TLDR;
- Our Strategic Fix: We restructured the entire account to optimise for free trial registrations ('Complete Registration') instead of chasing 'Purchase' conversions that fell outside Meta's 7-day attribution window.
- Verified Results: Generated 4,622 B2B software trial registrations at an average cost of just $2.38 per signup, spending a total of $11,013.
- Main Takeaway: For SaaS with a trial period longer than 7 days, optimising for a higher-volume, top-of-funnel event like a signup is far more effective than trying to make the algorithm track delayed purchases it cannot see.
We recently worked with Feedbackwhiz, a B2B software company that helps Amazon sellers streamline their feedback and review requests. Like many SaaS businesses, they faced a classic, expensive problem with their Meta ads: the gap between someone signing up for a free trial and becoming a paying customer was longer than the ad platform's tracking window.
Most people will tell you to optimise your campaigns for the final, money-in-the-bank conversion. It sounds logical. Why would you pay for anything less than a sale? But for a SaaS product with a 14 or 30-day trial period, that advice is a recipe for burning cash. Meta's algorithm works on a 7-day click attribution window. If a user clicks an ad, signs up for a trial, and then converts to a paid plan on day 15, the algorithm sees nothing. As far as it's concerned, your ad generated zero purchases. It concludes your ads are rubbish, throttles delivery, and sends your costs through the roof.
This is the attribution black hole that swallows SaaS ad budgets. You end up with campaigns that cannot scale, costs you cannot control, and a total lack of visibility on what is actually working. The client was stuck. They knew their product was good, but could not get the ad platform to work with their business model. Our solution was not to find a magic-bullet audience or a unicorn creative. It was to fundamentally restructure their approach, working *with* the algorithm's limitations, not against them. We decided to stop chasing invisible purchases and instead focus the entire strategy on the one conversion event we could track reliably and in high volume: the free trial registration.
The Root of the Problem: Fighting a Battle You Cannot Win
Before we look into what we did, it is worth properly understanding the mess so many SaaS advertisers find themselves in. When you set up a Meta ads campaign and choose 'Purchases' as your optimisation goal, you are giving the algorithm a very specific instruction: "Go and find me people who are most likely to complete a purchase, and show my ads to them." To do this, it needs data. Lots of it. It looks at the characteristics of people who have already purchased and then tries to find more people just like them.
Now, imagine you have a 14-day free trial. A potential customer sees your ad on Monday, clicks through, and signs up. They use the software, see the value, and on the 15th day—well outside Meta's 7-day window—they upgrade to a paid plan. From a business perspective, that is a win. The ad worked. But from Meta's perspective, that purchase never happened. The conversion pixel for 'Purchase' never fired within the attribution window tied to that ad click.
So, the algorithm gets no feedback. It has no data on what a 'purchaser' looks like for your business. It cannot learn, it cannot optimise, and it cannot find you more customers. Your campaign will get stuck in the 'learning limited' phase forever. Your cost per result will show as '—', and your ad spend will deliver a handful of clicks and nothing more. You are effectively flying blind, telling the world's most sophisticated advertising machine to find a needle in a haystack, but you have hidden the needle and set the haystack on fire.
It is common advice you see on forums: "always optimise for the final conversion." It is a fine principle in theory, especially for e-commerce where the purchase happens in minutes, not weeks. But for B2B SaaS, it is a trap. You end up fighting the platform's core mechanics. We knew we had to sidestep this fight entirely. The goal wasn't to get a few expensive, poorly attributed purchases. The goal was to build a predictable system that could acquire trial users at a cost so low that the trial-to-paid conversion rate on the back end would make the entire funnel profitable. The solution was to shift the goalposts.
Our Strategy: Building a Funnel That Speaks the Algorithm's Language
The core of our strategy was simple: stop asking the algorithm to do something it couldn't. Instead of optimising for the 'Purchase' event that was happening weeks down the line, we switched the primary campaign objective to 'Complete Registration'. This was the free trial signup. It was an action that happened immediately after a user clicked the ad, well within the 7-day attribution window.
This single change transformed the campaign. Instead of the algorithm getting zero conversion signals, it was now getting hundreds, then thousands. It could finally learn what a 'trial user' looked like and actively seek out more of them. We were now feeding the machine exactly the data it needed to perform.
However, just changing the objective is not enough. You have to build an entire campaign structure around it. We mapped out a multi-stage funnel designed to guide a potential customer from initial awareness to becoming an active trial user.
SaaS Acquisition Funnel Architecture
Top of Funnel (TOFU): The Engine Room This was where we spent the bulk of the budget. The single goal here was to reach cold audiences—people who had likely never heard of Feedbackwhiz—and get them to sign up for a free trial. All campaigns at this stage were optimised for 'Complete Registration'. This is where we did all our heavy testing on audiences and creatives to find the most efficient pockets of potential customers.
Middle of Funnel (MOFU): The Safety Net Not everyone who clicks an ad is ready to sign up straight away. People get distracted, they want to do more research, or they simply are not convinced yet. The MOFU stage was our safety net to catch these people. We set up retargeting campaigns to show ads specifically to users who had visited the website but had not yet started a trial. The message here could be slightly different—perhaps highlighting a specific feature or a customer testimonial—but the goal was the same: drive them back to complete their registration.
Bottom of Funnel (BOFU): Bridging the Gap to Purchase This is where we addressed the post-signup journey. Once a user was in their free trial, we used a small portion of the budget to retarget them. These ads weren't about getting a signup; they were about user activation and retention. The goal was to keep the product top-of-mind during their trial period, encourage them to use core features, and nudge them towards upgrading to a paid plan. This stage acts as a support system for the client's internal onboarding and email sequences, ensuring our hard-won trial users did not just sign up and disappear.
This full-funnel structure meant we had a specific job for every pound spent. TOFU for acquisition, MOFU for efficiency, and BOFU for activation. And all of it was built around a trackable, high-volume conversion event that the Meta algorithm could actually understand and optimise for.
Execution: A Relentless Process of Testing and Optimisation
A strategy is only as good as its execution. With the funnel structure in place, we began a methodical process of testing audiences, locations, and creatives to find the winning combinations and cut everything that wasn't performing. This wasn’t a 'set and forget' campaign; it was a constant cycle of analysis and adjustment.
First, we attacked targeting. For a B2B SaaS product aimed at Amazon sellers, you cannot just target 'e-commerce' and hope for the best. That is far too broad. You will get a mix of consumers, hobbyists, and business owners, most of whom are not your ideal customer. We had to get much more specific.
B2B SaaS Audience Layering Framework
Layer 1: High-Intent Lookalikes (TOFU)
Our top priority. We built Lookalike audiences from the most valuable custom audiences available: existing paid customers, recent trial signups, and website visitors who completed registration. This tells Meta to find more people who behave just like the ones who already converted.
Layer 2: Problem-Aware Interests (TOFU)
Next, we tested detailed targeting. Instead of broad interests like 'Amazon', we targeted audiences like 'Small Business Owners' and interests related to e-commerce platforms (Shopify, WooCommerce) and Amazon seller tools. These audiences are far more likely to contain our ideal customer.
Layer 3: Retargeting (MOFU & BOFU)
Finally, we built our retargeting pools. This included anyone who visited the website but did not sign up (MOFU), and anyone who had started a trial but had not yet purchased (BOFU). These are warm audiences who already know the brand.
Our testing revealed a few main winners:
- Lookalike Audiences: Despite what some people say, lookalikes can still perform exceptionally well when built from high-quality source audiences. Lookalikes of existing customers, trial signups, and website visitors who had completed registration were consistently strong performers. The algorithm had a clear signal of what a good customer looked like, and it delivered.
- 'Small Business Owners' Detailed Targeting: This audience proved to be a solid, scalable option. For a tool designed to help entrepreneurs on Amazon, this targeting was straightforward and performed reliably.
- What We Binned: We also tested broader e-commerce interests and even broad targeting with no interest layers. These simply didn't work. The cost per registration was too high, and the quality was questionable. We weren't afraid to turn off audiences that were underperforming and reallocate that budget to the winners. It is a simple discipline, but one many advertisers fail to follow.
We applied the same rigorous testing to location targeting. We tested three main setups: US-only, a group of 'Developed Countries', and a broad 'Worldwide' campaign. As expected, the US-only campaigns had the highest costs, while the Worldwide campaign delivered the cheapest registrations. For this campaign, where the primary goal was to acquire a high volume of trial users at the lowest possible CPA, the Worldwide campaign was the clear winner. We did, however, keep an eye on back-end metrics to ensure the quality of signups from different regions wasn't dropping off a cliff. Cheap signups are useless if none of them ever convert to paid.
The final piece of the execution puzzle was the creative. We split-tested a variety of ad copy and image combinations. The copy focused directly on the pain points of an Amazon seller: "Automate your Amazon feedback," "Fire up your Amazon ratings," "Unlock Amazon success." This is not clever, abstract branding; it is direct, problem-solution messaging. The creative itself was clean, showing the product's logo and a clear call-to-action. We systematically turned off underperforming ads and gave more budget to the ones that were delivering the lowest cost per registration. It is a grind, but it is how you turn a good campaign into a great one.
The Results: 4,622 Registrations at $2.38 Apiece
By sticking to this strategy of optimising for the trial signup and relentlessly testing every variable, we were able to build a highly efficient and scalable acquisition engine. The numbers speak for themselves.
Across all campaigns, we generated a total of 4,622 complete registrations for the software's free trial. The total amount spent was $11,013.42, bringing the average cost per registration to an incredibly low $2.38. For a B2B SaaS product, that is an exceptional result.
Campaign Performance Breakdown
A summary of performance across our primary Top of Funnel (TOFU), Middle of Funnel (MOFU), and Bottom of Funnel (BOFU) campaigns.
| Campaign | Registrations | Amount Spent | Cost Per Registration |
|---|---|---|---|
| LH - TOFU (Worldwide) - Complete Registration | 1,083 | $336.75 | $0.31 |
| LH - TOFU (Worldwide) - Trials | 1,013 | $217.57 | $0.21 |
| LH - MOFU - Complete Registration | 1,122 | $987.93 | $0.88 |
| LH - Promotions (Developed Countries) | 419 | $3,890.48 | $9.29 |
| LH - TOFU (Developed Countries) - Complete R... | 232 | $1,337.42 | $5.76 |
| Other Test Campaigns | 753 | $4,243.27 | Avg. $5.63 |
| Total | 4,622 | $11,013.42 | $2.38 |
What this table shows is the power of our testing methodology. The 'Worldwide' prospecting campaigns were unbelievably efficient, bringing in registrations for as little as $0.21 and $0.31. Our MOFU retargeting campaign was also effective, capturing 1,122 registrations from previous website visitors at just $0.88 each.
You can also see the cost difference in targeting developed countries, where the cost per registration was higher, ranging from $5.76 to $9.29. While these users might have a higher likelihood of converting to paid, for the goal of mass trial acquisition, the worldwide campaigns provided far better value. This data allowed the client to make informed decisions about where to allocate their budget. Towards the end of the main push, the client had to reduce their ad spend, but because we had built such an efficient and optimised system, the remaining active campaigns continued to deliver registrations at a great cost.
The Final Piece: Connecting Registrations to Revenue
A $2.38 trial signup is a fantastic metric, but it is not revenue. The job is not done until those trial users become paying customers. This is where the holistic, full-funnel approach becomes so important. Our work on the ad platform was designed to feed the client's business a steady stream of low-cost, qualified leads. What happens next is a partnership between our advertising efforts and their internal processes.
Our BOFU (Bottom of Funnel) campaigns played a supporting role here. By retargeting users who were already in their trial, we kept the product front and centre. These ads acted as reminders, encouraging users to log back in, explore features, and ultimately see the value that would convince them to upgrade. It is a way of using paid ads to support the onboarding process.
However, the real heavy lifting of converting trials to paid customers happens off-platform. It relies on the client having:
- A Solid Product: The software has to actually solve the problem it promises to solve. No amount of clever advertising can fix a bad product.
- Effective Onboarding: A smooth, intuitive onboarding experience is important to guide new users to their "aha!" moment as quickly as possible.
- Automated Email Nurturing: Automated email sequences that teach users, highlight benefits, and address common questions can dramatically increase trial-to-paid conversion rates.
Our strategy was never about just delivering cheap signups in a vacuum. It was about providing the fuel for the client’s own conversion engine. By engineering a system to acquire trial users at $2.38, we gave them a massive advantage. They could be confident that their user acquisition was profitable, allowing them to focus their internal resources on optimising the product and onboarding experience to maximise the lifetime value of each customer we brought in.
Conclusion: Stop Fighting the Platform and Build a System That Works
The biggest lesson from this campaign is that for SaaS advertising on platforms like Meta, common advice is often the worst. Chasing 'Purchase' conversions with a long sales cycle is a fool's errand. You are fighting the platform's fundamental mechanics and setting your campaigns up to fail.
Success comes from a pragmatic approach. You have to understand the limitations of the platform and build a strategy that works within them. By shifting our focus from the final, untrackable purchase to the immediate, high-volume trial registration, we gave the algorithm the data it needed to do its job. We turned an unpredictable, frustrating campaign into a reliable, scalable machine for user acquisition.
This is not just about flicking a switch on the campaign objective. It requires a full-funnel strategy, a relentless testing process, and a deep understanding of how to connect ad performance to real business outcomes. You need to know which audiences to prioritise, how to structure your funnel, what creative will resonate, and how your ads support the entire customer journey, not just the initial click.
If you're running a SaaS business and struggling to make your ads work, there is a good chance you're caught in the same attribution gap. If you are interested in building a predictable system for acquiring customers, we offer a free strategy consultation. We can have a look at your ad account together and figure out the right strategy to get you the results you're looking for.
Cite this Case Study
If referencing this research, data, or strategy in your publication, please use the attribution below.
Lukas Holschuh
Founder, Growth & Advertising Consultant
Great campaigns fail without expertise. Lukas and his team provide the missing strategy, optimizing your entire advertising funnel—from ad creatives and copy to landing page design.
Backed by a proven track record across SaaS, eLearning, and eCommerce, they don't just run ads; they engineer systems that convert. A data-driven partnership focused on tangible revenue growth.