When a US-based project management SaaS came to us in Q4 2025, their instinct was straightforward: spend more on Google Ads and LinkedIn to grow the pipeline. Their Google Ads CPL sat at $312, their LinkedIn CPL at $480, and their sales team was closing roughly 8% of all inbound leads. Before we touched a single bid or budget, we ran a full attribution audit, and what we found changed the entire strategy.
The Core Problem: Scaling a Broken Measurement System
The client was running last-click attribution across both Google Ads and their CRM. This meant Google Search was receiving credit for almost every closed deal, while LinkedIn Ads looked like a loss-maker on paper. In reality, LinkedIn was the first touchpoint for 61% of their closed-won accounts, but that data was invisible to anyone looking at the channel-level CPL report.
The practical consequence was dangerous: the team was about to cut LinkedIn spend by 40% and reallocate it to branded search, a channel that was already converting well but had limited incremental volume. Scaling branded search while cutting the top-of-funnel channel that was seeding it would have compounded the damage over two to three quarters. Before advising on any budget shift, we needed the attribution layer to reflect reality. For a deeper explanation of why this distortion is so common in B2B funnels, see our guide on multi-touch attribution and B2B ROI.
What the Attribution Fix Actually Involved
We implemented a linear multi-touch model using HubSpot as the source of truth, with UTM parameters enforced at every ad level across both platforms. Every lead source was mapped back through the full contact timeline, not just the most recent campaign interaction. We also set up offline conversion imports into Google Ads using the HubSpot-to-Google Ads integration, so that SQL (sales-qualified lead) status and closed-won events fed directly back into the bidding algorithm within 72 hours of the conversion occurring.
The technical setup took approximately three weeks: one week to audit and fix broken UTM parameters across 140 active ad variants, one week to configure the CRM pipeline stages correctly, and one week to validate that data was flowing accurately before we touched any campaign settings. This delay frustrated the client initially, but it was non-negotiable. Feeding a smart bidding algorithm with accurate conversion signals is the single highest-leverage action in any paid search account, as Google's own smart bidding documentation makes clear when explaining the volume and quality requirements for Target CPA to function correctly.
Campaign Changes Made After Attribution Was Fixed
With clean data in place, the campaign changes were relatively straightforward. On Google Ads, we switched from maximise conversions (which had been optimising toward form fills, many of which were low-quality) to Target CPA optimising toward SQL conversions, with a target of $180. On LinkedIn, we shifted budget from broad job-title targeting to a tighter account-based audience of 2,200 named accounts provided by the sales team, paired with single-image ads leading to a case study landing page rather than a generic demo request form.
We also ran a negative keyword overhaul on the Google Ads account, removing 318 irrelevant search terms that had been generating clicks from SMB job seekers and students. If your account has a similar problem, the process we use is covered in detail in our article on eliminating wasted spend with negative keywords. Within the first 30 days post-fix, the Google Ads cost per SQL dropped from $312 to $198, a 37% reduction before any additional budget was added.
Results at the 90-Day Mark
By the end of Q1 2026, the combined blended CPL across Google Ads and LinkedIn had fallen from an average of $396 to $182, a 54% reduction. The close rate from inbound leads improved from 8% to 13%, partly because the leads were better qualified and partly because the sales team now had accurate first-touch data telling them which content a prospect had engaged with before requesting a demo. Total pipeline generated in Q1 2026 was $2.1M, compared to $940K in Q4 2025, with the same overall ad spend of approximately $38K per month.
It is worth being precise about what drove this result. The attribution fix did not generate a single extra lead on its own. What it did was stop the algorithm from optimising toward the wrong signal, stop the team from misreading channel performance, and allow budget to flow to the touchpoints that were actually driving revenue. The creative and targeting changes amplified an already-improving system rather than carrying the full weight of the improvement.
What This Pattern Looks Like in Other Accounts
This scenario, last-click attribution hiding the true value of a demand-generation channel while over-crediting a closing channel, appears in roughly 70% of the B2B paid accounts we audit at the growth stage. It is especially common in companies running both paid search and paid social simultaneously without a CRM integration that ties the two together. The symptom is usually a CPL that looks reasonable at the channel level but a close rate or pipeline quality that consistently disappoints the sales team.
If your Google Ads account is generating clicks but not pipeline, attribution is often a more likely culprit than ad copy or targeting. For a full breakdown of the structural issues that cause this, our article on why Google Ads don't generate quality leads covers the most common root causes we see across B2B accounts in the US and EU markets. Fixing measurement before scaling spend is not a conservative approach; it is the highest-return action available in most growth-stage B2B ad programmes.
Key Takeaways From This Engagement
- Audit attribution before diagnosing channel performance. Last-click models will mislabel your best channels as underperformers.
- Offline conversion imports into Google Ads are the single biggest lever for improving smart bidding accuracy in a B2B account.
- UTM hygiene across 100% of ad variants is a prerequisite, not an optional extra. One broken UTM corrupts an entire contact's attribution history.
- A tighter audience on LinkedIn (named account lists) consistently outperforms broad job-title targeting when the ICP is well defined.
- The 54% CPL reduction came from data integrity and targeting precision, not from increasing budget or rewriting ad creative.