When we launched Marketing Qualified Leads (MQLs) at the start of Q1, it turned our funnel management on its head. As the person running Marketing & Growth Operations at Hex, I’ve learned that the secret to pipeline efficiency isn’t just finding more leads — it’s knowing who sales should reach out to, and exactly when they’re most likely to convert. With so many signals in play, prioritization is everything.
Before MQLs, our model was a bit of a black box — confusing, and not great for driving accountability across marketing or sales. Our challenge was simple but critical: if we could identify both the right person and the right moment in their Hex journey, we could help sales spend their time where it counts, boost conversions, and keep our funnel running like a well-oiled machine.
Here’s how we started with the basics and evolved our approach — using data (and a bit of AI) to finally answer that who, when, and why.
How we got here
Okay, scoring leads isn’t new — but here’s how we did it at Hex. Each contact got points for things like persona, campaign engagement, website visits, industry, and which data warehouse their company used. We set a threshold, and when someone crossed it, sales got a notification. Suddenly, both marketing and sales had clear swim lanes and real accountability. Did it feel basic? Yes. Did it work? Also yes. It was the clarity we needed, even if a little old-school.
But let’s be honest: there was still a lot of gut feel involved. Month after month in the MQL Council — a working session I run to review performance and iterate on the model — I found myself poring over conversion data. Which titles convert? Which campaigns drive opportunity creation? Important work, but the “statistical significance” was questionable. I wanted confidence that the changes I was making would actually improve our conversion rate, not just vibes.
Then I remembered I work at Hex and have a fantastic data team! I reached out to Caleb, one of our data analysts, and said, “Can we figure out which features really matter for conversion?” His answer: “Yes, and we can make it repeatable.”
Together, we built an interactive app using regression analysis, so that I could finally move from gut feel to data-driven updates on demand.
Cracking the code on high-intent leads
Our data app takes every contact and every opportunity from the last twelve months, maps their attributes and engagement, and runs them through a regression model. It surfaces the traits and behaviors that actually predict pipeline creation, not just at a high level but for each segment (Enterprise, Mid-Market, SMB).
But here’s the real magic: Not everyone wants to dig into statistical coefficients or squint at regression tables. Hex’s AI features bridge that gap — turning all that data science under the hood into plain-English summaries and actionable recommendations to both our scoring logic and targeting strategy. Now, marketers and ops leaders (like me) can act on the data, not just admire it from afar — or worse, ignore it because it’s too technical.
Explore this demo app or create an account and clone it!
What the model actually revealed: segment stories that change the game
What surprised us most wasn’t just which signals moved the needle, but instead how the patterns shifted across segments.
What did we actually learn?
Hand-raisers win, everywhere.
Leads who request a demo (we call them hand-raisers) are far and away the most likely to convert, no matter the segment. It's nice when the classics hold up.
Seniority matters, but not always the way you think.
At Enterprise companies, targeting manager-level contacts worked best; in Mid-Market, directors were the sweet spot. For SMBs? Seniority faded in importance — intent trumped title (intent would be downloading eBooks, viewing the pricing page, signing up for a trial, etc.).
Industry matters.
Prospects in software, healthcare, and media/internet industries showed higher conversion rates — so we know where to double down.
Why does this matter? Now, when we run campaigns or adjust scoring, we can get hyper-specific in ways that match segment realities and evolve from a one-size-fits-all strategy.
Why this isn’t just another analysis
This isn’t just another one-off analysis. Because the app lives in Hex, anyone on GTM or data can update the data, rerun the model, and see what’s changed instantly. If someone on demand gen wonders how to hit next quarter’s target, they can get segment-specific answers in minutes. It’s fast, collaborative, and reusable. (Honestly, it’s also just more fun.)