A PPC manager rebuilds an account using textbook best practices. Traffic craters. Revenue drops 40%. Recovery takes six months.
This isn't a cautionary tale about Google Ads. It's a case study in what happens when marketing execution ignores business constraints.
Ana Kostic shared this story on PPC Live the Podcast, and the lesson extends far beyond paid media. She inherited a messy Google Ads account, applied every restructuring best practice she knew, and watched the client's revenue fall off a cliff. The new structure was technically sound. The business impact was catastrophic.
The Real Cost Wasn't in the Platform
Kostic's mistake wasn't strategic. It was contextual. By rebuilding the account from scratch, she wiped years of historical performance data that Google's algorithms had learned from. The platform had to relearn everything: which audiences converted, which keywords drove value, which bid adjustments actually worked.
The account eventually recovered, but only after roughly two and a half months of depressed performance. Full benefits took closer to six months to materialise.
For a CFO watching monthly revenue reports, that's not a learning period. That's a crisis.
Why Best Practices Fail Without Business Constraints
Here's what most marketing playbooks miss: best practices assume infinite runway. They assume the business can absorb short-term disruption while the new approach proves itself. They assume cash flow can handle a 40% revenue dip for a quarter or two.
Most businesses can't.
Kostic's insight is deceptively simple: You have to think about the business first. But operationalising that insight requires asking questions that most marketers skip entirely.
Before recommending any major structural change, she now asks about cash flow, margins, growth goals, and how much short-term disruption the company can realistically absorb. These aren't marketing questions. They're finance questions. And they should be non-negotiable inputs to any significant campaign restructure, platform migration, or channel reallocation.
The Gradual Migration Model
Kostic's current approach inverts the typical agency playbook. Rather than replacing campaigns overnight, she introduces changes gradually, allowing new structures to learn while existing campaigns continue delivering results.
Her philosophy: We like it slow and boring.
From a CFO's perspective, this is exactly right. Slow and boring means predictable. Predictable means forecastable. Forecastable means the board doesn't get surprised by a revenue miss that marketing can't explain until the learning period ends.
The math here isn't complicated. If a restructure promises 20% efficiency gains but requires a 40% revenue dip for three months, the payback period might be negative. Run the sensitivity analysis before you touch the account.
What This Means for Marketing Leaders
Kostic's story is about PPC, but the pattern repeats across every marketing function.
Content teams retire high-performing legacy assets to launch a refreshed content strategy. Demand gen teams migrate to a new MAP and lose lead scoring history. Brand teams consolidate campaigns under a new creative platform and watch engagement metrics reset.
In each case, the new approach might be objectively better. But better doesn't matter if the business can't survive the transition.
The question isn't whether the change is correct. The question is whether the business can afford the learning curve.
The Communication Layer
One detail from Kostic's account deserves attention. When the restructure failed, her agency didn't assign blame. They focused on transparency, created a recovery plan, and supported both the client and the team until performance stabilised.
This is the part most marketing leaders underestimate. When a major initiative underperforms, the instinct is to defend the strategy or blame external factors. Neither approach builds trust.
What builds trust: acknowledging the miss, explaining the root cause, presenting a recovery timeline with milestones, and reporting against that timeline until performance normalises.
If you can't explain to your CFO exactly why revenue dropped and exactly when it will recover, you don't have a communication problem. You have a planning problem.
The Sales Conversation Gap
Kostic's other recommendation is worth flagging: spend more time talking to sales teams. Those conversations reveal the language customers actually use, which often differs from the keyword data platforms provide.
This isn't just a PPC insight. It's a RevOps alignment issue.
Marketing teams that operate in isolation from sales build campaigns around platform signals. Marketing teams that embed with sales build campaigns around buyer language. The second approach converts better because it matches how prospects actually describe their problems.
If your marketing team's last substantive conversation with sales was a pipeline review, you're missing signal.
AI Doesn't Change the Fundamentals
Kostic advocates for products like Performance Max, but she's clear that AI shouldn't replace good decision-making. Advertisers should still put guardrails in place and test gradually.
This is the right framing. AI tools accelerate execution, but they don't eliminate the need for business context. An algorithm optimising for conversions doesn't know your cash flow constraints. It doesn't know your margin structure. It doesn't know that a 40% revenue dip for two months means you miss payroll.
The human layer isn't optional. It's where business constraints get translated into platform guardrails.
The Pilot Framework
For any major marketing restructure, the Kostic case suggests a simple pre-flight checklist:
What's the maximum revenue disruption the business can absorb? Over what time horizon? What's the expected learning period for the new approach? What's the recovery plan if performance doesn't stabilise within that window?
If you can't answer these questions with numbers, you're not ready to execute.
The Uncomfortable Truth
Best practices are averages. They describe what works across a population of accounts, campaigns, or businesses. They don't describe what works for your specific business with your specific constraints.
The 40% revenue drop wasn't a failure of PPC knowledge. It was a failure to translate platform best practices into business-safe execution.
Every marketing leader will face a version of this decision: the technically correct approach versus the approach the business can actually survive. The ones who build durable careers learn to ask the finance questions before they touch the account.
