Developers claiming 10x speedups from AI often aren't 10x faster on their core tasks. Instead, they're tackling new side projects that were previously impossible, creating a perception of "infinite" speedup. However, these new tasks are often less economically valuable, inflating the true productivity gain on business-critical work.

There's often a massive gap between a company's strategic goals and where development teams actually spend time. In one case, only 2% of capacity was spent on the top strategic goal because teams are "magnets for requests" that derail progress on the big picture.

As companies scale, the supply of obvious, valuable work dwindles. To stay busy, employees engage in "hyper-realistic work-like activities"—tasks that mimic real work (e.g., meetings to review decks for other meetings) but generate no value. It's a leader's job to create a sufficient supply of *known valuable work*.

Leaders in large companies often lack visibility into the day-to-day workflows that drive results. They see inputs like salaries and outputs like KPIs, but the actual process of how work gets done—the institutional know-how—is a black box that walks out the door every day.

Leaders focus on increasing reports because headcount is an objective metric for promotion, unlike subjective assessments of business impact. This creates an incentive for managers to accumulate people, even if it's not the most impactful business decision.

The primary obstacle to analyzing engineering output was the technical difficulty of synthesizing massive, unstructured data from disparate sources like code repositories, documents, and Slack. It wasn't a cultural issue or lack of tools; it was a data fragmentation problem that AI can now solve.

A 4x productivity increase was achieved by using data transparency to identify bottlenecks and underperforming resources. The primary value wasn't merely measuring output, but diagnosing *why* some teams struggled and bringing them up to the standard set by top performers within the same organization.

Many teams fall into a "busyness trap," engaging in activities that don't advance core objectives. This creates a hidden tax on productivity, as effort is spent on work that doesn't move the needle. The key is shifting focus from simply being busy to working on the right, high-impact tasks.

Organizing by function (e.g., all sales together) seems efficient but incentivizes teams to optimize their individual metrics, not the company's success. This sub-optimization prevents cross-functional learning and leads to blame games, ultimately harming the entire customer value stream and creating a non-learning organization.