You Killed Your Best Marketing Channel. The Dashboard Told You To.

Here's a scene we've watched play out a dozen times.

A business owner opens their marketing dashboard. Google Ads: $28 cost per lead. Facebook Ads: $45 cost per lead. The decision writes itself. Cut Facebook. Shift the budget to Google. The spreadsheet says it's obvious.

Three months later, Google Ads CPL has climbed to $52. Branded search volume has dropped 20%. The pipeline is thinner. And nobody can explain why, because the channel they cut was "underperforming."

It wasn't underperforming. It was doing the most important job in the entire marketing system. The dashboard just couldn't see it.

The Invisible Subsidy Your "Worst" Channel Provides

Marketing channels don't operate in isolation. They feed each other. And the channel that looks worst on a last-click dashboard is often the one keeping everything else alive.

The specific number is damning: 30% of paid search clicks are driven by other forms of advertising, most of which are upper-funnel brand and video budgets. Your Google Ads aren't just capturing demand from thin air. They're catching demand that Facebook, email, and content created.

Think of it this way: Google Ads is a net at the bottom of a waterfall. Facebook Ads is the rain that feeds the river. Cut the rain, and the waterfall dries up. But if you're only measuring the net, you'll never see the connection.

Why Your Dashboard Is Structurally Designed to Mislead You

The problem isn't bad data. It's that last-click attribution answers a question nobody should be asking: "Which channel did the customer click last?"

That question ignores everything that happened before. And everything before is where the actual persuasion occurred.

Here's how severe the distortion is. Analytic Partners' data, reported through Mi3 Australia, found that 88% of social media impressions aren't captured by traditional attribution models. Not 10%. Not 30%. Eighty-eight percent. Meanwhile, search ROI is overvalued by up to 336% compared to marketing mix modelling.

That's worth sitting with. Your dashboard says Google delivers 3x the return of Facebook. The actual data says Google is getting credit for work Facebook did.

Google's own research describes the 7-11-4 rule: consumers typically need 7 hours of engagement, across 11 touchpoints, in 4 different locations before buying. Last-click attribution credits only the final one. The other 10 touchpoints? Invisible. Unfunded. Eventually, cut.

The Brand Doom Loop (and Why 84% of Businesses Are Stuck in It)

Les Binet and Peter Field analysed 996 IPA Effectiveness Awards case studies spanning 30 years of data. Their finding: the optimal budget split is roughly 60% brand building, 40% sales activation.

But here's what matters for this article: they also found that shifting all spending to activation costs over 50% of total effectiveness. Not 10%. Not 20%. Half your marketing effectiveness disappears when you stop building the brand.

The reverse isn't true. Shifting entirely to brand only costs about 20% of effectiveness. The risk is asymmetric: cutting brand is far more destructive than cutting activation. Yet when budgets get tight, brand channels get cut first every time.

This creates what Gartner's June 2026 research calls the brand doom loop:

• Sales soften, so you cut brand spending
• Attribution shows activation channels "working better," so you shift more budget there
• Brand equity erodes. The 95% of buyers who aren't in market right now stop building memory of your business
• Activation costs rise because you're fighting harder for a shrinking pool of people who remember you
• Sales soften further. Repeat from step 1.

The pattern is consistent across every dataset. Cut the channels that build awareness, and the channels that capture demand get more expensive until they stop working entirely.

You're Optimising for 5% of Your Market

The Ehrenberg-Bass Institute's research, published through LinkedIn's B2B Institute, established the 95/5 rule: at any given time, only about 5% of your potential buyers are actively in market. The other 95% won't need you for months or years.

Google Ads is brilliant at capturing the 5%. Someone searches "plumber Adelaide" and your ad appears. That's not brand building. That's demand harvesting. It only works because the demand already exists.

Facebook, content, email, and brand activity are what build your business's presence with the 95%. They're planting seeds that won't germinate for six months. They're making sure that when someone's hot water system breaks next February, your business is the one they think of first.

Byron Sharp calls this mental availability: the probability that a buyer will think of your brand in a buying situation. It requires consistent exposure across multiple touchpoints over time. It cannot be built by search ads alone, because search ads only reach people already looking for you.

When you cut Facebook because the CPL looks bad on a dashboard, you're not cutting a marketing channel. You're cutting the pipeline that feeds future demand into every other channel. The effects take 2-3 months to show up, which is exactly long enough for the business owner to believe the cut was a good decision before reality arrives.

The $0.35 You're Losing on Every Dollar

Analytic Partners quantified the cost of making channel decisions based on siloed attribution: 35 cents of opportunity lost for every $1 spent. A business investing $5,000/month in marketing is leaving $1,750/month on the table by measuring channels in isolation.

That's not a rounding error. Over a year, it's $21,000 in wasted potential. For an SME spending $10,000/month, it's $42,000.

The fix isn't complicated. It just requires accepting that your dashboard is a partial view, not the full picture.

Sam Tomlinson draws the distinction clearly: attribution divides credit with no regard for causal relationship. Incrementality identifies outcomes that would not have happened without the activity. They are fundamentally different questions, and the first one is the one almost every SME is answering.

Kaushik offers a memorable test for incrementality: "If you fired everybody in marketing, how long until the company misses the team?" If the answer is "immediately," you're mostly doing activation. If the answer is "three months," you have brand equity doing the heavy lifting. Most businesses have no idea which answer applies to them.

What This Means for Your Business

This isn't a theoretical problem. If you've ever cut a marketing channel because the CPL looked bad compared to another channel, you may have already experienced the consequences. Here's what to do instead.

The most expensive marketing decision isn't spending too much on an underperforming channel. It's cutting a channel that was quietly making everything else work.

Further Reading

• Analytic Partners ROI Genome: Brand Marketing Drives Sales ROI - Data on how 45% of advertising impact crosses channel boundaries
• Brainlabs: Proving Paid Social's Search Halo Effect - 17 Meta Conversion Lift studies showing +19% incremental search visits
• Gartner: 84% of Companies Stuck in the Brand Doom Loop - 2026 survey of 426 marketing leaders on brand investment
• Avinash Kaushik: Attribution Is Not Incrementality - Why most marketers are answering the wrong measurement question
• WARC: The Multiplier Effect - How over-investing in performance reduces ROI by 40%