associated with these deals was $342,000. Spending the same budget and using better targeting based on attribution data.
What prevents most companies from doing this?
Companies usually fail to invest in the necessary infrastructure to track attribution. Hubspot + integrations + proper UTM hygiene + sales team actually using the data = not trivial to implement.
Additionally, there is a requirement for patience. An attribution model needs sufficient volume before any patterns will emerge, not just three months of data but twelve months that provide trends you can take action on.
However, the alternative is to guess about content ROI forever. Produce based on what feels right instead of producing based on proven data ROI. Use vague claims about brand building vs. concrete revenue numbers to justify budget defense; you can't optimize what you don't measure, and you can't prove value without attribution. Simple as that.