1. Home
  2. Knowledge Base
  3. Universal Analytics Backup
  4. How to estimate the cost of a Universal Analytics Backup?

How to estimate the cost of a Universal Analytics Backup

This article discusses how to estimate the cost of your UA Backup using the UA Backup Utility provided by Analytics Canvas.  The utility exists within Analytics Canvas, however no subscription is required to use the UA Backup Utility.  

As some site are enormous, with billions of sessions, while others are small, with just thousands of sessions, the size of the backup and the amount of data processing required will vary considerably.

Pricing is variable based on the total amount of data returned, determined by a sum of the total rows delivered in each table.

Generate an estimate that is within your budget by following the suggestions below. 

⚠️

There is no 'one-click' download of Universal Analytics Data. Each site must determine what data they want to backup. The UA Backup Utility generates a set of summary tables for use in a Looker Studio report template, a series of default tables in a Table Library that are available for inclusion and customization, and a custom query builder to supplement or create your backup.

No Subscription Required!

While the utility exists within Analytics Canvas Online, there is no subscription required to use it since each backup is priced individually. 

Analytics Canvas Starter, Pro, Premium, and Enterprise customers have the option of creating their own backups manually using the main application.  There are no additional fees required in this case.  However, customers using the backup utility to save time and ensure a successful load on larger backups, will be charged the same pricing as non-subscribers. 

Pricing

Pricing is $198 per View for up to 20 million rows of data (that's a lot of data!).  

For that you get:

  1. The Looker Studio report
  2. Access to the Analytics Canvas Table Library for more detailed tables to include in your backup,
  3. A custom query builder to add your own tables.

The Looker Studio report is yours to keep.  You can customize it, white-label it, remove one or more pages, etc.  The report and data are yours and in your Google Accounts (Looker Studio, Drive, or Google Cloud Platform), with no further obligations to Analytics Canvas.

For sites that don't want the Looker Studio report, simply deduct $99 and 10M rows from your price.

For sites that need more than the included rows, either for the dashboard tables or custom tables, additional data is available at a rate of $99 per 10 million rows.  

Bulk pricing is available for agencies and large sites.  See below for details.

  Most websites can get a reasonable backup within 10M rows!

Pricing Examples

  1. A backup is created that ONLY includes the Looker Studio report - no supplemental tables.  The row estimate is less than 10M rows.  Pricing is $198. 
  2. A backup is created that does not include the report - just custom and supplemental tables.  The row estimate is less than 10M rows.  Pricing is $99. 
  3. A backup is created with the Looker Studio Report, supplemental tables, and custom tables.  The row estimate is 28.5M rows.  Pricing is $297. 
  4. A backup is created with only custom tables. The row estimate is 91M rows.  Pricing is $990.

Price Guarantee

You will only pay the lower of the estimated price and the total rows processed for your backup.

If the total rows delivered is more than the estimated rows, there is no additional cost.

If the total rows delivered is less than the estimated cost, we will refund the difference between the estimated price and what the actual price should have been.  

Refunds will be issued no later than 30 days after the UA API is turned off. 

Bulk Pricing

Bulk pricing is available for individual backups that exceed 100M rows.  Pricing drops to $69/10M rows for backups between 100M and 200M rows, and then to $39/10M rows for anything beyond 200M. 

For example:

  1. 99M row backup: 10* 99 = $990 
  2. 175M row backup: (10*99) + (8*69) = $1,542
  3. 500M row backup: (10*99) + (10*69) + (30*39) = $2,850

For agency pricing, contact-us.

Estimating the row count in a backup

Rows are the best proxy available to determine the amount of processing required for your backup.

A popular news site with 1M monthly readers will generate billions of hits, require potentially thousands of API calls, and would take a very long time to run.  Whereas a mom-and-pop e-commerce shop can have their backup done in just a few minutes with 100K rows or less. 

The amount of rows delivered will vary based on:

  1. total number of sessions that visited the website
  2. total number of hits that those sessions generated 
  3. the years of history of data required / retained 
  4. the number of tables and the combination of dimensions in each table 
  5. the time granularity of each table (day vs week, month, or year)

The rows dowloaded may be greater than the total rows delivered to the dashboard tables.  Some dashboard tables are summarized after some light data-prep (such as striping parameters from page paths). 

Generating an Estimate in 10 Mins or Less

We have built a really easy tool for estimating the size of a backup.  Once you have your account setup, you can generate a backup estimate and refine it in just a few minutes. 

Here are the steps to generate a quick initial estimate:

  1. Go to https://cloud.analyticscanvas.com and login. If you don't have an account, create one.
  2. Go to UA Backups on the left navigation menu.
  3. Create a new backup, providing a name, selecting the View ID, and defining the time period you want backed up. You can also select to include the Looker Studio report in this setup wizard. 
  4. Select which data destination you want, including BigQuery, Sheets, Excel, or CSV.
  5. Select a ViewID to backup and click Next.

The table library will generate an estimate for you at this stage. This is only an initial estimate.

Image

The initial estimate may be higher or lower than you anticipated, and might be higher or lower than your budget.  Finalize the estimate by making the changes outlined below. 

For complete documentation on how to generate an estimate, refer to this article

🛑

The initial estimate is based on the default selections made when creating the backup. No backup is complete until redudant tables are removed, tables are customized with custom dimensions and metrics, and custom tables are built to satisfy specific requirements.

Finalizing Your Price

You can make the following changes to affect the price of your backup:

  • Remove redundant tables. For example, there are 4 goal tables included in sets of 5 goals each. Remove tables which do not apply to your site. 
  • If selecting dashboard tables and data tables, ensure that there is no overlap. 
  • Change time periods from Day to YearWeek, YearMonth, and / or Year on one or more tables.
  • Change level of detail on tables from the Table Library from Low, to Medium, to High
  • Modify the tables provided to include custom dimensions and metrics, events, or goals as you see fit.
  • Add your own custom tables.
  • After making customizations, use the sort and search fields to identify tables that you do not want or need, then delete them from your backup.  

After some fine-tuning, the vast majority of websites can get a backup for $198 or less.  Those who want or require more detail have the option of purchasing more data in 10M row increments. 

The best approach is to start with a budget. Talk to the client or marketing leads about what they’re willing to spend, and how much detail they would like. 

There is a direct relationship between level of detail and number of rows.

Sites with tens of millions of sessions can produce a backup with just a few million rows if they want their data aggregated by year instead of day and choose low detail tables.  

Sites with just a few million rows can have tens of millions of rows to backup if they want clientID, hitTimeStamp, and other high-cardinality dimensions.

FAQ

Can I make multiple estimates?

Yes! Make as many estimates as you'd like, there is no charge to generate an estimate or to run a preview for a View.

How can I reduce the price?

To reduce the price, modify or remove tables to reduce the total row count.  You can also reduce the time period from the full history (the default setting), to a shorter time period that you're comfortable with.  You may also change tables from Daily to Week, Month, or Year to get a more aggregate view with less overall rows.

What kind of training and support is available?

Our knowledge base contains a number of articles to help you get up to speed with the backup utility and generate your backups.  If you need additional support, you can reach out to our support team and they would be happy to help.

What if I have multiple Views to backup?

You can make as many estimates for as many views as you'd like.  Pricing for multiple views is shown in the application for each backup specification.  For bulk orders or those with lots of sites, contact-us.

Do I need a subscription to Analytics Canvas?

No! A subscription to Analytics Canvas is not required to use the UA Backup Utility.  You will be placed into a 30 day trial when you first sign-up.  This allows you to use the main application together with the Utility during an evaluation period.  You can verify your backup, QA the results, or blend UA and GA4 data together during this time without a subscription.

After the 30 day trial, the subscription automatically converts to a UA Backup plan. Under this plan, there is no monthly subscription fee.  You pay only for the backups you run.  You will no longer have access to the main application, unless you subscribe to a Starter, Pro, Premium, or Enterprise plan.

Do I have to place a separate order for each backup?

No - you can add credits to your account in bulk and apply them on a case-by-case basis. Contact-us for more information.

Are there any ongoing or hidden fees?

Not from Analytics Canvas.  You are responsible for your Google Workspace account for the Looker Studio report and Google Sheets, and your own GCP account for the data in BigQuery.  Costs for holding the data in BigQuery are minimal, as outlined here.

 
 
Was this article helpful?

Related Tutorials/Video