Case 03 · Multi-Service B2B Search · Materials Testing Lab · Canada
CMH Metlabs runs a full-service metallurgical and materials testing lab — failure analysis, coating analysis, chemical analysis, mechanical testing, NDT inspection, and more — serving industrial, aerospace, energy, and manufacturing clients across Canada. That range is a strength for the business and a real risk for a Search campaign: each service line has its own technical vocabulary, its own buyer, and its own risk of bleeding budget into a neighboring specialty's keywords if the account isn't segmented carefully.
A generic "metallurgical testing" campaign structure would blur eight genuinely different buying intents into one undifferentiated pool — someone searching for weld inspection and someone searching for polymer degradation testing are looking for completely different services, and treating them the same wastes spend on irrelevant clicks for both.
Segmented the campaign into Failure Analysis, Metallurgy Testing, Chemical Analysis, Coating Analysis, Mechanical Testing, Polymer & Plastics, NDT Services, and Generic — then actively built out targeting for each one rather than just filtering: over 2,200 phrase, exact, and broad-match keywords added specifically to match how an engineer searching for weld inspection writes a query versus one searching for polymer degradation testing. Each ad group also got its own responsive search ads — 13 in total — so the ad copy spoke to that specific discipline instead of a generic "lab testing" message.
Created a dedicated conversion goal in June 2025 and continued refining which actions counted toward it through 2026 — including and excluding specific actions like call-button clicks and GA4-sourced form submissions as the account's tracking setup matured, so reported conversions reflected real inquiries rather than tracking noise.
Through September and October 2025, cost per conversion climbed past CA$100 as broad-match keywords across all eight ad groups picked up increasingly loose, lower-intent traffic — a common failure mode when broad match is left to expand unchecked across multiple technical service lines at once.
On October 27, 2025, paused 614 broad-match keywords in a single coordinated pass across every ad group — Generic (230), Metallurgy Testing (112), Mechanical Testing (91), Chemical Analysis (67), Failure Analysis (69), Coating Analysis (39), and Polymer & Plastics (6) — cutting the account back to the keyword sets each ad group had been specifically built around.
Adopted a portfolio Maximize Conversions bid strategy for 2026, adjusting Target CPA and daily budgets on an ongoing basis as performance data came in — scaling spend up from roughly CA$50–90/day in early 2025 to CA$200+/day by mid-2026 as the account consistently proved it could convert efficiently at higher volume.
The account's real turning point is the October 2025 cleanup — everything since has held at a stronger baseline:
Before the October 2025 cleanup, the account was chasing volume across eight service lines with keyword matching left too loose to tell them apart. After it, every dollar had to earn its place in a specific, high-intent ad group — and once that discipline was in place, the account could scale spend with confidence instead of scaling risk.
If your business covers multiple services and your Google Ads account treats them all the same, let's talk. I'll show you exactly where the overlap is costing you — no pressure.
Message me on WhatsApp →