Most marketing turnarounds fail because the incoming leader treats them as optimization problems. They aren't. If the business is in sustained decline, the marketing function has usually drifted from the customer it was built to serve. No amount of channel optimization fixes that.
Here's what typically happens. A company finds product-market fit with a specific customer. It scales aggressively through performance channels. Over time, the growth machinery optimizes for short-term metrics — last-click conversions, same-day ROAS, lower-funnel transactions — rather than the original customer insight. The targeting broadens. The creative gets generic. The numbers still look okay for a while, until they don't.
Les Binet and Peter Field documented this pattern extensively in their IPA research. Their analysis of nearly a thousand campaigns found that the optimal balance between brand building and sales activation is roughly 60:40. Yet most performance-heavy organizations are inverted — 80% or more of spend goes to activation, starving the brand of the long-term equity that ultimately drives sustainable growth. Emotional campaigns, they found, are nearly twice as likely to produce top-box profit growth over the longer term. The industry heard the message. Most still haven't acted on it.
By the time a new marketing leader walks in, the org is usually over-concentrated in one or two paid channels, under-invested in brand, and measuring against metrics that reward efficiency rather than growth. The funnel is being starved of new customers because acquisition costs look unhealthy — but they look unhealthy partly because the targeting has drifted away from the customers who actually want the product.
The instinct is to immediately diversify channels and test new things. Resist it — at least for the first few months. You probably have meaningful non-incremental spend in your portfolio that's hiding in plain sight. Remarketing campaigns targeting customers who were going to buy anyway. Affiliates taking credit for organic demand. TV or display spend that nobody has seriously measured for lift.
Uber learned this the hard way. In 2018, they ran incrementality tests that revealed their Meta ads weren't actually bringing in new riders — the attributed conversions were people who would have signed up regardless. They cut $135M in wasteful spend and reinvested it into growth drivers like Uber Eats and global expansion. Separately, they discovered $100M being lost to outright ad fraud. Without incrementality testing, they would have continued funding both indefinitely.
Cutting non-incremental spend is the highest-leverage move in a turnaround because it simultaneously improves your economics and frees up budget to reinvest. But it requires an honest incrementality assessment, and most marketing orgs don't have one because the existing attribution model tells a more comfortable story.
This is the part that sounds obvious and isn't. "Know your customer" is the most repeated and least practiced principle in marketing.
Going back to the customer means making a specific choice about who you serve and being willing to narrow. It means changing creative direction, rethinking media strategy, and pushing back on institutional momentum. It means telling the performance team that a broader audience is not better if it's less differentiated. And it means delivering actual tools — personas, brand voice, positioning — that the rest of the organization can execute against. Having a customer strategy in a slide deck doesn't count. If you haven't delivered artifacts the team can use, you haven't done the work.
Most turnarounds I've observed get the "who is our customer" question right and then fail to operationalize it fast enough. The window between articulating the direction and delivering the executional tools is where velocity dies.
Channel diversification is critical for any business with concentration risk. But it's a sequencing problem, not just a strategy problem.
You can't reallocate 30% of your dominant channel's spend into new channels overnight. The new channels need to be tested, optimized, and scaled — and your P&L is dependent on the concentrated position in the meantime. You have to build the new while maintaining the old, which requires patience and tolerance for a period where overall efficiency might look worse before it looks better.
The biggest opportunities are usually in the long tail — influencer ecosystems, partnerships, content-driven channels — rather than just shifting spend between major ad platforms. Low barriers to entry rarely lead to outsized returns. The channels that take more work to build tend to be the ones your competitors aren't in.
Airbnb's experience during the pandemic is the most cited recent example of this, and it's cited for good reason. When COVID hit, they cut performance marketing spend to zero. Traffic dropped only 5%. That discovery led to a permanent strategic shift — they reduced overall marketing spend by 28%, redirected toward brand building, and launched their first major brand campaign in five years. The result: 90% of traffic remained direct or unpaid, revenue grew 40% to $8.4B in 2022, and they posted their first full-year profit. Their CEO now describes the role of marketing as "education, not buying customers."
Not every company has Airbnb's brand equity starting point. But the directional lesson applies broadly: if you're in a turnaround and every dollar is going toward lower-funnel activation, you're fighting the symptom while the disease progresses. Brand equity compounds. Performance spend doesn't.
I've come to believe that a turnaround is fundamentally an organizational challenge, not a strategic one. The strategy is usually not that complicated. The hard part is having a leadership team that can execute against it — that defaults to analytical thinking, is comfortable with ambiguity, and will challenge existing approaches rather than run inherited playbooks.
This sometimes means replacing most of the leadership team. Not because the people lack talent, but because the turnaround requires different instincts than the steady-state role did. The longer you wait to make those changes, the more the strategy stalls.
It's not ideas. It's not strategy. It's rate of experimentation.
How fast can you test? How fast can you measure? How fast can you iterate? Most marketing organizations are constrained not by a lack of ideas but by data infrastructure, team capability, and technical limitations that slow down their ability to learn. Improving that learning velocity — through better tooling, stronger analytical capability, and fewer manual workflows — is often the single highest-ROI investment in a turnaround.
And you have to get the measurement right. If the org is optimizing against lower-funnel transactional metrics, every decision will favor short-term demand at the expense of the customer relationships that drive long-term value. Shifting to acquisition and lifetime value as the primary measures is uncomfortable, because the feedback loops are longer and the numbers are less flattering in the short term. Do it anyway.