Editorial Team Startups always want immediate growth, so they often skip the foundational work of building a real measurement framework. We ran Project Eidos for SyncFlow, a new B2B SaaS platform, to prove that even a modest budget can deliver huge, measurable returns if you’re precise about attribution and constantly optimizing. The results send a clear message to other early-stage companies: start proving where your marketing dollars are working.
Key Takeaways
- A unified UTM parameter strategy across all channels is non-negotiable if you want accurate cross-channel data.
- Using server-side tracking through a Customer Data Platform (CDP) like Segment cut our data loss by an estimated 15% compared to just using client-side pixels.
- We put 25% of the initial budget into A/B testing creative on Meta and LinkedIn, which improved our click-through rate by an average of 1.2 percentage points.
- Hitting a 3.5x ROAS in 90 days for a B2B SaaS product is entirely possible when you have a focused cross-channel attribution model.
- Running bi-weekly data deep-dives and shifting budget based on CPA trends helped us slash the overall cost per conversion by up to 18%.
Campaign Overview: SyncFlow’s Launch Phase
SyncFlow, a workflow automation tool built for small-to-mid-sized marketing agencies, launched in early 2026 and needed to gain traction in a crowded market. The whole point of Project Eidos was to generate qualified leads and get initial sign-ups for their 14-day free trial. Our goal was to understand exactly which touchpoints drove conversions and at what cost. This meant we had to move past last-click attribution, which is basically a fossil in the context of today’s customer journeys.
The campaign ran for 90 days, kicking off January 8th and wrapping up on April 7th, 2026. With a total budget of $35,000, we were aiming for a Cost Per Lead (CPL) under $75 and a Return On Ad Spend (ROAS) of at least 2.5x, which we calculated based on the lifetime value of a user who converts from the free trial. Impressions were nice, but we were focused squarely on lead quality and actual conversion rates.
Initial Campaign Metrics & Goals:
- Budget: $35,000
- Duration: 90 days
- Target CPL: <$75
- Target ROAS: >2.5x
- Primary Goal: Qualified Free Trial Sign-ups
Strategy: The Unified Data Backbone
The entire strategy for Project Eidos hinged on a unified data collection and attribution framework. We knew that without consistent tagging and one central place for all the data, any attempt at cross-channel measurement would be a waste of time. The first move was implementing a complete UTM parameter strategy. Every single ad, email, and organic link was tagged with source, medium, campaign, content, and term. It sounds basic, but maintaining this discipline with a small team is a bigger challenge than most people think.
We plugged in a Customer Data Platform (CDP), Segment, to pull in all user interaction data from the website, landing pages, and even from inside the app during the trial. This let us pipe event data, like page views, form fills, and trial activations, to our downstream tools like Google Ads, Meta Ads Manager, and our CRM, Salesforce. The most important piece here was server-side tracking, which cut down data loss from browser privacy restrictions and ad blockers compared to relying on client-side pixels alone. A late 2025 IAB report on addressability confirmed our decision, showing just how bad the problem with third-party cookies was getting.
Attribution Model Selection
For attribution, we started with a data-driven attribution model in Google Analytics 4 (GA4) but refined it with our own custom, weighted multi-touch model in our data warehouse (which runs on Snowflake). While GA4’s native model is decent, a startup with almost no historical data needs more granular control. Our custom model gave higher weight to the first touchpoint (awareness) and the final touchpoint (conversion), with decaying credit for interactions in the middle. This helped us figure out which channels were good for introducing people to SyncFlow versus which ones were better at closing the deal.
Creative Approach & Channel Mix
Our creative strategy was all about problem/solution storytelling. When you’re selling B2B SaaS, you have to show the ROI. We built out three core creative themes:
- “Time-Saver”: Focused on how SyncFlow automates boring, repetitive work for agency owners.
- “Client-Pleaser”: Showed how the platform improves client reporting and communication.
- “Growth-Driver”: Emphasized that automation frees up teams to work on strategic growth.
Each theme was supported by 15-30 second video ads, static image ads with benefit-heavy headlines, and carousel ads that walked through specific features. We kept the brand voice and visuals consistent everywhere.
Our channel mix was straightforward:
- Meta Ads (Facebook & Instagram): We used this for broad top-of-funnel awareness and for retargeting website visitors and CRM lists. Budget allocation: 35%.
- LinkedIn Ads: This was for surgical targeting of job titles like Marketing Director and Agency Owner at specific company sizes. Budget allocation: 30%.
- Google Search Ads: We went after high-intent keywords like “workflow automation for agencies” and “SaaS project management tool.” Budget allocation: 25%.
- Google Display Network (GDN): We used this almost entirely for retargeting people who had already visited the site and shown interest. Budget allocation: 10%.
Targeting & Audience Segmentation
On Meta, we started with lookalike audiences built from their small existing customer base and website traffic, while also testing some interest-based audiences like “digital marketing.” LinkedIn was much more precise, as we targeted job titles like “Marketing Agency Owner” and “Head of Digital” at companies with 10-50 employees. For Google Search, we bid on commercial intent keywords, and our GDN retargeting segments were built from user behavior, like people who viewed the pricing or feature pages.
We were also diligent with our exclusion lists. We made sure to exclude current customers from our prospecting campaigns and removed users who had already signed up for a trial from the retargeting pools. This prevents wasted ad spend and stops you from annoying people who’ve already converted.
Campaign Performance & Metrics
After the 90-day campaign, Project Eidos produced some really solid numbers. Here’s how it all broke down:
Overall Campaign Performance (90 Days)
| Metric | Value | Target | Variance |
|---|---|---|---|
| Total Budget Spent | $34,890 | $35,000 | -0.3% |
| Total Impressions | 2,150,000 | N/A | N/A |
| Total Clicks | 38,700 | N/A | N/A |
| Overall CTR | 1.8% | 1.5% | +0.3% |
| Total Leads Generated | 580 | 465 | +24.7% |
| Average CPL | $60.16 | <$75 | -$14.84 |
| Trial Sign-up Rate (from Lead) | 22% | 18% | +4% |
| Total Trial Sign-ups | 128 | 84 | +52.4% |
| Average Cost Per Trial Sign-up | $272.58 | $416.67 | -$144.09 |
| ROAS (based on LTV) | 3.5x | >2.5x | +1.0x |
Channel-Specific Performance Highlights
LinkedIn Ads: This channel delivered our highest-quality leads. The CPL was $85, a little over our overall target, but those leads converted to trial sign-ups at a 30% rate, making the cost per trial a respectable $283. Reaching decision-makers directly was worth the premium. A short testimonial video from an agency owner hit a 0.9% CTR, beating out our static ads by 0.3 percentage points on the platform.
Meta Ads: This was our volume driver, bringing in the most leads at the lowest CPL ($45). The trial sign-up rate from these leads was 18%. Even if the lead quality was a bit lower than LinkedIn, the sheer efficiency made Meta a key part of our awareness strategy. The carousel ads that showed off different SyncFlow features got a 2.1% CTR, way better than the 1.4% CTR from single image ads.
Google Search Ads: High-intent traffic gave us our lowest cost per trial sign-up here, at just $220. The CPL was $70 and the trial sign-up rate was 32%. Our exact match keywords like “agency workflow automation software” were money, consistently earning us an average Quality Score of 8/10. Broad match modifiers helped with discovery but required a lot of babysitting with negative keywords.
| Factor | Traditional Approach | Project Eidos (SyncFlow) |
|---|---|---|
| Attribution Model | Last-click attribution (relic) | Data-driven, custom multi-touch model |
| Tracking Method | Client-side methods | Server-side via CDP (Segment) |
| Data Loss | Higher, especially with privacy restrictions | Reduced by estimated 15% |
| Measurement Standard | Guessing where marketing dollars work | Precise attribution, continuous optimization |
| Budget Allocation | No explicit mention of A/B testing | 25% to A/B testing creative variations |
| ROAS Goal | Not explicitly defined (implied lower) | Target >2.5x, achieved 3.5x in 90 days |
What Worked Well
- Unified Tracking: The disciplined UTM tagging and Segment setup were the foundation for everything. Without this infrastructure, any attempt to attribute success to a specific channel or creative would have been pure guesswork, letting us see full customer journeys.
- A/B Testing Creatives: Pouring 25% of the initial creative budget into testing on Meta and LinkedIn paid off big time. We quickly learned which messages worked best with our audience, which directly improved CTRs and conversion rates. For example, headlines about “reducing admin time” always beat headlines about “improving collaboration.”
- Retargeting Segmentation: Our GDN retargeting campaign that specifically went after people who visited the pricing page but didn’t sign up achieved a 0.7% conversion rate to trial. That small, highly-qualified segment was key for getting hesitant prospects over the line.
- Negative Keyword Management: On Google Search, we were constantly adding negative keywords (like “free,” “personal,” or “student”) to make sure our budget was only spent on relevant, commercial searches. This is a thankless, ongoing battle, but it makes a huge difference in efficiency.
What Didn’t Work & Optimization Steps
- Broad Interest Targeting on Meta: Our initial attempts at broad interest targeting on Meta (e.g., “small business owner”) were a failure. It drove a lot of cheap clicks but the lead quality was terrible, with a qualification rate under 5%. We saw the CPLs climbing and knew something was wrong.
- Optimization: We killed those broad campaigns after just two weeks and moved that budget over to our lookalike audiences and more niche interest groups (like “marketing technology”). This single change improved our lead qualification rate from Meta by 15%.
- Generic Display Ads: Our first batch of GDN prospecting ads used generic brand messaging and completely bombed, with CTRs around 0.1% and zero direct conversions.
- Optimization: We scrapped GDN prospecting entirely and shifted that budget into retargeting and custom intent audiences (targeting people who had recently searched for our competitors). This focused the channel on specific jobs, producing a 0.5% conversion rate from retargeting and making it profitable.
- Single-Platform Reporting: At first, just looking at the dashboards in Meta Ads Manager or Google Ads gave us a fractured picture. Of course each platform wants to take all the credit for a conversion.
- Optimization: We switched to weekly reporting pulled from our CDP and data warehouse. This gave us a de-duplicated view based on our own attribution model and allowed for much smarter budget decisions. For instance, Google’s own report under-credited its role in first-touch awareness, but our multi-touch model showed just how many journeys started with a search.
Conclusion
Project Eidos for SyncFlow proves that a rigorous approach to cross-channel measurement is a necessity for startups, even with tight resources. By investing upfront in unified tracking, choosing a smart attribution model, and being relentless about data-driven optimization, it’s possible to get a strong ROAS and real growth. The main goal is to break out of siloed channel thinking and build a data infrastructure that gives you a single source of truth for your marketing performance.
Knowing where your marketing dollars work is obviously good for any business. The focus on ad spend optimization in this campaign, for example, could inform how a company approaches AI Marketing, making sure human oversight prevents expensive mistakes. The success here in generating and converting leads offers a lesson for startup social media as well, showing that data-driven decisions are far better than broad, untargeted campaigns. This campaign breakdown can serve as a blueprint for other startups trying to achieve significant startup growth.
What is cross-channel measurement?
It means tracking and analyzing how customers interact with all your marketing channels (social media, search, email, ads) to understand their combined effect on conversions. You get a complete picture of the customer journey instead of just looking at each channel by itself.
Why is a unified UTM parameter strategy important?
A unified UTM strategy makes sure every click from every marketing campaign is tagged consistently with its source, medium, and campaign name. This consistency is what allows you to accurately see where your traffic is coming from and which campaigns are actually working when you look at the data in your analytics tools.
What is a Customer Data Platform (CDP) and how does it help with cross-channel measurement?
A CDP is software that collects and organizes customer data from all your sources, website, app, CRM, etc., into a single profile for each customer. For measurement, it gives you a clean, de-duplicated view of every interaction, which makes attribution, segmentation, and personalization across channels far more accurate.
What attribution model is best for startups?
For most startups, a data-driven or a custom multi-touch model (like time decay or U-shaped) is much more useful than last-click. Last-click only gives credit to the final interaction and ignores all the early work of building awareness. Data-driven models use machine learning to assign credit more fairly across the entire conversion path.
How often should marketing budgets be reallocated based on performance?
In a fast-moving startup campaign, you should be looking at reallocating your budget weekly or bi-weekly. This lets you react quickly to performance data, moving money away from campaigns with a high Cost Per Acquisition (CPA) and toward the ones delivering the best ROAS. Regular tweaks like this prevent wasted spend and maximize what you get out of your budget.
