Maaz Zindani

Loading the index…

↑↓ move↵ openesc close4072 pages

Help for owners

My Looker Studio dashboard is slow, stale or shows errors

Short answer

Looker Studio keeps query results in memory for a freshness period set on each data source, so numbers can look old even when the source is current. Reports that use the owner's credentials fail when that person loses access. Slowness usually comes from heavy queries on a large source. Check the data source's freshness and credential settings first.

  • Written for business owners
  • Updated
  • By Maaz Zindani, Senior Software Engineer

Does this sound like your situation?

  • The dashboard shows yesterday's numbers although the source is up to date.
  • Charts show a data set configuration error or no data.
  • The report takes a long time to load, especially with many charts.
  • It broke after someone left or changed a Google account.
  • Viewers see different numbers or cannot see data that I can.

Why does this happen?

  1. Results are cached for a freshness period

    Looker Studio stores report queries and results in memory, and repeated queries inside the freshness window are served from memory. BigQuery and SQL databases default to 12 hours, Google Sheets to 15 minutes, and Google ad and measurement products refresh every 12 hours, a rate that cannot be changed. Data may leave memory before the full interval ends.

  2. Manual refresh has limits

    Report editors can refresh all data sources at any time. Viewers can refresh only if the editor turns on viewer data refresh. A one-minute cooldown applies to manual refreshes.

  3. The data source uses one person's credentials

    With owner's credentials, viewers do not need access to the underlying data because the owner's access is used. With viewer's credentials, each viewer needs their own access. A source set to the owner's credentials depends on that person keeping access.

  4. The queries are heavy

    Google advises matching freshness to how often data changes, writing efficient queries and using extracted data sources for smaller data sets, up to 100 MB, so viewers query a snapshot. For BigQuery it suggests materializing heavy views.

  5. The source itself changed

    A renamed column, a moved spreadsheet or a changed sheet tab breaks fields in the report even though Looker Studio is working.

What can I do about it today?

  1. Check the date on the numbersCompare one figure with the source. If it is old, the freshness setting is the first suspect.
  2. Refresh from the reportIf you are an editor, refresh the data and wait a minute before trying again.
  3. Find who owns each data sourceOpen the data source list and note the owner and which credential type is used. Do not change credentials yet.
  4. Take screenshots of the errorsNote which chart and which message. This saves hours.
  5. Make a copy of the reportUse a copy for experiments, so the live one is not damaged.

When should I bring someone in?

  • The report feeds decisions or client reports that must be right.
  • The source is a large database and every chart loads slowly.
  • The data comes from several sources with numbers that do not agree.

How do I handle it?

  1. Trace one number back to its sourceI follow a single figure from the chart to the data source, to the query, to the raw data.
  2. Set freshness on purposeI match each source's freshness to how often the data changes, and say what each choice costs in speed and quota.
  3. Move credentials to a stable ownerData sources belong to an account the company controls, not to one person's login.
  4. Prepare the data before the dashboardHeavy calculations move into a prepared table or view, so the report only reads the result.
  5. Document each sourceA list of sources, owners, refresh settings and who to ask.

The price depends on what I find, so I quote in writing after a short discovery call. Tell me what is happening and I reply within 24 hours.

What do owners ask about this?

Why does Looker Studio show old data?

Results are cached for the freshness period set on the data source. Editors can change it or refresh manually.

Can viewers refresh the data?

Only if the editor enables viewer data refresh in the report settings, with a one-minute cooldown.

Does a shorter freshness setting fix slowness?

No. Google notes that shorter refresh intervals give more current data but cost more in queries and quotas.

What are extracted data sources?

Google describes them as snapshots for smaller data sets, up to 100 MB, which load faster and update on a schedule.

Sources

How was this page made?

I drafted this page with AI assistance from the official documents listed under Sources. Found a mistake? Tell me.

Next step

Want this off your desk?

Send me a few lines on where things stand. I reply within 24 hours.