When Client Y, a B2B SaaS tool for internal team communication, came to us, they weren’t lacking effort, they were lacking clarity.

​

Over the past 24 months, they’d run hundreds of ad campaigns, cycled through hundreds of variations, and tested nearly every combination of channel, offer, and creative. But despite all that, growth had flatlined and CAC kept climbing.

​

The real problem?

​

They had no idea what was actually working.

There was no documentation of past tests. No unified learnings. Just scattered ad accounts, burned budgets, and a marketing team that was constantly starting from scratch.

​

Step 1: Build the Learnings Engine

We didn’t rush to launch new ads. We paused - and built the system first.

We audited every historical campaign and tagged it across 5 dimensions:

• Funnel stage

• Offer type

• Creative angle

• Channel

• Audience

​

Then we built a centralized learnings database, so every test going forward would compound insight, not waste spend.

​

From there, we launched tight A/B tests in 7-day cycles, each one mapped to a specific hypothesis:

• Does urgency outperform social proof?

• Do product screenshots beat founder-focused creatives?

• Should we lead with “async productivity” or “calendar freedom”?

​

Each test added a brick to the system. And the results started to snowball.

Within 60 days:

• CTR improved by 37%

• CPL dropped by 22%

• We reactivated a cold LinkedIn audience that had been dismissed months ago

​

But that was just the top of the funnel.

​

Step 2: Fix the Sales Bottleneck

As we followed leads through the funnel, we found another issue: low ACV leads were going through the exact same sales process as enterprise ones.

​

Every demo, even for a $49/month account, was eating SDR time and clogging the pipeline.

​

So we built a self-serve flow specifically for smaller accounts:

• Replaced demos with targeted landing pages + recorded walkthroughs

• Added chat-based support with guided onboarding

• Introduced a “start now” CTA with Stripe checkout tied to feature limits

​

The impact was immediate:

• Self-serve signups increased 73%

• SDR bandwidth freed up for true mid-market and enterprise leads

• Speed to revenue improved by 12 days

​

Step 3: Stack the Wins

Over 6 months, we kept compounding what worked and ditching what didn’t. The database grew. The ads got sharper. The funnel got cleaner.

​

By month three:

• CAC had dropped from $295 to $159 - a 46% reduction

• Lead-to-paid conversion rose from 4.8% to 7.1%

• Self-serve MRR jumped from $22K to $38K/month

• Payback period on paid acquisition dropped below 30 days