Paid Media: 5 Moves to Boost ROAS in 2026

For digital advertising professionals seeking to improve their paid media performance, understanding the granular mechanics of a campaign teardown is non-negotiable. We’re not just throwing money at algorithms anymore; we’re surgically dissecting every impression, every click, every conversion. How can you genuinely move the needle for your clients without this forensic approach?

The “GrowthEngine Pro” Campaign Teardown: A B2B SaaS Case Study

I recently led a campaign for “GrowthEngine Pro,” a fictional but highly realistic B2B SaaS platform offering advanced analytics and reporting for marketing teams. Their goal was straightforward: drive qualified leads for their enterprise-tier subscription. We had a healthy budget, but the competitive landscape for B2B SaaS is brutal, demanding precision.

Initial Strategy: Balancing Awareness with Conversion

Our strategy for GrowthEngine Pro wasn’t revolutionary, but it was robust. We aimed for a full-funnel approach, knowing that enterprise sales cycles are long. The plan involved:

• Top-of-Funnel (ToFu): LinkedIn Ads for broad awareness, targeting specific job titles (e.g., “Head of Marketing,” “CMO,” “VP of Analytics”) at companies with 500+ employees. We focused on thought leadership content like whitepapers and industry reports.
• Middle-of-Funnel (MoFu): Google Search Ads for high-intent keywords (“enterprise analytics platform,” “marketing dashboard software comparison”). This was where we expected to capture users actively researching solutions.
• Bottom-of-Funnel (BoFu): Retargeting across Meta Ads and Google Display Network for users who engaged with our ToFu content or visited key pages on the GrowthEngine Pro website but hadn’t converted. This segment received direct demo requests and free trial offers.

Our initial hypothesis was that a strong content play on LinkedIn would fill the pipeline, and high-intent search would seal the deal. Retargeting would act as a safety net, nudging fence-sitters.

Campaign Setup and Initial Metrics

Here’s how we kicked off the GrowthEngine Pro campaign:

The initial CPL was a shocker. We were significantly over our target, and the ROAS was abysmal. This wasn’t just a slight underperformance; it was a flashing red light screaming for intervention.

Creative Approach: What We Thought Would Work

For ToFu, we designed sleek, professional-looking ad creatives featuring data visualizations and bold claims about efficiency gains. Our whitepapers were gated, requiring an email address. MoFu ads were more direct, highlighting specific features and benefits of GrowthEngine Pro, with CTAs like “Get a Demo” or “Start Free Trial.” BoFu creatives used testimonials and urgency, reminding users of the value they’d seen. We used Canva Pro and Adobe Photoshop for all design work, ensuring brand consistency.

Targeting: The Broad Strokes

LinkedIn targeting was extensive: job titles, company size, industry. For Google Search, we started with exact and phrase match keywords, alongside a tightly controlled broad match modifier strategy. Retargeting audiences were segmented by website visits (all visitors, demo page visitors, pricing page visitors) and LinkedIn content engagers.

What Worked, What Didn’t, and the Hard Truths

Here’s the breakdown of our first 30 days:

What Didn’t Work (And Why I Was Kicking Myself)

• LinkedIn ToFu CPL: This was the biggest disappointment. Our CPL on LinkedIn for whitepaper downloads was averaging $180, far exceeding our overall target. The quality of these leads was also questionable; many were junior roles, not decision-makers. My initial assumption was that a high-value asset would naturally filter for quality, but I was wrong. The targeting, while specific, was still too broad for the intent we needed.
• Google Search Broad Match: While we used broad match modifiers, we still saw significant spend on irrelevant searches. Terms like “growth engine for small business” or “free analytics tools” were eating budget without converting. This is where a robust negative keyword strategy is paramount – a lesson I’ve learned repeatedly, yet it always seems to bite you if you’re not vigilant.
• Generic Retargeting: Our BoFu retargeting, while generating some conversions, had a higher CPL ($250) than anticipated. We were showing the same “Request a Demo” ad to everyone who’d engaged, regardless of how they engaged. This lack of personalization was a missed opportunity.

What Did Work (And Gave Us Hope)

• Google Search Exact Match: Our exact match keywords were performing beautifully, with an average CPL of $60 and a strong conversion rate. This confirmed high intent was present for users actively searching for specific solutions.
• Specific Landing Page Performance: The landing page for demo requests, which was designed with clear value propositions and social proof, had a conversion rate of 12% for direct traffic, indicating that when we got the right person there, it worked.

One evening, after reviewing the data, I realized our strategy was too static. We were treating a long sales cycle with short-term, generic tactics. We needed to be more agile, more surgical. This wasn’t a “set it and forget it” campaign, despite what some clients might wish for.

Optimization Steps: Turning the Ship Around

We immediately implemented a series of aggressive optimizations for the remaining 60 days of the campaign.

1. Overhauling LinkedIn Strategy

• Audience Refinement: We tightened LinkedIn targeting to focus exclusively on C-suite and VP-level titles in specific departments (Marketing, Data, Operations) at companies with 1,000+ employees. We also layered in “skills” targeting (e.g., “data visualization,” “marketing automation”) to capture more relevant professionals.
• Content Gating Adjustment: Instead of a whitepaper, our ToFu LinkedIn ads now promoted a live webinar featuring an industry expert, requiring registration. This immediately filtered for higher intent.
• Bid Strategy Shift: Moved from cost-per-click (CPC) to target cost per result (CPR), allowing LinkedIn’s algorithm to optimize for registrations.

2. Google Search: Negative Keywords and Bidding

• Aggressive Negative Keyword Expansion: We added over 300 new negative keywords, focusing on terms related to “free,” “small business,” “personal,” and competitor names (unless specifically targeted). This was a daily task for the first week, then weekly.
• Value-Based Bidding: Switched from maximize conversions to target ROAS bidding for our exact match campaigns. This allowed us to tell Google that not all conversions were equal, prioritizing those with higher potential value.

3. Retargeting Segmentation and Dynamic Creative

• Tiered Retargeting: We split our retargeting audiences into three tiers:
  1. High Intent (Pricing/Demo Page Visitors): Served direct “Request a Demo” ads with aggressive social proof and limited-time offers.
  2. Medium Intent (Other Key Page Visitors/LinkedIn Engagers): Served ads promoting case studies, testimonials, and detailed feature breakdowns.
  3. Low Intent (General Website Visitors): Served brand awareness ads, reminding them of GrowthEngine Pro’s core value proposition.
• Dynamic Creative Optimization (DCO): We implemented DCO on Meta Ads, allowing the platform to dynamically assemble ad variations based on user data. We tested different headlines, body copy, images, and CTAs (e.g., “See How It Works,” “Get a Personalized Demo,” “Download the Report”). This was a game-changer for improving relevance.

Results After Optimization (Remaining 60 Days)

The changes paid off. Here’s how the metrics evolved:

While the overall CPL of $150 was still higher than our initial target of $83.33, the quality of leads had dramatically improved, leading to a much better attributed ROAS. We also saw a significant increase in MQL-to-SQL conversion rates post-optimization, which isn’t directly reflected in ad platform ROAS but was critical for the client.

My biggest takeaway from this turnaround? Never be afraid to admit a strategy isn’t working and pivot aggressively. The data doesn’t lie, and sticking to a failing plan out of stubbornness is the quickest way to burn a client’s budget and trust. I’ve seen it happen countless times, and frankly, I’ve been guilty of it myself in earlier stages of my career. The industry moves too fast for complacency.

For instance, a client last year, a financial tech startup, insisted on running broad match keywords despite mounting evidence of irrelevant traffic. It took a full month of demonstrating wasted spend through detailed search term reports before they agreed to pause those campaigns. Sometimes, you have to be the bad guy to be the effective guy.

Conclusion

Successful paid media performance isn’t about setting and forgetting; it’s about continuous, data-driven optimization. Be prepared to dissect, diagnose, and decisively adapt your campaigns, because the initial plan is rarely the winning one. Your ability to quickly pivot based on real-time data is your most valuable asset. For more insights into maximizing your returns, explore our article on 2026 Marketing ROI Secrets. Additionally, understanding common pitfalls can save you significant budget, as highlighted in Ad Optimization Myths: 5 Truths for 2026.