Case 02 · Long-Term Account Management · B2B · IT Asset Liquidation
When the wrong audience kept clicking,
we rebuilt around the right one — twice.
The problem
The client liquidates IT assets in bulk — memory, servers, network equipment, hard drives — and needed leads from B2B buyers and sellers moving volume, not individuals with a single laptop to sell. When the account started in June 2023, the existing phone call campaigns, display ads, and dynamic search ads were doing the opposite: pulling in people selling one computer, or shoppers looking to buy a single used laptop.
That mismatch is a common trap with dynamic search ads on a niche B2B service — the algorithm matches on surface-level keyword relevance, not buyer intent, so it fills the funnel with the wrong audience even while metrics look reasonably healthy on the surface.
What I did
Split one blended budget into five product-specific campaigns
Replaced the generic setup with dedicated campaigns for Memory ($7/day), Server & IT Asset Management ($45/day), Hard Disk Drives ($20/day), Website Visits ($50/day), and other US locations outside California ($20/day) — each with its own keyword set and ad copy matched to that specific asset category.
Built a dedicated exact-match ad group for bulk B2B buyers
In August 2023, created a separate ad group targeting bulk and B2B buyers specifically, using exact-match keywords instead of broader match types — since this business's audience is narrow enough that broad matching kept pulling in irrelevant intent.
Ran aggressive, ongoing negative keyword filtering
Continuously pruned consumer-resale and job-seeker traffic — terms like "computer recycling," "sell gaming PC," "used laptop buyers," and employment-related searches — removing hundreds of irrelevant keyword matches across campaigns as new junk patterns surfaced.
Layered Performance Max once Search traffic was clean
Added PMax campaigns in March 2024 and again in October 2024 specifically to reach corporate buyers for server-based solutions — deliberately sequenced after Search targeting was already refined, so PMax wasn't scaling the same wrong-audience problem.
Diagnosed and rebuilt through a 3-month performance dip
From January to March 2025, cost per conversion spiked 212%, then 156%, then 52% in successive months as a new geo-targeted campaign underperformed. Rather than waiting it out, switched bid strategy from Maximize Conversions to Maximize Clicks, renamed and restructured ad groups, ran an extensive keyword cleanup pass, and paused the underperforming campaign entirely — recovering to a 117.3% conversion increase by April.
Results over time
This account's real story is in how it responded to a downturn, not just the highlight months:
Full results (May 2025 snapshot)
The key shift
Before, the account measured traffic. After, it measured whether that traffic was actually a bulk buyer or seller — and when a new campaign broke that alignment in early 2025, the response was to rebuild it, not ride it out. That willingness to admit something stopped working is what turned a three-month slump into a 117% recovery instead of a slow bleed.
Recognize this in your own account?
If your ads are pulling in traffic that looks active but isn't the right buyer, let's talk. I'll show you exactly where the mismatch is — no pressure.
Message me on WhatsApp →