Here are the questions we hear most often about the Digital Analytics Program (DAP).
1. What tool is being used to collect the data?
2. Who can use the tool?
3. Can my agency use it for free?
4. Is a Gmail or Google account the only way to access this data?
5. Does this tool have to go on all websites? Or just top-level domains?
6. Will agencies need to obtain a “cookie waiver” to use the tool?
7. Are the metrics we collect available to Google's corporate advertising partners?
8. With many people coming in via dynamic IP services, how can you tell the difference between new and returned visitors?
9. Where can I find benchmarks for things like bounce rate, returning users, unique visitors, etc.?
10. What kind of testing was done to ensure that the code doesn't cause conflicts with other script-based code already in use?
1. What tool is being used to collect the data?
We’re using Google Analytics Premium to collect and analyze the Web traffic data.
2. Who can use the tool?
The initial roll-out, scheduled for October 15, 2012, is limited to executive branch agencies of the U.S. federal government.
3. Can my agency use it for free?
Yes. There is no cost for federal government agencies to use this tool.
4. Is a Gmail or Google account the only way to access this data?
A Google account is required, but a Gmail email address is not—any email address can be used to create a Google account. You should sign up for the DAP using your work email address.
5. Does this tool have to go on all websites? Or just top-level domains?
6. Will agencies need to obtain a “cookie waiver” to use the tool?
No. Since we are stripping out user IP addresses, agencies do NOT need any special permission, waivers, or privacy impact assessments.
The cookie used is considered Tier 2 (multi-session without PII). This tier encompasses any use of multi-session Web measurement and customization technologies when no PII is collected (including when the agency is unable to identify an individual as a result of its use of such technologies).
See page 5 of OMB Memo M-10-22 (PDF, 102.75 KB, 9 pages, June 2010).
A privacy impact assessment and sign-off by the privacy official is not required until Tier 3.
Agencies should incorporate this information into their existing website privacy policy. For an example of the disclaimer language you can use, see
HHS.gov or
DHS.gov.
7. Are the metrics we collect available to Google's corporate advertising partners?
No. None of the federal government data tracked as part of the DAP will be shared with or available to Google’s corporate advertising partners.
8. With many people coming in via dynamic IP services, how can you tell the difference between new and returned visitors?
Because of dynamic IP addresses and browser cookie deletion, unique, and new vs. returning visitors metrics can never be 100% accurate, regardless of the Web analytics tool used. If an IP address is different (dynamic) each time a visitor visits a website, a visitor will be counted as new. Similarly, if a visitor deletes their browser cookies, he or she will be counted as a new visitor.
9. Where can I find benchmarks for things like bounce rate, returning users, unique visitors, etc.?
There are several reliable sources of competitive/benchmark data, including commercial providers like Compete, Hitwise, ComScore, and Nielsen, as well as free tools like Google Trends. Each source/solution has a specific methodology to measure Internet traffic, so their findings may not always be consistent.
There are also a variety of Web analytics blogs that publish the Google Analytics Monthly Benchmark Report, a report that Google Analytics users receive by email on a monthly basis if they opt-in to share data anonymously with Google. For example, Google Analytics Benchmark Averages for Bounce Rate are as follows:
-
40-60% Content websites
-
30-50% Lead generation sites
-
70-98% Blogs
-
20-40% Retail sites
-
10-30% Service sites
-
70-90% Landing pages
Ultimately, while there are many sources for benchmarking information, your Web data interpretation and analysis should be based on the website/department/agency goals that make sense within your organization. Industry-based benchmark percentages only make sense to measure against when they are in the proper context, and align to your specific website, the mission of your site, and other factors specific to YOUR needs and goals.
10. What kind of testing was done to ensure that the code doesn't cause conflicts with other script-based code already in use?
Preliminary testing has been conducted to ensure that the government-wide code script does not conflict with other script-based code. However, it's impossible to test every possible scenario when the possibilities are unknown, especially with custom code already deployed on agency websites. To ensure that the government-wide code snippet does not conflict with the other existing script-based code, each implementation will be handled uniquely, driven by the agency Web infrastructure, existing code, etc.
