Data Architecture

Data warehouse vs data lake vs lakehouse is the architecture argument that eats the most meeting time and produces the least clarity. All three will store your numbers and feed a dashboard; the difference appears eighteen months later in the size of the bill and the number of people it takes to run. This guide compares the three on the factors that actually move the answer: what each architecture is and when you are forced to agree what the data means, how the bill is genuinely built once compute and salaries are counted, the governance and UK GDPR duties that do not change whichever you pick, the skills and monthly run effort each demands, which workloads belong where, realistic migration paths with the failure mode of each, and a weighted scoring framework you can complete with your own numbers.