Client stories

Three accounts. Three different leaks.

Representative, anonymized examples show how we diagnose paid search. Figures illustrate plausible engagements and are not guarantees.

Regional home services · 42 days

Calls looked healthy. Booked jobs did not.

What was happening?

The account counted every 45-second call as equal. Sales records showed that emergency terms, broad geography, and several partner queries produced calls but few booked jobs.

What changed?

Call outcomes returned to Google Ads, radius targeting followed actual service capacity, and query groups split by job value. A new landing path exposed service area and arrival time before the call.

Illustrative outcome

CPL moved from $118 to $82 while booked-job rate moved from 21% to 31%.

B2B workflow software · 8 weeks

Demo volume hid a pipeline problem.

What was happening?

Broad non-brand campaigns generated activity at the top of the funnel. CRM stages showed that many demos came from students, tiny teams, and comparison traffic with no near-term project.

What changed?

Qualification language moved into ads and pages. Search terms were grouped by use case, offline opportunity stages became the primary value signal, and weak informational clusters lost budget.

Illustrative outcome

Non-brand spend fell 18% while qualified demos increased 27%.

Specialty retailer · 6 weeks

One asset group treated every product the same.

What was happening?

A catch-all Performance Max campaign ignored margin, stock position, and new-customer value. Brand demand also made the blended return look stronger than acquisition performance.

What changed?

Products split by margin and inventory state. Titles adopted high-value attributes from search data. Brand reporting, new-customer value, and feed diagnostics moved into the weekly decision memo.

Illustrative outcome

Wasted spend fell 23%; new-customer ROAS moved from 2.4× to 3.1×.

Your next 30 minutes

Find the most expensive wrong assumption in your account.

Get a free strategy call