[{"data":1,"prerenderedAt":121},["ShallowReactive",2],{"case-en-data-lakes":3},{"_path":4,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":8,"description":9,"summary":10,"slug":11,"code":12,"label":13,"tags":14,"challenge":17,"approach":18,"outcome":19,"seo":20,"body":23,"_type":114,"_id":115,"_source":116,"_file":117,"_stem":118,"_extension":119,"sitemap":120},"\u002Fen\u002Fportfolio\u002Fdata-lakes","portfolio",false,"","Bringing scattered data together","This is a method note: it describes how we approach this kind of problem, not a write-up of a named client.","How we approach it when data lives in ten systems and every question means a manual export — and nobody trusts the numbers.","data-lakes","M-03","Data",[13,15,16],"Analytics","Architecture","Data is scattered across several systems. Every business question means a manual export and reconciling spreadsheets — and two reports give two different numbers.","We model the questions before choosing storage. We build one source of truth that fills incrementally, and keep it reachable by the people who ask the questions.","Done right, answers take minutes instead of days, and the same question always returns the same number. The team decides from data because it finally trusts it.",{"title":21,"description":22},"Bringing scattered data together — method note | Ag3","How Ag3 approaches unifying scattered data: questions first, one source of truth, incremental — instead of one more data silo.",{"type":24,"children":25,"toc":108},"root",[26,33,40,45,51,97,103],{"type":27,"tag":28,"props":29,"children":30},"element","p",{},[31],{"type":32,"value":9},"text",{"type":27,"tag":34,"props":35,"children":37},"h2",{"id":36},"the-problem-isnt-storage",[38],{"type":32,"value":39},"The problem isn't storage",{"type":27,"tag":28,"props":41,"children":42},{},[43],{"type":32,"value":44},"\"We need a data lake\" is often the answer to the wrong question. The real trouble isn't where to put the data — it's that nobody can reliably answer a simple business question without half a day of reconciling exports by hand.",{"type":27,"tag":34,"props":46,"children":48},{"id":47},"how-we-approach-it",[49],{"type":32,"value":50},"How we approach it",{"type":27,"tag":52,"props":53,"children":54},"ul",{},[55,67,77,87],{"type":27,"tag":56,"props":57,"children":58},"li",{},[59,65],{"type":27,"tag":60,"props":61,"children":62},"strong",{},[63],{"type":32,"value":64},"Questions first.",{"type":32,"value":66}," We map the decisions the data should support. The model follows the questions, not the other way round.",{"type":27,"tag":56,"props":68,"children":69},{},[70,75],{"type":27,"tag":60,"props":71,"children":72},{},[73],{"type":32,"value":74},"One source of truth.",{"type":32,"value":76}," One agreed definition for each concept — \"active customer,\" \"revenue\" — so two reports can no longer contradict each other.",{"type":27,"tag":56,"props":78,"children":79},{},[80,85],{"type":27,"tag":60,"props":81,"children":82},{},[83],{"type":32,"value":84},"Incremental.",{"type":32,"value":86}," We connect the sources that matter most first and deliver value as we go, not after a long silence.",{"type":27,"tag":56,"props":88,"children":89},{},[90,95],{"type":27,"tag":60,"props":91,"children":92},{},[93],{"type":32,"value":94},"Reachable.",{"type":32,"value":96}," The result stays queryable with tools the team already knows — it doesn't require a new department.",{"type":27,"tag":34,"props":98,"children":100},{"id":99},"what-changes",[101],{"type":32,"value":102},"What changes",{"type":27,"tag":28,"props":104,"children":105},{},[106],{"type":32,"value":107},"Questions that used to take days get answered in minutes. The same definition holds everywhere, so arguments about \"whose number is right\" disappear. Decisions lean on the data because the data is finally trustworthy.",{"title":7,"searchDepth":109,"depth":109,"links":110},2,[111,112,113],{"id":36,"depth":109,"text":39},{"id":47,"depth":109,"text":50},{"id":99,"depth":109,"text":102},"markdown","content:en:portfolio:data-lakes.md","content","en\u002Fportfolio\u002Fdata-lakes.md","en\u002Fportfolio\u002Fdata-lakes","md",{"loc":4},1784880623054]