Reach reports

This article is about running reach reports if you do not have access to query tool functionality. It may not be applicable for your DFP environment. For more details, contact your DoubleClick representative.

To learn how to use reach reports using query tool functionality, see Reach reports.

Use reach reports to determine the number of unique visitors exposed to different advertisers, orders, line items, or ad units in your network over a given time period. Note that a single visitor who accesses your site using multiple browsers may be counted as multiple visitors, since there's no way to recognize that it's the same visitor across each browser.

When scheduling reach reports, keep in mind that they are generated when the day has ended according to Pacific Time and that they need a longer time to generate than a regular delivery report. Consequently, we recommend that you schedule monthly reach reports so that they run on the third day of a new month instead of on the first day.

How does Google calculate reach?

Google estimates the number of unique visitors for your campaigns and inventory based on data from a specific set of cookies. The cookies in this set were all "born" before the start date of the order, line item, or query in question.

Cookies that are born just after the start date of the order, line item, or query, are more likely to be from an expired or deleted cookie than to be an actual new user. We avoid looking at statistics of cookies that are less likely to represent an actual new visitor.

For example, if you were looking at reach for a line item during a particular month, the data used in the calculations would be based on the impressions that were generated by cookies meeting this specific criteria.

To calculate reach using this example, let’s say that visitors with cookies meeting this criteria saw an ad an average of 5 times each, and there were 1,000,000 total impressions for the ad. The reach is calculated by dividing 1,000,000 by 5 to come up with a reach of 200,000 unique visitors.

See a detailed example with calculations

Below is some data for a sample campaign, shown in 30-day increments. The initial cookie set is collected and remains the same for reach calculations throughout the campaign.

Time period Impressions for this time period Total campaign impressions to date Total cookies in the cookie set Impressions generated by the cookie set
30 days 1,000,000 1,000,000 100,000 500,000
60 days 1,000,000 2,000,000 100,000 100,000
90 days 1,000,000 3,000,000 100,000 0

As illustrated by the sample data, impressions generated by the cookie set decrease over time. This is expected and attributed to historical cookie expiration data (e.g. approximately 50% of cookies expire after 30 days).

Calculating reach

When calculating reach, the figures are based on the impressions generated from the cookie set. To calculate the impression average from that group:

   Impressions generated by the cookie set / Total cookies in the cookie set

Google assumes that the average impressions generated by cookies meeting the criteria is equal to the impressions generated by unique users:

   Average impressions by cookie set = Average impressions by unique users

30-day calculations

Using the 30-day data above, the average impression count in the sample group is calculated:

   Impressions generated by the cookie set / Total cookies in the cookie set

   500,000 / 100,000 = 5

Using the 30-day data above, the total unique user count for the given time period is calculated:

   Total impressions / Average impressions by unique users

   1,000,000 / 5 = 200,000

The reach for this 30-day period of the campaign is 200,000 unique users.

60-day calculations

Using the 60-day data above, the average impression count in the sample group is calculated:

   Impressions generated by the cookie set / Total cookies in the cookie set

   (500,000 + 100,000) / 100,000 = 6

Using the 60-day data above, the total unique user count for the given time period is calculated:

   Total impressions / Average impressions by unique users

   2,000,000 / 6 = 333,333

The reach for this 60-day period of the campaign is 333,333 unique users.

90-day calculations

Using the 90-day data above, the average impression count in the sample group is calculated:

   Impressions generated by the cookie set / Total cookies in the cookie set

   (500,000 + 100,000 + 0) / 100,000 = 6

Using the 90-day data above, the total unique user count for the given time period is calculated:

   Total impressions / Average impressions by unique users

   3,000,000 / 6 = 500,000

The reach for this 90-day period of the campaign is 500,000 unique users.

Growth in reach number vs growth in sample group impressions

Using the calculations above, and as evidenced by the 90-day calculation example, even if no new impressions are generated by the sample group, reach should continue to steadily increase. For extended campaigns, line items or queries, the actual rate of increase may slow down over time.

The graph shown below is an illustrated example of the expected upward trend of reach values, even if no further new impressions are generated by the cookies matching the activity criteria.

 

Please note that the reach report does not include impressions from Ad Exchange or AdSense.

Generate a reach report

To view a reach report:

  1. Click the Reports tab then, under the Reach link, select the type of report you'd like to see (either advertisers, orders, line items, ad units, or entire network).

  2. Use the date picker to select a date range:

    If you're viewing an order or line item report, you must select Month or Week if you want to customize reporting dates.

    If you want to schedule the report to run automatically, select Week for weekly reports and select Month or Month and year to run it monthly.

    If you change the reporting dates, the report won't refresh on its own; you must first click the Done button.

  3. Click the edit report link next to the report name to expose all reporting options, then select your reporting criteria dimensions (rows) and metrics/fields (columns) to define the information displayed in your report.

    If you are scheduling the report to run automatically, ensure that you've selected either Week for weekly reports or Month or Month and year for monthly reports.

  4. (Optional) Select whether to display reach data for all items, or for some subset (for example, all orders or line items, or only orders or line items that have started or completed).

  5. You can save, share, or schedule your reach reports, just like any other report.

 
Data availability for reach reports

Data is available for inclusion in reach reports from July 1, 2010 to yesterday's date. You can break down reporting data in any of the following ways:

  • Weekly breakdown: View reporting data for each calendar week in the selected full (past) weeks. This option is available at the advertiser, ad unit, order, line item, or entire network level.

    A calendar week is the period spanning from Sunday to Saturday. If you select a different time period that spans multiple calendar week boundaries, the resulting report will automatically shift the start and end dates to match those of the closest calendar week. For example, you might select a start date of Friday the 16th and an end date of Thursday the 22nd for your report. In this case, the start date is rounded back to the previous Sunday (the 11th) and the end date is rounded ahead to the following Saturday (the 24th). We recommend always selecting a Sunday start date and a Saturday end date.
  • Monthly breakdown: View reporting data for each calendar month in the selected full (past) month. This option is available at the advertiser, ad unit, order, or line item level. It is the default setting for advertiser and ad unit reports.

    As with the weekly breakdown, the start and end dates for the report will automatically shift backward or ahead if you select any dates other than the first and last of the month.
  • Lifetime: Lifetime reach reports display data for the last six months of an order's lifetime (or its entire lifetime, if it ran for less than six months); beyond six months, it's difficult to calculate an accurate value for reach. This option is available for orders and line items and is the default setting. If the order or line item stops running, the number remains fixed in place.

At this time, you cannot request a custom time period or a real-time report for reach reports. Reach statistics are based on large amounts of data that need to be processed every time a new report is generated for a defined time period. Reach calculations are updated once a day in the evening (PT).

Reports containing "N/A" or "-" values

If you see "N/A" or "-" in your report, it means that there's not enough data to accurately calculate your reach for that row. There are two general cases when you may see "N/A" or "-" in a reach report:

  1. Reach data for any given period is only shown if the calculated error rate is <0.25. This is to make sure we only show data that we're confident is correct. If the error rate is greater than that, reports will display reach as "N/A' or '-'. Because reach reports are based on sampling, the error rate tends to be higher when the sample is smaller, so higher numbers of unique users and generally higher traffic will lead to a reduction in the number of unavailable values.

  2. Aggregations in reach reports are disabled, and show '-' because adding up the unique visitors doesn't make sense. For instance, if you have 2 line items in an order with 100 unique visitors each, and you run a report by order x line item, the number of unique visitors for the order is going to be somewhere between 100 and 200. Since there may be duplicates between the line items, it most likely won't be 200, so the aggregated row for the order will show '-'. In order to see aggregates, you can run a report with the order dimension only.