A cache is a temporary data storage system. Fetching cached data can be much faster than fetching it directly from the underlying data set, and helps reduce the number of queries sent, minimizing costs for paid data access.
Data freshness refers to how up-to-date the data in a report is. Different types of reports have different requirements or expectations for data freshness. If you are measuring ad performance on your site or app, for example, daily updates might be sufficient. Reports based on social media analytics, on the other hand, may want their data updated multiple times in a day.
Report performance is a measure of how quickly the report loads. Fetching data directly from the underlying data set can be slow, which in turn makes your reports sluggish to load and respond to viewer changes, like applying filters and date ranges. In addition, for some data sources, such as BigQuery, fetching data directly can cost you money.In this article:
Set data freshness for a data source
Looker Studio automatically refreshes all the cached data for each data source used by your report at certain intervals. To change the default data refresh rate:
- Edit the data source. You can do this from within a report, or from the Data Sources home page.
- At the top, click Data freshness.
- Under "Check for fresh data," select a new refresh option, if available.
- Click SET DATA FRESHNESS.
Different connectors support different data refresh rates, as follows. Note that Google marketing and measurement products, such as Google Ads, Google Analytics, Campaign Manager 360, Search Console, YouTube Analytics, and others, refresh every 12 hours. That rate can't be changed.
|Google Cloud Storage||
|MySQL, PostgreSQL, MS SQL Server, CloudSQL for MySQL||
|Other Google ad and measurement products||
Every 12 hours
* Default refresh rate
How the cache works
Every component in a Looker Studio report has its own individual cache. That cache contains only the data needed to create that component's default visualization. For example, the cache for a table with 2 dimensions and 3 metrics, having a default date range of the last 28 days, and filtered by some value (country = Belgium, for instance) will contain only the data needed to display that table.
Here's another example:
Let's say your data source connects to a custom SQL query:
SELECT * FROM tableName. The results of this query are not cached, since the cache applies to charts and components, not the data source as a whole.
Now, in your report, you add a scorecard with the metric
Sum(revenue). The underlying query for the scorecard would be something like
Sum(revenue) FROM (SELECT * from tableName)).
When a component in your report requests data, if that query matches a previously received query, then the new data request is served from the cache.
If the request can't be served from the cache, Looker Studio requests the data from the underlying data set. For example, if a viewer changes the date range or filters the table described above by a new country, then Looker Studio requests the new data from the underlying data set. The results of that new query are then added to the table's cache, and Looker Studio remembers the new query for future use.
When the cache refreshes, all the old cached data and remembered queries are discarded. New queries generated by components on the report go directly to the underlying platform and the responses are added to the cache for those components.
Refresh report data manually
Report editors can refresh the cache at any time:
- View or edit the report.
- In the top right, click and then click Refresh data .
This refreshes the cache for every data source added to the report.
How to tell if report data is cached
You can see if data is coming from the cache by viewing the report and looking in the bottom left corner. When all the charts on the current page are being served from the cache, you'll see a lightning bolt icon along with the time and date of the last update .
Blending and cached data
For a blended data source, the cache will use the setting that satisfies the desired refresh times for all of the data sources included in the blend.
For example, if you blend a Sheets data source having a refresh time of 15 minutes, with a BigQuery data source having a refresh time of 4 hours, the resulting blended data source will have a refresh time of 15 minutes.
Data freshness and embedded reports
Viewers can't refresh the data in an embedded report. (The data will refresh as usual when the cache expires.)