“Did it work?”

That’s where things tend to fall apart. Not because the strategy was bad, but because the tracking was.  If you’re not using distinct lead sources and clean UTM structures when testing new initiatives, you’re basically trying to evaluate performance with half the story missing. And no, this isn’t overkill, it’s the difference between insight and guesswork.

Why You Need New Lead Sources (Yes, Even If It Feels Like it is “In the Weeds”)

It’s tempting to just dump new leads into existing buckets like “Paid Social” or “Display” and call it a day. It feels cleaner. Less work for everyone in the setup, fewer things to explain in reports. It’s also how you end up with zero clarity on what actually drove results. When you create a dedicated lead source for a new initiative, you can finally answer questions like:

• Are these leads actually incremental, or just coming from people who would’ve converted anyway?
• Is this channel bringing in quality prospects, or just inflating volume?
• What’s the real cost per lead when it’s not hiding inside blended numbers?

Without that separation, everything looks… fine. And “fine” is usually where good ideas go to die.

UTMs: The Unsung Hero of Not Guessing

Lead sources tell you where something came from. UTMs tell you what specifically made it work. Think of UTMs as the difference between “this came from paid social” and “this came from a retargeting campaign on Meta using that one video everyone argued about but secretly works.”

With a solid UTM structure, you can quickly see:

• Which platform is actually pulling its weight
• Which audiences are worth scaling
• Which creative deserves more budget (and which should quietly disappear)

Without UTMs, optimization becomes a lot of “we think this is working,” which is not a great place to be when budgets are on the line.

“But Doesn’t This Mess Up Our Media Mix Reporting?”

Short answer: no. Slightly longer answer: it actually makes it better.

There’s a common worry that breaking out new lead sources will fragment reporting and make it harder to evaluate performance at the channel level. But that’s only true if you stop at the fragmented view, which you shouldn’t. This is why we make sure that we can roll granular data back up into bigger buckets because what you can’t do is retroactively break apart blended data once it’s been lumped together. It’s like cooking: you can combine ingredients later, but you can’t un-bake a cake. By keeping new initiatives separate at the start, you get:

• A much better understanding of what’s actually driving results
• Clear performance at the initiative level
• Clean roll-ups for channel-level reporting

Better Testing = Better Decisions

When you isolate new initiatives properly, a few good things happen:

• You stop over-crediting your legacy channels just because they’ve been around longer
• You can spot incremental gains instead of assuming everything is business as usual
• You make faster, more confident optimization decisions

And maybe most importantly—you can actually prove when something is working (or not) without squinting at blended dashboards and hoping for the best.

The Bottom Line

Using distinct lead sources and UTMs isn’t about making reporting more complicated. It’s about making your insights more trustworthy.  Because at the end of the day, testing without clean tracking isn’t really testing—it’s just… trying stuff and hoping it works.

And that’s not a strategy.