AMA: Counterpart Marketing Lead, Erika Barbosa on Tech Stack

It starts with clarity on what data actually needs to move between systems and why.It’s easy to fall into the trap of syncing everything, but that usually leads to unnecessary complexity. I’ve found it’s more effective to map the customer journey first and pinpoint the key moments where data needs to flow between tools. For example, lead capture from a form to enrichment, then routed via the CRM and synced to marketing automation or the product.I try to keep the setup as simple as possible. A fe ...Read More

I try to build testing into the process wherever it makes sense, rather than treating it as something separate or extra. Not every campaign needs an A/B test, but when there’s a clear hypothesis or decision to inform, it’s a great way to learn and improve over time.Before launching anything, I define what I’m testing and what I’m hoping to learn. For example, if I’m testing a landing page headline, the goal isn’t just to see which one converts better, but to understand why one message might reso ...Read More

I try to start with the outcome we’re driving toward, not just the channels we’re using. That makes it easier to track performance in a way that’s meaningful and not just a report full of vanity metrics.For most campaigns, I’m looking at a mix of channel-level performance like click-throughs, conversions, and engagement, paired with deeper funnel metrics like pipeline and revenue impact. We’ll pull that together through the CRM and whatever reporting layer we’re using, whether that’s native dash ...Read More

Good segmentation starts with clean data. No matter how sophisticated your tools are, if the data is messy or incomplete, the targeting will fall flat. So I put a lot of emphasis upfront on standardizing fields, defining ownership across systems, and building regular audits into the process.Once the foundation is solid, I segment based on a mix of firmographics, behavior, and lifecycle stage. That might mean creating different streams for personas, company size, product interest, or recent activ ...Read More

Attribution modeling is tricky because most traditional models can over credit the last or first touch without showing the true impact of each channel. To get closer to reality, I lean heavily on incrementality testing.Incrementality testing helps isolate which channels are genuinely driving new leads or conversions versus those that are just capturing activity that would have happened anyway. That means setting up controlled experiments, like holdout groups, where a portion of the audience is e ...Read More