Defensive Strategy: Controlling the Rate of Decline

Not every strategy is about growth. Sometimes the most important strategic question is: how do we manage a decline we can't prevent?

In late 2023, I was engaged by a top-five global open source technology company. Their core revenue model was search - and large language models were threatening to fundamentally reshape how people find information online.

The question wasn't whether LLMs would impact their business. It was how much, how fast, and what they could do about it.

Sizing the Threat

The first step was getting precise about the problem. Not "AI is disrupting search" - that's a headline, not an analysis. We needed to understand the mechanics.

At the time, LLMs were primarily used for information gathering - answering questions, summarizing content, explaining concepts. These were the same tasks that drove search traffic. But not all search traffic is created equal.

The revenue that mattered was commercial intent search - people looking to buy something, compare products, find services. These searches generated ad clicks. And in late 2023, LLMs weren't yet capturing that commercial behavior at scale. People were using ChatGPT to learn things, but they were still using search engines to buy things.

We also identified a buffering effect: Google and Microsoft were integrating AI summaries directly into search results. This meant that even as users adopted LLM-based tools, the major search engines would retain much of that traffic by incorporating AI into their own experience. For a company whose search revenue came through partnerships with these engines, this actually slowed the disruption.

Our model projected approximately 25% decline in paid search volume and roughly 20% revenue decline due to user mix shifting toward AI-first information gathering. Significant - but not existential. And critically, not immediate.

The Wrong Response

The instinct when facing disruption is to either deny it or panic. Denial leads to inaction. Panic leads to scattershot initiatives that drain resources without strategic coherence.

The company had been leaning toward panic. Internal discussions were circling around radical pivots - completely abandoning search, building their own LLM from scratch, acquiring AI startups. All of these had merit in isolation. None of them made sense as a first response.

Why? Because the company was still generating hundreds of millions in search revenue. That revenue was declining, not disappearing. And every dollar spent on a speculative pivot was a dollar not spent on managing the transition.

The Framework: Defend, Diversify, Own

We built the strategic response around three pillars:

Defend the core. Search revenue was declining but would remain the largest revenue source for years. We focused on optimizing what was still working - improving search quality, strengthening distribution partnerships, and maximizing revenue per query during the transition period. The goal wasn't to stop the decline. It was to slow it and extract maximum value from the existing model while it lasted. Diversify revenue. The company had several underutilized products - including a mailing client with a substantial user base. These products had been treated as secondary to search for years. We recommended aggressive cross-selling and monetization of these existing products. The user base was already there. The infrastructure was already there. The revenue potential was real and could be captured without building anything new. Own the AI experience. This was the bold move. Rather than ceding the AI search experience to Google and Microsoft, we recommended the company build its own LLM-powered search capability. The logic: if users were going to interact with AI instead of traditional search results, the company should own that interaction. Serve ads within AI conversations. Provide AI-generated answers alongside traditional results. Control the experience rather than watching someone else replace it.

The Financial Model

Strategy without numbers is just storytelling. We built scenario models that quantified each pillar:

The full strategy didn't promise growth. It promised stability - which, in the context of a potentially devastating disruption, was the right target.

Execution Speed

Within six months of engagement, the company launched three new monetization projects. That speed mattered. Defensive strategy has a shelf life - the longer you wait to execute, the less runway you have.

We prioritized initiatives by time to revenue impact:

This sequencing ensured the company was generating defensive revenue immediately while building toward the larger strategic shift.

The Lesson

Defensive strategy isn't about stopping decline. You can't always stop decline. Market shifts happen. Technology disrupts. Competitive dynamics change.

What you can control is the rate of decline and whether you're building offsetting revenue while the transition happens. The companies that navigate disruption successfully aren't the ones that find a magic solution to make the threat go away. They're the ones that buy themselves time, diversify their revenue base, and position to own the next version of their market.

The worst response to disruption is paralysis. The second worst is panic. The right response is a clear-eyed assessment of the threat, a quantified plan to manage it, and the discipline to execute at speed.