- More Than Media Marketing Newsletter
- Posts
- Don't Be Afraid To Cut Your Losers
Don't Be Afraid To Cut Your Losers
Even though a keyword may feel relevant for paid search, the data might not always support that assumption.

I’ve been having conversations with a few clients recently about paid search keywords, and an interesting theme has been popping up. We’ve been addressing certain keywords, or keyword themes, that feel as if they should be relevant to my clients’ products/services, but actual performance doesn’t support that assumption.
The majority of the paid search campaigns I’m running for clients are set up with the intention of targeting high-intent keywords that will contribute to generating leads. Given that goal, one of the main ways we evaluate performance is by looking at last-touch performance in terms of CPL and pipeline:spend.
When analyzing recent keyword performance I found that a specific subset of keywords for some clients have been underperforming compared to others in the same campaign. The interesting piece is that at face value these keywords are definitely relevant for these clients, and that can make the decision to pause these keywords anything but straightforward.
My approach for these scenarios usually boils down to taking one of two approaches:
If the keyword itself is relevant but not performing, I often segment that keyword into its own ad group, and start crafting specific copy for the associated ad and landing page. Sometimes this can take a week or two depending upon the landing page development needs.
Otherwise, I pause the underperforming keyword(s) and allow the campaign to invest that budget in keywords which have shown stronger performance recently.
When choosing the second option, I explain to my clients that this doesn’t mean these keywords are off the table forever. Instead, we’re focusing our short term investment on what we know works, while making some optimizations for other keyword themes where we believe there’s an opportunity.
I’ve seen this have a dramatic positive impact on campaign performance in some instances. While it may feel uncomfortable, or like we’re leaving opportunities on the table, the data rarely lies and has proven to be a helpful guiding factor when making decisions thus far.
Have questions, considerations, or critiques? I’d love to hear them! If you’re reading this via email, just hit respond. Otherwise, you can find me on LinkedIn.