# Quicksight data incremental refresh on weekly basis

**URL:** <https://community.amazonquick.com/t/quicksight-data-incremental-refresh-on-weekly-basis/21036>\
**Category:** Q&A\
**Tags:** analysis, Business-Intelligence-Engineer, quick-sight\
**Created:** [November 9, 2023, 3:20pm UTC](https://community.amazonquick.com/t/quicksight-data-incremental-refresh-on-weekly-basis/21036 "2023-11-09T15:20:56Z")\
**Posts on this page:** 8\
**Page:** 1

<div class="post-metadata">

**Author:** ![Zeeshan](https://avatars.discourse-cdn.com/v4/letter/z/c4cdca/32.png) [@Zeeshan](https://community.amazonquick.com/u/Zeeshan)\
**Post date:** [November 9, 2023, 3:20pm UTC](https://community.amazonquick.com/t/quicksight-data-incremental-refresh-on-weekly-basis/21036/1 "2023-11-09T15:20:56Z")

</div>

Hello team ,  
I just wanted to know how my data will incrementally refresh if i do not have date level column,  
in my sql query i have week\_first\_date level data. i want to understand the process of incremental data refresh on week level and do i need to put dynamic dates in query ?  
and which window unit should i select to optimize my run time and what filters it will use date column or week\_first\_date

---

<div class="post-metadata">

**Author:** ![Sanjeeb2022](https://d2orw1595917va.cloudfront.net/user_avatar/community.amazonquick.com/sanjeeb2022/32/20947_2.png) [@Sanjeeb2022](https://community.amazonquick.com/u/Sanjeeb2022)\
**Post date:** [November 10, 2023, 2:44am UTC](https://community.amazonquick.com/t/quicksight-data-incremental-refresh-on-weekly-basis/21036/2 "2023-11-10T02:44:46Z")

</div>

Hi @Zeeshan - Welcome to AWS Quick Sight community and thanks for posting the question. Can you please have a look on the below documentation - [Refreshing SPICE data - Amazon QuickSight](https://docs.aws.amazon.com/quicksight/latest/user/refreshing-imported-data.html)

Tagging @WLS-D @David_Wong for their advise as well.

Regards - Sanjeeb

---

<div class="post-metadata">

**Author:** ![Zeeshan](https://avatars.discourse-cdn.com/v4/letter/z/c4cdca/32.png) [@Zeeshan](https://community.amazonquick.com/u/Zeeshan)\
**Post date:** [November 10, 2023, 9:59am UTC](https://community.amazonquick.com/t/quicksight-data-incremental-refresh-on-weekly-basis/21036/3 "2023-11-10T09:59:43Z")

</div>

Hello @Sanjeeb2022 ,  
Thank you for replying but thing is on website it is not mentioned how week level data will work in the refreshing part.  
in my query i have week level data of sales but i want to check how week level incremental refresh will work in QS  
Thanks

---

<div class="post-metadata">

**Author:** ![Sanjeeb2022](https://d2orw1595917va.cloudfront.net/user_avatar/community.amazonquick.com/sanjeeb2022/32/20947_2.png) [@Sanjeeb2022](https://community.amazonquick.com/u/Sanjeeb2022)\
**Post date:** [November 10, 2023, 11:49am UTC](https://community.amazonquick.com/t/quicksight-data-incremental-refresh-on-weekly-basis/21036/4 "2023-11-10T11:49:04Z")

</div>

Hi @Zeeshan - From your data set, how do you determine weekly data? Is there any field available which will tell or use to find the incremental. If not, basically you can do a full refresh on weekly basis.

Regards - Sanjeeb

---

<div class="post-metadata">

**Author:** ![ErikG](https://d2orw1595917va.cloudfront.net/user_avatar/community.amazonquick.com/erikg/32/14524_2.png) [@ErikG](https://community.amazonquick.com/u/ErikG)\
**Post date:** [November 10, 2023, 12:07pm UTC](https://community.amazonquick.com/t/quicksight-data-incremental-refresh-on-weekly-basis/21036/5 "2023-11-10T12:07:09Z")

</div>

Hi @Zeeshan,  
does your ` week_first_date` contains a date?  
Maybe you can define something like

 ![grafik](https://d1wl4jav5dp96l.cloudfront.net/original/2X/f/fc59e729c6527c3a4bb3c8b31911581c661ab0c0.png)  
BR

---

<div class="post-metadata">

**Author:** ![royyung](https://d2orw1595917va.cloudfront.net/user_avatar/community.amazonquick.com/royyung/32/3302_2.png) [@royyung](https://community.amazonquick.com/u/royyung)\
**Post date:** [November 10, 2023, 3:07pm UTC](https://community.amazonquick.com/t/quicksight-data-incremental-refresh-on-weekly-basis/21036/6 "2023-11-10T15:07:50Z")

</div>

when incremental refresh happens, it will get the date column from your data source and use Window size to define how many data need to be updated

In this screenshot, I have defined Windows size(number) to be 3 and Window size (unit) to be Day. SPICE will remove data on or after 3 days ago and ingest again. Assume today is Nov 10, it will remove data on or after Nov7 and ingest again  
 ![image](https://d1wl4jav5dp96l.cloudfront.net/original/2X/8/88487f445e636d5c8e0209707bf11300919dafc3.png)

Same logic applies on week, if you define Window size (unit) to be week and Window size(unit) to be 2, it will remove data on or after 14 days ago and ingest again. Assume today is Nov10, it will remove data on or after Oct28 and ingest again

---

<div class="post-metadata">

**Author:** ![Zeeshan](https://avatars.discourse-cdn.com/v4/letter/z/c4cdca/32.png) [@Zeeshan](https://community.amazonquick.com/u/Zeeshan)\
**Post date:** [November 20, 2023, 2:09pm UTC](https://community.amazonquick.com/t/quicksight-data-incremental-refresh-on-weekly-basis/21036/7 "2023-11-20T14:09:20Z")

</div>

Hello ErikG  
yes it is week first date like first date is monday so date is 20/11/2023,  
will it work on week level or daily level because i want to refresh on hour level.  
so that is what i want to check if we can update it as week level or day level or daily on week\_first\_date column.  
regards

---

<div class="post-metadata">

**Author:** ![Zeeshan](https://avatars.discourse-cdn.com/v4/letter/z/c4cdca/32.png) [@Zeeshan](https://community.amazonquick.com/u/Zeeshan)\
**Post date:** [November 20, 2023, 2:14pm UTC](https://community.amazonquick.com/t/quicksight-data-incremental-refresh-on-weekly-basis/21036/8 "2023-11-20T14:14:14Z")

</div>

Hello @royyung  
i understand your concept of incremental refresh,  
suppose i have weekly data in my data set because i want to save my cost of storage in AWS i will not pull data on date level cause it will increase size of data set and it will cost me more.  
I wanted to check if we have option to refresh data on the week\_first\_date which is a weekly column, so i can run incremental refresh on week\_firstdate coumn so my numbers of rows will be reduced and cost will be saved and my ingestion will also saved.  
regards
