Reach & Frequency Reporting

Owned by Abhinav Rao
Last updated: Mar 30, 2022

Reach & Frequency Reporting is now available in SpringServe by turning on the Reach & Frequency Reporting beta feature. With this feature, clients have the ability to run reports that show metrics the number of users they reach (namely the Reach) and how often they reach these users (namely Frequency) over time over supply & demand sources.

Reach & Frequency Reporting

Once this feature has been enabled for your account by your Account Manager, you will have the ability to run Frequency & Reach reports. On turning the beta feature for the account, users will see two new reporting templates available to run. These two reporting templates are called:

  • Reach Report

  • Frequency Report

The Mechanics of the Reach & Frequency Report Templates

This section explains the details of these two templates with emphasis on time parameters, dimensions, filters, metrics and interpreting the data.

Reach Report

The Reach report shows the details of the number of unique users, devices, ips and the impressions that have been served to these users. This report addresses custom use-cases where users can choose to run these metrics by custom time periods like Days, Weeks, Months or Cumulative.

Listed below are the specific considerations to be cognizant of when running this report:

Time Parameters:

  • Date Ranges: All Date Ranges available to existing Supply, Demand, Pod and Advertiser Domain Reports are available for this report.

    • The one difference is that if users select a custom time period, the max look back available to be selected is 365 days. Any time period beyond 365 days is ineligible for selection.

    • If users try to re-run a saved report for a time period that might exceed 365 days as of the time of running this report, then a message like the one shown in the screenshot below will be displayed

  • Intervals: All Intervals available to existing Supply, Demand, Pod and Advertiser Domain Reports are available for this report with the exception of hour.

  • Time Zone: This report can only be run in UTC.

Dimensions:

  • There are a few specific Demand, Supply Parameters are ineligible to be added to this report as Dimensions.

    • The ineligible parameters are greyed out when you select either of these report templates.

  • All Direct Connect, Programmatic Connect, Partner Segments Inventory (excluding Country) are ineligible to be added to this report as dimensions.

    • The ineligible parameters are greyed out when you select either of these report templates.

Filters:

  • There is a filter called Cadence that is selected by default and the value selected by default is All Days.

    • If this filter is deselected, kindly note that the metrics in this report view will change to that of a Supply Report.

    • To run this report, please keep the cadence filter selected and the value selected to that of All Days. This filter value is the main differentiator between this report and all other reports in SpringServe and thus, do not de-select this if you intend to run Reach & Frequency Data by custom time periods.

The screenshot shows the filters container for a reach report with the Cadence filter pre-selected and the value of “All Days” pre-populated.

  • There are a few specific Demand, Supply Parameters are ineligible to be added to this report as Filters.

    • The ineligible parameters are greyed out when you select either of these report templates.

  • All Direct Connect, Programmatic Connect, Partner Segments Inventory (excluding Country) are ineligible to be added to this report as filters.

    • The ineligible parameters are greyed out when you select either of these report templates.

Metrics:

Attached below is a screenshot highlighting all the metrics available to be selected in this report with the ones in teal being the default selections.

Numbers:

  • Imps: This is the sum of Impressions aggregated over the the report parameters defined by time range, dimensions, filters, and date ranges.

  • Unique Users: Count of Unique Users aggregated over the the report parameters defined by time range, dimensions, filters, and date ranges.

  • Unique IPs: Count of Unique IP's aggregated over the the report parameters defined by time range, dimensions, filters, and date ranges.

  • Unique Devices: Count of Unique Devices aggregated over the the report parameters defined by time range, dimensions, filters, and date ranges.

Calculations:

  • User Reach: Impressions/Unique Users. This is indicative of the number of impressions seen by each user.

  • IP Reach: Impressions/Unique IP's. This is indicative of the number of impressions seen by each user IP.

  • Device Reach: Impressions/Unique Devices. This is indicative of the number of impressions served to each device.

Financials:

Note that the financial metrics available to this report do not have third party fees and thus the financial metrics available are Revenue, Media Cost and their derivative financial calculations.

Conditions

Users can set conditions on this report to return only those lines which meet those conditions. For example, you can run a report to return data filtered by a condition where unique users >= 10.

Users can use the Quick Add buttons to select a metric or click the +New Condition button and a dropdown will appear. Select your desired metric from the dropdown and set the condition.

Below is a screenshot showing the first page of a results container for a report run on test data for the last month with the data broken out by day.

Frequency Report

The Frequency Report is almost entirely similar to the reach report with one key difference. All the mechanics detailed in the above section are applicable to this report and thus this section will highlight only the difference between this report and the reach report.

Time Parameters:

  • Intervals: This report can only be run for a Cumulative Interval. If cadence is selected as a dimension, the Interval will default to Cumulative.

Dimensions:

  • The default dimension selected is Cadence.

Cadence is a set of pre-defined intervals for users who want to run these reports by the pre-defined Cadence Intervals. All possible values of the cadence intervals are displayed in the screenshot in the next section.

Interpreting Cadence

The screenshot below shows all applicable values for Cadence. Note that the values for cadence displayed will be dependent on the data range selected.

Note that “Days 1-2” implies that the metrics are being displayed for the current day and two days ago. “Days 3-4” implies that the metrics are being displayed for the time period from three days and including 4 days ago. Note that the lower and upper bounds for each cadence interval are inclusive, which means that Days 1-2 is inclusive of data up to 2 days ago from the end date of the date range selected.

To enunciate this logic, consider an example report that a user runs the frequency report for the last 30 days on March 30th at noon UTC.

  • In this case, the date range considered for this report will be March 30th and then we will count 30 days back from March 30th, which for the sake of this example will be March 1st.

  • Secondly, the values displayed in the cadence will be limited to the date range selected.

    • In this case, the values displayed will be such that the lower bound of the cadence interval can not exceed the date range. Thus, since Days 31-60 falls outside the 30 day date range, this and all following cadences will not be displayed in this report.

  • Here is an output for the report parameters defined above:

  • Days 1-2 is limited to and including March 30th - 29th

  • Days 3-4 is limited to and including March 28th - 27th

  • Days 5-7 is limited to and including March 26th - 24th

  • Days 8-15 is limited to and including March 23rd - 16th

  • Days 16-30 is limited to and including March 15th - 1st

If users run this report for a custom date range in the past, this logic still holds true. So, consider that a user runs a report for the month of January 2022 but runs this on March 30th at noon UTC, then these cadence intervals above will be run but the interpretation of the cadence interval in this report will be thus:

  • Days 1-2 is limited to and including Jan 31st - 30th

  • Days 3-4 is limited to and including Jan 29th - 28th

  • Days 5-7 is limited to and including Jan 27th - 25th

  • Days 8-15 is limited to and including Jan 24th - 17th

  • Days 16-30 is limited to and including Jan 16th - Jan 2snd

  • Days 31-60 is limited to and including only Jan 1st

    • The reason for this is because the date range limits the cadence to only 31 days of data and thus the report will display this cadence interval but limit the data queried to just the 1st day of January.

Running Frequency & Reach Reports

When users run either of these reports, a modal will appear that displays a large Frequency & Reach report Modal with the following details:

Modal Title: Large Frequency & Reach Report

Text: This report is very large and will take over 2 minutes to run. You will be able to view it at the Download Reports page when it is complete. It will continue to be processed if you navigate away from this page.

Additionally, users can choose to add a name to this report.

The main thing to note is that due to the volume of data that needs to be processed to run these reports, these reports have a data lag of 8 hours and they will almost always take more than 2 minutes to run. Please note that there is a note highlighting this specifically in the results container while this report is being processed.

Users must also be cognizant of the fact that adding more incremental parameters to this report will increase the complexity of the queries used to process the data thereby increasing the processing time needed to run these reports.

Frequency Reports Edge-Case

There is one edge-case to be aware of in some cases where users can see a value for Cadence called “Interval_unknown” on running custom reports. If users see a value called 'Interval_unknown” this is indicative of data for the last day of the date range for the hours that fall outside 365 hours in a custom report.

Consider a Frequency report and reach report run for a custom date range of 03/29/21 00:00 - 03/29/22 23:59

The Reach report will show data like this for the last day:

And if you correlate this data with the Unknown_interval shown in the frequency report, these numbers will match.

To enunciate this data, 365 days back in the UI is 03/29/21 00:00 - 03/29/22 23:59 since the start date of 03/29/21 00:00 falls out of the 365 days before 03/29/22 23:59, it falls into interval unknown. To explain it another way, it falls out of the cadence intervals but it's in the start and end dates and thus, we report on it.

Running Frequency & Reach Reports via the API

These reports are available to be run via the REST API. Here are the conditions to be aware of while running these reports via the REST API:

  • Start date and End date are required.

  • Start date cannot be before 365 days ago

  • Timezone must be UTC

  • Interval must be cumulative when cadence is a dimension

  • Interval cannot be Hour when all_days cadence filter

  • Require async when using API since reports generally take longer than 2 mins.

Here’s an example API Call in our test environment with the Auth token hidden for running these reports after users have received a valid Authorization token.

curl --location --request POST 'https://staging-admin-console.springserve.com/api/v0/report' \ --header 'Content-Type: application/json' \ --header 'Authorization: ________________' \ --data-raw '{ "interval":"day", "timezone":"UTC", "start_date":"2021-08-26", "end_date": "2022-01-01", "demand_tag_ids":[24275], "async":true }'

It is important to note that these reports will not work in the API when the flag "async":true is not set in the CURL call.