B2B SaaS Teardown: What Drove 2.3x ROAS (and What Didn’t)

For and marketing professionals, understanding exactly what makes a campaign soar or sink is paramount. We offer practical guides on content marketing, marketing analytics, and everything in between, because seeing real-world data and dissecting strategy is the only way to genuinely learn. Today, we’re pulling back the curtain on a recent campaign we managed for a B2B SaaS client – a deep dive into the numbers, the creative, and the brutal reality of what worked and what absolutely did not. Ready to see how the sausage is made?

The Campaign Teardown: “AI-Powered Automation for SMBs”

I’ve been in this game for over a decade, and one thing remains constant: every campaign, no matter how meticulously planned, has its surprises. This particular campaign, launched for “AutomateNow AI” – a B2B SaaS platform specializing in AI-driven workflow automation for small to medium-sized businesses – was no exception. Our objective was clear: drive qualified leads for their flagship product, increasing demo sign-ups and ultimately, new customer acquisition.

Initial Strategy & Budget Allocation

Our strategy hinged on a multi-channel approach, focusing on platforms where SMB decision-makers spend their time. We allocated our budget across LinkedIn Ads for B2B targeting precision, Google Ads for intent-based searches, and a smaller slice for programmatic display to build brand awareness. The total campaign budget was $75,000 over a 90-day duration.

Here’s the initial breakdown:

• LinkedIn Ads: 50% ($37,500)
• Google Search Ads: 35% ($26,250)
• Programmatic Display (via The Trade Desk): 15% ($11,250)

Our initial hypothesis was that LinkedIn would deliver the highest quality leads, given its robust professional targeting capabilities. Google Search would capture immediate intent, and display would serve as a top-of-funnel awareness driver, nurturing prospects who might not yet be actively searching.

Creative Approach: Show, Don’t Just Tell

For AutomateNow AI, the core challenge was demonstrating the tangible benefits of AI automation without getting bogged down in technical jargon. Our creative team, which I personally oversee, focused on a “problem-solution-outcome” narrative. We developed several ad variations:

• LinkedIn Carousel Ads: These were our workhorses. Each slide depicted a common SMB pain point (e.g., “Drowning in manual data entry?”), followed by the AutomateNow AI solution, and then a clear outcome (e.g., “Save 10+ hours/week!”). We used short, punchy copy and animated GIFs that showed the software in action – a critical element, in my opinion.
• Google Search Ads: Standard text ads, tightly focused on high-intent keywords like “AI automation for small business,” “workflow automation tools,” and “CRM integration AI.” We experimented with dynamic keyword insertion and various call-to-actions (CTAs) like “Get a Free Demo” or “See How We Save You Time.”
• Programmatic Display Ads: Primarily static banner ads and short HTML5 videos. These were more brand-focused, highlighting key benefits and a strong brand identity.

We specifically designed landing pages for each channel, ensuring message match. The LinkedIn ads drove traffic to a landing page featuring a short video demo and a prominent demo request form. Google Ads led to pages with case studies and detailed feature breakdowns. The display ads, being more top-of-funnel, often directed users to a blog post or a gated whitepaper.

Targeting Precision: Who Were We After?

This is where the rubber meets the road. For LinkedIn, we targeted:

• Job Titles: Business Owner, Operations Manager, Office Manager, CEO, Founder, Head of Sales (for SMBs).
• Company Size: 1-200 employees.
• Industries: Professional Services, Marketing & Advertising, E-commerce, Healthcare (non-clinical roles), Financial Services.
• Skills: Business Process Automation, Digital Transformation, CRM, Project Management.

On Google, our targeting was keyword-driven, but we also applied audience layering using in-market segments for “Business Software” and “Marketing & Advertising Services.” For programmatic display, we used lookalike audiences based on existing customer data, combined with firmographic targeting similar to LinkedIn.

What Worked: The Unexpected Wins

The LinkedIn Carousel Ads with animated GIFs were an absolute powerhouse. Their performance blew our initial projections out of the water.

That 1.8% CTR on LinkedIn is phenomenal for B2B. Typically, I’d be happy with 0.8-1.0% on cold audiences. The animated GIFs truly made the difference, cutting through the noise in a way static images couldn’t. Users could immediately grasp the product’s functionality without leaving their feed. We saw a 45% higher CTR on these dynamic ads compared to their static counterparts.

Google Search Ads also performed admirably, particularly for branded and high-intent keywords. Our average CPL for Google was around $90, slightly better than LinkedIn initially, but the volume was lower. The prospects coming from Google were often further down the funnel, exhibiting stronger purchase intent.

What Didn’t Work: The Hard Lessons

Our programmatic display campaign was, frankly, a bit of a dud. Despite using lookalike audiences, the CPL was significantly higher, nearing $250 for a demo sign-up. The quality of leads was also questionable; many were early-stage researchers rather than decision-makers ready for a demo. It felt like we were shouting into the void, rather than having a targeted conversation. This is where I’ll offer an editorial aside: don’t chase impressions just for vanity. If your display spend isn’t generating measurable, high-quality leads, cut it. Fast. It’s a common trap in marketing, especially for those new to the B2B space.

Another area of concern was the cost per qualified lead (CPQL). While our initial CPL for demo sign-ups was $125 on LinkedIn, a significant portion of these leads weren’t truly “qualified” by the sales team’s standards – meaning they didn’t meet specific revenue, employee count, or immediate need criteria. Our CPQL, when factoring in the sales team’s disqualifications, was closer to $300 initially. This was a brutal wake-up call.

Optimization Steps: Course Correction in Real-Time

We didn’t just sit there and watch the budget burn. Here’s how we iterated:

1. Aggressive A/B Testing on LinkedIn: We continuously tested headlines, ad copy, and even different animated GIF sequences. We found that questions directly addressing pain points (“Is manual data entry draining your team’s productivity?”) performed better than declarative statements. We also introduced a new ad variation showcasing a specific ROI metric from a fictional client case study, which boosted CTR by another 10%.
2. Landing Page Overhaul: Recognizing the qualification issue, we added more explicit qualification questions to our demo request forms, such as “Company Revenue” and “Number of Employees.” This immediately filtered out some unqualified leads, increasing the quality of submissions, even if it slightly reduced the raw number of conversions. It dropped our CPL slightly, but our CPQL significantly.
3. Refined Google Ads Keywords: We aggressively pruned underperforming keywords, focusing more on long-tail, highly specific phrases. For example, “AI automation for accounting firms” performed far better than just “AI automation.” We also increased bids on keywords that consistently delivered high-quality leads.
4. Retargeting with Testimonials: We paused the broad programmatic display campaign entirely. Instead, we reallocated that budget to a highly targeted LinkedIn and Google Display Network retargeting campaign. This campaign specifically showed ads with strong customer testimonials and case studies to anyone who had visited our landing pages but hadn’t converted. This strategy was a game-changer. Our retargeting CPL was $60, and the CPQL for these warm leads was a fantastic $150 – a 22% reduction from our initial cold CPQL.
5. Sales-Marketing Alignment: We instituted weekly syncs with the sales team. Their feedback on lead quality was invaluable. They helped us understand nuances in job titles and company types that were truly ready for our solution, allowing us to refine our LinkedIn targeting even further. For instance, we discovered that “Operations Coordinator” often led to dead ends, while “Director of Operations” was a goldmine.

Final Performance Metrics (After Optimization)

The final ROAS of 2.3x was a solid win for AutomateNow AI, especially considering the B2B SaaS sales cycle. We not only hit our lead generation targets but significantly improved the quality of those leads over the campaign’s duration. This campaign reinforced a core belief of mine: consistent, data-driven optimization is more important than a perfect initial plan. You simply cannot predict every variable.

A quick anecdote to drive this home: I had a client last year, a logistics software provider, who insisted on using only static images because “video production was too expensive.” We finally convinced them to test some simple screen-capture GIFs demonstrating their product’s UI. The engagement metrics soared. Sometimes, the simplest visual elements have the most profound impact. Always test your assumptions!

One final thought on reporting: we used Google Looker Studio (formerly Data Studio) for our client dashboards, pulling data directly from LinkedIn Campaign Manager and Google Ads. This allowed for real-time visibility and fostered trust, as they could see our adjustments and their impact firsthand. Transparency in reporting is non-negotiable.

This campaign, while successful, also highlighted the evolving landscape of B2B marketing. The increasing sophistication of AI-powered tools means that generic targeting and static creative are no longer enough. You need to speak directly to pain points, demonstrate value immediately, and be prepared to pivot your strategy based on hard data. There’s no room for guesswork.

The key takeaway from this campaign teardown is simple: relentless iteration fuels success. Even with a well-researched strategy, the market will always throw curveballs, so build agility into your marketing process from day one.