1. Home
  2. Knowledge Base
  3. Universal Analytics Backup
  4. Troubleshooting Common Errors With the UA Backup and Looker Studio UA Backup Report

Common Errors with the UA Backup and Looker Studio Reports

This article is about common issues that can be encountered when generating a UA backup or connecting that backup to the Looker Studio reports. If you can't find the solution to your problem in this article or our knowledge base, please contact our support team

This article relates to the Universal Analytics Backup Utility offered by Analytics Canvas. 

Known Issues

Google Sheets files not generating in some cases

June 12th: We are investigating an issue with backups to Google Sheets.  A small percentage of backups where some files were too big for Sheets resulted in a partial export.  Please check the CSV download to ensure that the full backup was complete.  

If your Sheets did not generate or your Looker Studio report link fails, please contact support to open a ticket.  We will resolve the issue programmatically, however since the data is safely landed, we may not be able to resolve this before June 30th.  Please use the CSV files to verify your backup.  Find the CSV files under the "Exports" link if it is available. 

⚠️

If the status log said "All partitions loaded, launching export", or if the CSV files are available, your data has downloaded and it is simply a matter of generating the Sheets files and the Looker Studio report.

Error MessageAction
Generate UA Backup Tables task failed: Sheet not found.

June 17 + later: Do not retry at this time. Your backup has loaded to your Analytics Canvas account, but Google Sheets file generation has failed.

Some reasons why this can happen include:

  • the folder is deleted
  • the folder or directory is renamed
  • the user loses access to the folder

If you purchased the dashboard, the links will not be established.

Check your CSV output to verify that all tables and data have exported. You can import those files into Drive and open them in Sheets.

We will notify you about next steps.

Generate UA Backup Tables task failed: Parameter validation failed for "spreadsheetId" : The parameter value must be a non-empty string

June 17 + later: Do not retry at this time. Your backup has loaded to your Analytics Canvas account, but Google Sheets file generation has failed.

If you purchased the dashboard, the links will not be established.

Check your CSV output to verify that all tables and data have exported. You can import those files into Drive and open them in Sheets.

We will notify you about next steps.

Generate UA Backup Tables task failed: A task was canceled.

June 17 + later: Do not retry at this time. Your backup has loaded to your Analytics Canvas account, but Google Sheets file generation has failed.

This error can happen include one or more of your files are at or above the Google Sheets export limit. The process is designed to create as many Sheets as are required, exporting to CSV anything that is too big for Sheets. However, if there are multiple large files, the process can time-out, resulting in a failed load to Sheets.

If you purchased the dashboard, the links will not be established.

Check your CSV output to verify that all tables and data have exported. You can import those files into Drive and open them in Sheets.

We will notify you about next steps.

No source items found for Looker Studio linking API.

June 17 + later: Do not retry at this time. Your backup has loaded to your Analytics Canvas account, but Google Sheets file generation has failed.

If you purchased the dashboard, the links will not be established.

Check your CSV output to verify that all tables and data have exported. You can import those files into Drive and open them in Sheets.

If there is no CSV output, contact support with the name of your backup and subscriptionID (under Admin > Subscription and Quota).

We will notify you about next steps.

Generate UA Backup Tables task failed:DMUABackupExportingQueryIDGet returned no IDs

Do not retry. Your backup has loaded to your Analytics Canvas account, but the transfer of your data to your BigQuery dataset has failed.

If you purchased the dashboard, the links will not be established.

Some reasons why this can happen include:

  • the dataset was deleted before the export took place
  • the permissions on the dataset were removed prior to the export taking place
  • the backup encountered another error after the download and before exporting to BigQuery

If you purchased the dashboard, the links will not be established.

We will notify you about next steps.

Image

UA Backup Frequently Asked Questions

How long will it take / Why hasn't it started / What's taking so long? 

At this time, we cannot provide individual updates or accurate estimates as to when your job will complete. From May 2023 to May 2024, backups were running within a few hours to a few days. At this point, with this much traffic, we are unable to provide updates and time-to-completion estimates on individual backups.  

There is a significant line-up to get your backup running.  Once the backup begins, it can take a few hours to a few weeks.  Factors which affect runtime include the Analytics API's responsiveness to each query, network traffic, the size of your backups / number of API calls required to retrieve your data, etc. 

⚠️ Please ensure you have placed your order and RUN the backup.  It will not run automatically by purchasing credits, you must apply the order ID to the backup you want to run, or add credits to your account:

You can check on your backup as often as you'd like. The status log will update as the job progresses. You will receive an email when the job is complete. 

How to interpret the Run log

Once your status has changed to "Running", if you keep the window open after purchasing, or if you click the "Running" link on the main /UAbackups page, you can see a run-log.  

If your backup says "Backup Run Started" in the log, or says "Backup Status: Running" at the top of the backup log window, it is in the queue waiting to begin loading partitions. 

If the run has started, you can measure progress by the number of partitions loaded vs the total partitions planned when it gets to loading partitions.  For a while it may simply show you the query plan and validation of each of the queries you entered. 

The log may show 1 or more failed partitions.  These are requests to the API that returned an error.  The backup utility will hold these until the end, then retry them in accordance with API rules and limitations.  If there are still failed partitions at the end of the run, you will be emailed and given the opportunity to continue retrying or to remove that partition from the load.  In the vast majority of cases, retrying after a long delay tends to result in successfully loading failed partitions.  Contact support if you have retried more than twice and are still getting a failed result. 

Image

If the log says "The data for this backup has been backed up to your Analytics Canvas account. The job is now queued for export to the data source you defined", it means that the data has been backed up to your Analytics Canvas subscription and is safely downloaded.  Our next step is to deliver the data to you, which may not take place until early July. 

Image

What does "Pending" / "Purchased" / "Running" / "Failed" etc. mean? 

If you have created a backup, it will have a status associated with it.  This status will tell you what point you are at in the process.  The status messages are at the bottom of the Step-by-Step guide.
StatusDefinition
Beta ConfigThe backup was configured during the beta period. It is no longer editable. Copy it to put it into the Pending state, delete the Beta Config backup, then rename your copy to the original filename.
PendingThe backup configuration has been started but has not been run. You can continue to edit the backup and you can delete it. You can also run a Preview or if ready, make your purchase and initiate the backup process.
PreviewA Preview has been requested and is now available. Click the link of the word "Preview" to bring up details of BigQuery dataset or download the preview tables.
PurchasedThe backup has been configured and paid for, but has not run yet. You can no longer change the View, the Time Period, or modify any tables that have been exported. You can add tables, the dashboards, and you can duplicate the backup when it has this status.
RunningThe backup is currently running. Clicking the Edit button will bring up the status of the backup.
Paused for QuotaThis is either the daily quota limit of 10,000 API calls per day, OR an hourly limit on retries if a particular table or partition failed.
CompleteThe backup has been run to completion. You can copy it, edit it to add new tables, but you cannot modify tables that have already been run.
FailedThe backup run has failed. Refer to the error messages and action accordingly. Most often you simply need to re-run.

When will you stop taking orders? 

UPDATE June 27th:  We have stopped taking orders to focus on delivering data for backups that were ordered on or before this date.  Thank-you!

Can I order in bulk, or do I need to purchase each backup separately?

Placing multiple individual orders for the same quantity of the same product might raise the suspicions of your Credit Card company or financial institution.  We have seen a number of orders flagged for security review, blocking new orders from being placed.  

If you know you will be making more than 1 purchase, it is best to make a bulk order and add the credits to your account.

Image

What happens if my backup doesn't finish before July 1st? 

Due to overwhelming demand, we are filling orders on a 'best-efforts' basis. If your order is not filled, you will be refunded completely. If your order is partially filled, you will only be billed for the total rows delivered.

What are the new limits on order size?

We will need to review any orders that are over 80M rows.  Please ensure your backup is efficient by following this guide on our pricing page. You can then reach out to support to request an exception.  We will assess and determine whether we can even start the job.  If the job starts, it will be delivered on a best-efforts basis. 

Can I get a refund?

YES! If your order hasn't started yet, we can pull it out of the queue and allow someone else to get through.  Please email support@analyticscanvas.com with your order number, subscriptionID, name of the backup, and the email you used to create / login to your account. 

If your order delivered, but you're having issues with Sheets or Looker Studio report generation, please allow us to deliver that in July before requesting a refund.  If your data has been downloaded, it is a simple technical issue in transferring the data from your Analytics Canvas account to your BigQuery or Sheets and generating the Looker Studio report link. 

I'm having trouble with BigQuery

Ensure that you have BigQuery setup for use with an application and that you have the correct permissions applied.  Note that there are TWO places where the Service Account must be authorized, the Project and the Dataset.

Image

One or more of my jobs Failed, what should I do?

If you received an email that says your job failed, and when you login you see the failed notification next to a backup, click the link that says "Failed" and follow the on-screen prompts.

Image

You will first be shown a summary of the error. Click "Manage Detailed Errors"

Image

If you see a list of tables that have failed, it is because one or more partitions, or the entire table itself was unable to load successfully at the time the job was run. 

The Backup Utility will have retried the maximum allowable times before showing you this error message.  Retry each table at least twice before contacting support or excluding the table from the backup if it will never succeed. 

Image
⚠️

There are some queries which will simply not return data. This is a known issue affecting the UA API. If the same table fails repeatedly, you may have to exclude it from your backup.

I received an email saying my backup failed, but when I login it shows complete or running.

Our operations team is working tirelessly to get backups over the line.  If we noticed your backup had failed and it doesn't require action on your part, we are manually retrying them.  Often this is all it takes to get your backup over the finish line. 

If you login and your backup says "Complete", please verify the outputs and your Looker Studio reports.  

If your backup says "Running", wait for the next email and check back. 

My estimates seem counterintuitive - what is your policy?

Our estimate algorithm is imperfect and was trained on a limited number of Views over a short time period.

If a view has an unusual traffic pattern, is heavily weighted towards the end of the backup period, or has other anomalies in it, the estimates may be counterintuitive when run under different scenarios. 

Our policy is that your final price will be the lower of the estimate or the total rows delivered. Refunds for cases where the delivery was lower than estimated will begin in July 2024. This is because you have until the end of June to add tables to the row limit you've paid for.

Can someone get on a call and help me?

Unfortunately our calendar is now booked through the end of June.  We will be starting webinars soon.  For now, please contact support if you have any questions. 

Can you review my backup and tell me if it's missing anything?

We've provided a generic Table Library and Looker Studio report template that serve as the starting point for your backup. Only you / someone intimate with the website can know if the backup is complete. Please review our detailed guide on what to consider for your UA backup.

Why do I only see 14, 26, 38, or 50 months of history when I can see more history in the UA web interface?

At some point in the website's history, an administrator enabled a Data Retention Policy that asked Google to remove specific data beyond a 14, 26, 38, or 50 month time period.  Unless your setting was "Do Not Automatically Expire", you will only have detailed data for the time period selected.

Not all dimensions and metrics are affected by this policy.  This is why when in the web interface you can often see sessions or pageviews that go back to the start of the website, but when making detailed queries, you do not see any data beyond the retention period.

Google has not published a list of affected dimensions, nor do they show in the meta-data whether a site has a data-retention policy in place.
https://support.google.com/analytics/answer/7667196?hl=en

Unfortunately, Google does not publish an API for testing which queries would be affected, nor is there an indicator on your account or in the response delivered.  The only one who can know about the data retention policy for a specific view is through the UA web interface.

For more details about data retention policy and how to find the policy set for your website, review this article.

Why do some tables have missing or incomplete time periods?

Certain tables contain only data that was collected - they do not return zeros for time periods where no data existed.  

Examples of such tables include: 

  • Audience Demographic and Interest
  • Ecommerce
  • Google Ads 
  • Events
  • Custom queries where custom dimensions were applied 

For the data categories listed above, the minimum and maximum date in each table may not align with the start and end date of recorded hits for the View.  

For example, if ViewID has a start date of January 1st, 2009, but eCommerce wasn't enabled until April 2020, there will be no e-commerce data prior to April 2020.  If the site then stopped e-commerce tracking on June 30th, 2023 when they switched to GA4, but the website continued to track hits under UA until October 2023, there will be no e-commerce data from July 1st 2023 onwards. 

What do I do when my Analytics Canvas trial expires? 

Simply convert your trial to a UA backup plan.  This plan is free of charge and allows you to continue making estimates and running backups.


Looker Studio UA Backup Report Issues

Why does my report contain demo or sample data? 

  1. If you selected Google Sheets as the destination and one or more tables were too big to fit within a Google Sheet, you had the option to continue to download and receive those tables as a CSV file, or to reduce the time period so that the data would fit within Sheets.

  2. The template didn't generate properly. In this case, the Looker Studio report generation URL, found under "Exports", doesn't include all of your tables.  This can happen if the load to Google Sheets  fails or if the data was too big for Sheets.

In the first case, you must manually replace the data sources in Looker Studio with the CSV files that were exported.

In the second case, contact Support.  There is a known-issue with Sheets where the export could fail to deliver.  As the data is securely in your Analytics Canvas account, it can be re-exported once the export issue is resolved.  Contact support for further assistance.

Why does my report only contain a limited amount of data?

Your View's Data Retention settings determined how much data would be available. See the above section on Data Retention

How do I manage access and ownership of the reports?

The options for sharing your Looker Studio reports are outlined in this article.

How do I generate my Looker Studio report?

Once your backup is complete, you must generate your Looker Studio report from the template.  You have until December 31st, 2024 to complete this part of the process.

I see the error "Failed to parse input string" in my Looker Studio report

In Looker Studio there is a bug where the data type of a given column in a data source will change from the correct type to a Date data type.  When that happens, the user will see the error "Failed to parse input string".  The solution is quite simple and outlined below.
Image

If you see the error above, click EDIT on the report, thego to the Data Source and identify the column that is causing the issue. 

Image

Edit the data source, then change the data type from Date to Number.

Image

Edit the metric setting on the scorecard. 

Image

Finally change the aggregation type from Count to Sum.

Image

The component should now load correctly. 

Changing the Date Range on All Time Site Metrics

In some versions of both the Sheets and BigQuery templates, the "Backup Start Date" and "Backup End Date" may not reflect the full values.  See the video below for instructions on how to reset it. 

Was this article helpful?

Related Tutorials/Video