Power Query Big Data Reconciliation

Ғылым және технология

This video shows how to build and complete a classic accounting reconciliation using Excel Power Query. Once the model is built, it will reconcile and identify differences between two data sets with several hundred rows in just a few seconds.

Пікірлер: 16

  • @rexbk
    @rexbk2 ай бұрын

    It is so useful and easy to understand. Thank you much.

  • @alexrosen8762
    @alexrosen87625 ай бұрын

    Super useful tutorial and well demonstrated. Thanks 🙏

  • @hafizasim4245
    @hafizasim42454 ай бұрын

    This is easy In practice its more complicated where files are from different systems havono references How to do that where there is no common key?

  • @christophertauss3640
    @christophertauss36403 ай бұрын

    Nice work. Thanks for sharing this. I will definitely see if this can work for us. I am not sure if rows can always be in exactly the same order for most of the comparisons I see, but maybe there is some way to impose some kind of sort and then do a similar analysis.

  • @rajanpradeepankarath8846
    @rajanpradeepankarath88467 ай бұрын

    Great video without using merge and joints. Would you make a video to reconcile 2 tables, both with 3 rows - date, particulars and amounts. The recon should be done first with particulars matching, and then if particulars dont match, with dates matching

  • @AZAdvancedAnalytics

    @AZAdvancedAnalytics

    7 ай бұрын

    Yes, I will be posting one on how to reconcile amounts with multiple fields involved. It will be the same basic process, however you create a unique key by concatenating the fields together before the pivot. i.e. Date&"-"&Number&"-"&Department, etc. Then parsing this back out once the zeros are filtered out, to see differences by field.

  • @rajanpradeepankarath8846

    @rajanpradeepankarath8846

    7 ай бұрын

    @@AZAdvancedAnalytics thank you, will wait for it. Please inform when video is available

  • @Vandalfoe
    @Vandalfoe5 ай бұрын

    At 8:10, instead of subtraction, could you just filter and keep the nulls?

  • @AZAdvancedAnalytics

    @AZAdvancedAnalytics

    5 ай бұрын

    Hello. If you filter and keep the nulls, you would be excluding the records that were not null, but had different amounts in both columns. However, you could use that method to count records that were in one table and not another.

  • @rcdny
    @rcdny7 ай бұрын

    Why not use merge queries and do joins

  • @AZAdvancedAnalytics

    @AZAdvancedAnalytics

    7 ай бұрын

    That's a great way to do the same thing! I recorded the "pivot table method" since I thought it would be easier to follow and learn for those already familiar with Excel Tables, and or Pivot Tables.

  • @GeertDelmulle

    @GeertDelmulle

    6 ай бұрын

    Indeed. The equivalent of Venn diagrams is PQ are those join types.

  • @iankr
    @iankr7 ай бұрын

    Hi Many thanks for that. But did you really need to add the queries to the Data Model?

  • @AZAdvancedAnalytics

    @AZAdvancedAnalytics

    7 ай бұрын

    Hi iankr. I don't think you have to, but working within the data model allows us to build the solution within that virtual space. Then, return just what we need to complete the report within Excel.

  • @christophertauss3640

    @christophertauss3640

    3 ай бұрын

    @@AZAdvancedAnalytics I was wondering the same thing, if adding to the Data Model was really needed, but thanks for a very interesting approach.

Келесі