In a sentence,
Distributed query rate limit exceeded means that your Funnel account is overworked.
The amount of work required to satisfy all the queries that are being made for a single account over a 60 second period is higher than our limit, so some of the requests are being throttled. This includes all queries being made, no matter where they come from, for example Google Data Studio, Data Explorer or Data Warehouse exports.
The limit is not directly related to the amount of data being returned to the report, it is rather the number of simultaneous queries and the amount of work required to satisfy them.
What can be done?
Tune your report
Google Data Studio sends at least one request to Funnel for every widget. Having a page with many widgets on it will result in many requests to Funnel, and there is a higher chance of being rate limited.
- Consider having fewer widgets per page and having more pages instead.
- Consider whether you can use a shorter time period.
- Try removing 'comparison' time periods.
- Filtering the report on 'Data Source' or 'Data Source type' may also help.
Refresh the report
This might seem counter-intuitive, but refreshing the report might help. We cache responses for an hour, so requests that have been satisfied in the last hour will be fulfilled from the cache and not be included in the rate limit.
For example, if your report has ten widgets, when you first load the report, 8 might get data and the other two get the rate limit error. If you refresh, the 8 will get cached data and the two that did not get data before should now get data, as there are fewer requests being included in the rate-limiting.
Time your report
As mentioned, the rate limit is for all requests over a 60 second period, so avoid loading your report during peak times if you have other requests with Funnel at that time, for example if you have many Google Sheets exports that run at 8 AM you are more likely to get rate limiting errors in your report if you load it at the same time.
View extracts are an experimental feature we are trying out to help with this issue. You create a view containing a subset of your fields and an optional filter, and we create an extract of that data. Your report can then query that extract, which has no rate limit imposed. You should see a significant improvement in performance and stability.
Contact us if you want to try out view extracts, or read more about view extracts.
Another option is exporting your data to BigQuery (or another Data Warehouse) and connecting that to your reports. This might be a good solution if you have very large amounts of data in Funnel. Contact us if you want to discuss this option more.