I have watched the same sequence run in several organisations, and it usually starts with a purchase.
A platform is bought before anyone has agreed which business problems it is for. Use cases get found afterwards to justify the spend. Adoption is slow, because the people who were meant to use it were not asked what they needed. The benefits never show up in the numbers. Confidence goes, the analytics lead leaves or is moved sideways, budget is cut, and the team breaks up. Eighteen months later a new leader arrives and buys a different platform.
The NewVantage Partners 2022 executive survey puts a number on where the difficulty sits: 91.9% of executives named culture, people and process as the biggest obstacle to becoming data driven. 8.1% named technology.
The alternative is straightforward, though it is slower to start.
Choose use cases for the value they carry, and check they line up with what the business is actually trying to do this year. Involve the people who will use the output from the beginning, not at rollout. Pick the technology once you know what you are delivering. Measure what the work turns out to be worth, put a name against that number, and report it.
Two things accumulate each time round: trust, and knowledge of how the business works. Both make the next piece of work easier and cheaper.
There is a second effect that gets less attention. Data teams are happier when the things they build get used. I have never met anyone who left analytics because their work was adopted.
The two-loop model and the eight-pillar framework behind it are what the data strategy work here is built on. The two-minute maturity assessment scores you across those eight pillars, and the data strategy sprints set out what a full engagement looks like.