Why sample design beats sample size
A sample drawn from the wrong population is not more reliable for being larger, only more confident. How we specify one that survives challenge.
Request the noteInsights
Short notes drawn from live research programmes. We publish where a finding is durable enough to be useful and general enough to share without compromising a client engagement.
A sample drawn from the wrong population is not more reliable for being larger, only more confident. How we specify one that survives challenge.
Request the noteBoards keep announcing supply chain relocation. Order books for the equipment that would make it real tell a considerably slower story.
Request the noteWhat enterprise buyers actually renew on, and why the headline metric increasingly measures pricing action rather than product value.
Request the noteCoverage decisions are treated as regulatory detail and priced as such. In most medtech diligence they are the single largest swing factor.
Request the noteCategory-level demand has held up better than modelled in some baskets and worse in others. The split is not where most forecasts placed it.
Request the noteGrid connection timetables constrain more transition projects than capital availability does. They are also public, and largely unwatched.
Request the notePlatform economics assume adviser inertia. Interview evidence points to a narrower set of triggers that break it faster than incumbents assume.
Request the noteIntegration synergy models rarely account for the operational cost of holding acquired customer relationships through a rebrand.
Request the noteNational footprint is the metric operators report. Route density is the metric that determines whether the unit economics ever close.
Request the noteMost engagements begin with a thirty-minute scoping conversation. There is no charge for it, and no obligation to proceed.