Direct vs. influenced
Gated content is easy to measure. Someone fills out a form, downloads an ebook, becomes a lead, eventually becomes a customer. I can tell you to the penny how much money that ebook directly influenced. That's direct attribution - a linear line from content to revenue.
But the direct number is always the lowest number. It's probably higher. Some people download an ebook after they've already started a free trial. The ebook doesn't get credit for that one - we just get the name from the form.
Influence is the bigger picture. A piece of content might influence $150,000 of net new pipeline in a year without ever being the converting touch. Someone keeps coming back to the blog once a week - that never converted them, but they continued to engage. We can see the impact of specific articles on any milestone, whether that's becoming a lead, an MQL, or a customer.
For company-wide reporting, we stick to direct. The number is lower but it's defensible. For smaller team decisions - what to write next, which topics are working - influence helps us decide. We'd never put influenced numbers in the company report because you'd have $17 million of pipeline "generated" by content with a ton of overlap.
Gated vs. ungated
This is where content measurement gets hard. Gated content with form fills - easy. Ungated content is where I struggle too.
We started moving to account-level tracking instead of contact-level. Using Reveal, we could see which companies were engaging with ungated content even without a form fill. It shifted from "how many people read this" to "which accounts are reading this, and did they eventually become customers."
Whenever you can gate the content or grab an email, that's the easiest way to get the clearest measurement. It's as easy to measure as paid. Ungated is where everyone struggles.
Working with content teams
I built an attribution model in dbt with all our touchpoints. So for any contact, you could see their full journey - John at Acme downloaded the Chrome extension via a Google search, then downloaded an ebook 17 days later, attended a webinar, and became a customer seven months after that first touch. All of that lives in Salesforce on the contact record so the content and demand teams can slice and dice it themselves.
Content usually isn't the thing that pushes someone past the finish line. But it shows up everywhere in the journey. The content team can see what influenced deals at every stage without needing me to build them a report.
What I learned
- 80% is good enough. You can get 80% of what you need with 20% effort. That last 20% of accuracy is where you spin your wheels. With tracking issues and privacy, you're never getting to 100%. Get agreement that 80-85% is close enough and move on.
- Don't give too much data at once. When I hand over everything, nobody asks questions - they're just confused. Spoonfeed a little bit. They get one answer, now they have a follow-up question, now they want another sliver. That's when it's working.
- Come with the problem, not the report. The best content teams come to me with the question they're trying to answer, not the dashboard they've already designed in their head. When someone says "build me X," they use it once and never open it again. When they come with the problem, we solve it together and they actually use it.
- Start simple. Google Analytics with goals and conversions set up properly gives you a lot. I like to keep myself honest by making sure my data matches GA closely enough. The rank order should be the same - if our top five match, the discrepancies are just server-side vs. third-party differences.