When a mid-market SaaS company came to us in Q1 2026, they were spending $28,400 per month on Google Ads and generating roughly 18 sales-qualified leads. Their cost per SQL sat at $1,578, which is nearly three times the industry benchmark for B2B SaaS paid search. Over 90 days, we restructured their attribution model, rebuilt their campaign architecture, and fixed a landing page issue that was quietly killing conversions. The result: cost per SQL dropped to $614, a 61% reduction, while monthly ad spend stayed within 5% of the original budget.
The Starting Situation: Leads That Sales Would Not Touch
The client's CRM showed 140 form fills per month, but their sales team was only booking discovery calls with about 13% of those. The rest were a mix of students, competitors, and people who had clicked on broad-match keywords like "project management software" with no intent qualifier. Marketing was celebrating a low cost per lead of $203, while sales was frustrated by a pipeline that went nowhere.
The first thing we did was redefine the conversion event. Instead of counting form fills, we started tracking only leads that reached a specific thank-you page AND had a session duration over 90 seconds on the pricing page. This single change made the real SQL count visible inside Google Ads for the first time. It also revealed that one campaign, accounting for 34% of total spend, had generated zero SQLs in the prior 60 days.
This mirrors a pattern we see repeatedly: teams optimise for the wrong event, and Google's algorithm happily delivers volume against that event at low cost, while the actual pipeline stays empty. If you want to understand why this happens at a structural level, our breakdown of why Google Ads campaigns fail to generate quality leads covers the mechanics in detail.
Phase 1: Attribution and Tracking Fixes (Days 1-21)
We implemented a two-stage conversion import from Salesforce into Google Ads using the offline conversion import API. Stage one fired when a lead was created in the CRM. Stage two fired when a sales rep marked the lead as SQL, typically within 48-72 hours. Both events were imported with their actual conversion dates, not the date of import, which prevented attribution lag from distorting bidding signals.
We also set up a multi-touch model in GA4 to understand which channels were assisting SQLs versus claiming last-click credit. Paid search was responsible for 41% of first touches but only 19% of last touches, meaning it was doing significant top-of-funnel work that last-click reporting was erasing entirely. For more on choosing the right model for B2B pipelines, see our guide to multi-touch attribution for B2B ROI.
Phase 2: Campaign Restructure and Keyword Pruning
Before the restructure, the account had 11 active campaigns with heavy overlap and no clear separation between brand, competitor, and solution-aware traffic. We consolidated to four campaigns: branded, competitor conquest, solution-aware (e.g. "resource planning software for agencies"), and retargeting. Each campaign had a single bidding goal tied to the SQL conversion event, not the form fill.
Keyword pruning removed 847 active keywords that had accumulated spend over 90 days with zero SQL contribution. We also added 214 negative keywords across the account, specifically targeting job-seeker terms, free-tool queries, and academic research patterns. The wasted spend recovered from this step alone was approximately $6,200 per month, about 22% of the original budget, which was reallocated into the solution-aware campaign.
- Removed 1 campaign responsible for 34% of spend and 0 SQLs
- Added 214 negatives targeting zero-intent traffic patterns
- Reallocated $6,200/month from pruned keywords into high-intent ad groups
- Switched all campaigns from Target CPA on form fills to Target CPA on SQL events
- Reduced active keywords from 1,340 to 412, tightening match type to phrase and exact
Phase 3: Landing Page Parity and Messaging Alignment
The solution-aware campaign was sending traffic to the homepage, which had no mention of the specific use case the ads were referencing. A visitor clicking "resource planning software for agencies" expected to land on a page about agency resource planning. Instead they got a generic SaaS homepage with five different use-case tabs and no clear CTA hierarchy. Bounce rate on that traffic was 74%.
We built two dedicated landing pages, one per primary ad group theme, with a headline that matched the search intent, a single CTA (book a 20-minute demo), and a short qualification form asking company size and current tool. Removing the open-ended "tell us about yourself" field and replacing it with a dropdown reduced form abandonment by 38%. The qualification questions also pre-filtered leads, so sales only received submissions from companies with 10 or more seats, which matched the client's ICP.
The combination of message match and form qualification pushed the landing page conversion rate from 2.1% to 5.6% on solution-aware traffic. That improvement, combined with the keyword pruning, is what drove the majority of the SQL cost reduction. Getting landing page fundamentals right is often the highest-leverage fix available, and the common failure modes are documented in detail in our post on why B2B landing pages fail to convert.
Results at Day 90 and What Came Next
By the end of the 90-day engagement, monthly SQL volume had increased from 18 to 46 while total spend moved from $28,400 to $28,900. Cost per SQL dropped from $1,578 to $614. Pipeline value attributed to paid search in the CRM increased by 187% quarter over quarter. The sales team's close rate on paid-search SQLs was 24%, compared to 9% before the campaign restructure, because the leads arriving in the CRM were now pre-qualified by both keyword intent and the form.
The client used these results to make a budget case to their board for a 40% increase in paid search investment in Q3 2026, this time with a clear cost-per-SQL target and a Salesforce attribution model to defend every dollar. The lesson is not that any single tactic drove the outcome. It was the combination of measuring the right thing, removing budget from campaigns that looked efficient but were not, and aligning ad creative with landing page experience. Each step alone would have moved the needle modestly. Together, they compounded.