A UK-based SaaS company selling an HR and payroll platform to mid-market businesses (50-500 employees) came to us in Q1 2026 with a familiar problem: their Google Ads account was spending £18,000 per month and generating leads at £340 each, most of which their sales team described as "low intent" or outright irrelevant. After 90 days of structured changes across campaign architecture, keyword strategy, and landing page alignment, cost per lead dropped to £133 and the qualified lead rate improved from 22% to 58%.
The Starting Point: What the Account Actually Looked Like
The account had one broad campaign containing 14 ad groups, all pointing to the platform homepage. Match types were a mix of broad and phrase, with no negative keyword list of any substance. The client had added broad match terms like "HR software" and "payroll system" without layering in audience signals or demographic exclusions, which meant significant budget was being consumed by searches from job seekers, students, and solo freelancers, none of whom were the target buyer.
Quality Scores across the top-spend ad groups averaged 4/10. The homepage landing page had a generic headline, no mention of company size or industry, and the only conversion action was a contact form buried below the fold. This is a pattern we see repeatedly, and it is well documented in why most B2B landing pages fail to convert paid traffic. The click-to-lead rate was sitting at 1.4%, against a realistic benchmark of 3.5-5% for this category.
Month 1: Restructuring Campaigns and Killing Wasted Spend
The first priority was stopping the bleed. We audited every search term report from the previous six months and built a negative keyword list of 310 terms covering job-seeker intent, free-tool seekers, and irrelevant verticals. Within two weeks, impression share on irrelevant queries dropped by 47% and average CPC fell from £4.80 to £3.60 without touching bids directly. A detailed framework for this process is covered in our guide on eliminating wasted spend with negative keywords.
We also split the single campaign into three separate campaigns by funnel stage: branded, competitor, and non-brand informational versus non-brand commercial. Each got its own budget, bid strategy, and audience layers. Branded campaigns ran Target CPA at a lower target since those clicks converted at 4x the rate of non-brand. Non-brand commercial terms were shifted to maximise conversions with a CPA target set at 20% above the previous average to give the algorithm room to learn without overspending.
Month 2: Landing Page Alignment by Segment
We built three dedicated landing pages tied to the new campaign structure. The non-brand commercial page led with a headline specifically calling out "HR and payroll software for teams of 50-500", included three customer logos from recognisable mid-market brands, and placed a demo booking CTA above the fold. The competitor page used a direct comparison table. Each page had a single conversion path, with the form limited to four fields: name, work email, company size, and current software.
According to Google's research on landing page performance, reducing form fields and improving message match are among the highest-impact conversion levers available. In our case, removing three non-essential form fields alone lifted form completion rate by 34% in the first two weeks of the new pages going live. The click-to-lead rate moved from 1.4% to 3.1% by the end of month two.
Month 3: Tightening Lead Quality with Audience and Bidding Adjustments
Volume was improving but lead quality still needed work. We layered in LinkedIn Matched Audiences via Google's customer match feature, uploading a list of 2,400 existing customers to create a similar audience, then applied a +35% bid adjustment for users matching that profile. We also added negative bid adjustments of -70% for mobile devices on the commercial campaigns, since the client's sales data showed that mobile-originated leads converted to closed deals at less than a quarter of the rate of desktop leads in their segment.
We set up offline conversion tracking by pushing CRM stage data back into Google Ads via the API, so the algorithm was optimising toward SQL (sales qualified lead) events rather than raw form fills. This is the same principle behind the attribution approach we describe in our piece on multi-touch attribution for accurate B2B ROI measurement. Once the algorithm had 30 SQL conversion events to learn from, CPA on non-brand commercial campaigns dropped a further 18% without any manual bid changes.
The 90-Day Results
Across the full 90-day period, total spend held roughly flat at £17,200 per month on average. Cost per lead fell from £340 to £133, a 61% reduction. The qualified lead rate (leads accepted by sales as SQLs) rose from 22% to 58%. Monthly SQL volume went from approximately 12 to 47, a 291% increase on the same budget. The client's sales team closed 11 new accounts in the final 30 days of the engagement, compared to 3 in the equivalent month the previous quarter.
The changes that drove the most impact were, in order: negative keyword hygiene, dedicated landing pages with specific audience messaging, and offline conversion tracking feeding SQL data back to the bidding algorithm. Campaign restructuring mattered but mainly as an enabler for the other two. No new ad formats, no Performance Max, no increases in budget were required to achieve these results.
What This Means for Other B2B SaaS Advertisers
The conditions in this account are not unusual. A single broad campaign pointing to a homepage, with no negative keywords and no landing page specificity, is the default state of a large proportion of B2B Google Ads accounts we audit. The benchmark CPL of £133 for a mid-market HR SaaS product is still above the category median, but it is within a range where the unit economics work for a product with an ACV above £8,000. If your current CPL is more than 3x that figure, the issue is almost certainly structural rather than a budget problem.
The methodology here is repeatable: audit search terms, restructure by intent, build dedicated pages with segment-specific messaging, and close the feedback loop between CRM and bidding. None of these steps require a large budget increase. They require time, rigour, and a willingness to pause spend on keywords that feel important but are not producing revenue.