How to Analyse Paid Media Performance Beyond In-Platform Data in B2B & B2C Marketing
In this episode of The Revenue Room, we break down a real-world paid media experiment that challenges one of the most common assumptions in marketing: spend more equals more growth.
After increasing Meta ad spend by 25%, revenue only grew marginally - and media efficiency dropped. On the surface, platform performance looked strong. But once we analysed the data across CRM, GA4, and attribution layers, it became clear that the issue wasn’t acquisition - it was something much deeper.
This episode is designed as a practical podcast for marketing teams, founders, and operators looking to scale paid media without losing efficiency. We walk through how to properly analyse performance across channels and why relying on in-platform metrics alone can lead to the wrong conclusions.
In this episode, we cover:
Why increasing ad spend doesn’t scale revenue linearly
How to use media efficiency ratio (MER) to evaluate true performance
The impact of retention, repeat purchase rate, and lifetime value on growth
Why revenue lag can distort performance in longer sales cycles
What incrementality data reveals about wasted spend
How to diagnose performance across Meta, Google, GA4, and CRM data
Why marketers must take a top-down approach to analysis, not just channel-level reporting
If you’re running paid media, managing marketing budgets, or trying to improve business performance, this episode will help you think more strategically about where your growth is actually coming from.
Transcript
The Revenue Room – Episode 019 Highlights
Hosts: Harry Hughes & Dan
Topic: A live client test where Meta spend went up 25% and revenue rose by just under 5%, what the analysis behind that uncovered, and why paid media performance has to be diagnosed from the top down using more than in-platform data.
The Test: 25% More Spend, 5% More Revenue
Harry walks through a test run with a bookings-based client over roughly two sales cycles.
Meta spend increased by around 25%
Revenue grew by just under 5%
Media efficiency ratio (MER) declined, sitting just above break-even ROAS
In platform, everything looked healthy: more bookings and a falling CPA. The team spent two weeks pulling together every data source they could to find out why efficiency had dropped.
Finding 1: A Retention Problem, Not an Acquisition Problem
The business serves two customer types, roughly 50/50 by revenue. For customer type A, two years of data showed:
Booking volume had stayed within the same range
Revenue had grown only because average order value had increased
Repeat purchase rate was declining
Bookings attributed to direct were falling, another sign of fewer returning customers
New customers were coming in, but existing ones were leaving at a faster rate. Acquisition wasn’t broken. Retention was.
Finding 2: Revenue Lag
Customer type B has a much longer sales cycle. That side had grown by 34%, but the revenue needed another two to four months to land. Reviewed six months later, the same period would show a healthier MER.
Finding 3: Not Every Booking Was Incremental
The team layered several views on top of each other: bookings, click attribution CPA, their standard settings (7-day click and 1-day view for prospecting, 7-day click for retargeting) and Meta’s incremental attribution.
Between 5 and 10% of the bookings Meta reported weren’t incremental
Extra spend on customer type A may not have been driving extra bookings
What Happens Next
Wait three to four months for the lagged revenue to land, and compare it with the forecast
Run a full retention analysis on customer type A
Reduce Meta spend on customer type A by 10 to 15% and see whether back-end volume falls by the same amount
Don’t Be Steered by Platform Reps
Dan’s related example: an audit of a Series D business putting 150k to 200k a month through paid, mostly Google, after being encouraged to run full funnel across several markets.
Display showed around 60,000 clicks in platform, against 5,000 to 7,000 paid users actually landing on site
Month to month, spend couldn’t be matched to any movement in site traffic
No UTM data was being stored in the CRM, so sign-ups couldn’t be traced to a source
The fix was to strip it back: switch display off, return to high-intent search, check whether registrations and revenue moved (they didn’t), then scale back up based on evidence.
His wider point is that a raise is not a reason to pour budget into a channel. Ask what you would do without the money: test one channel and one ICP, then scale gradually.
Proving LinkedIn’s Impact
When attribution gets lost in direct and organic, Dan suggests two things:
Export the list of companies engaging with your paid and organic content, as early evidence you are reaching the right accounts
Run a geo holdout: activate a new state or market, for example with a webinar, and measure sign-ups and downstream actions there over the following six months
Always Analyse Top Down
Harry’s method starts with marketing spend against revenue, then works downwards.
Has MER improved or declined?
Then by channel, comparing attribution settings
Then GA4 sources over time
Then CRM and UTM data
Going straight to channel level produces wrong answers. A 40% increase in LinkedIn thought leader spend won’t show as LinkedIn conversions. It tends to show up later in Google, organic and direct.
Cast the Net Wider Than Your Channels
A drop in MER isn’t always a media problem. Layer in:
PR, trade shows and out-of-home activity in the comparison period
Sales conversion rates and lead quality
New competitors or pricing pressure
Retention
Dan describes doing this at their own startup: knowing the baseline traffic, then measuring the uplift each time a new activity such as PR or paid social was added.
Harry’s advice is to go in open-minded, even though the budget is yours to defend, and to look at a couple of years of data, not six months.
Key Takeaways
More spend doesn’t scale revenue in a straight line
Track MER over time, not just platform CPA
Check retention and revenue lag before blaming acquisition
Compare attribution settings, including incremental
Treat platform rep advice with caution
Start from the top and work down
Store UTM data in your CRM so you can trace what is working