department-of-veterans-affairs / va.gov-team

Public resources for building on and in support of VA.gov. Visit complete Knowledge Hub:
https://depo-platform-documentation.scrollhelp.site/index.html
284 stars 206 forks source link

Analytics Implementation or QA Support for Debt Resolution Portal - Referring traffic (URL) reporting #95669

Closed Tom-Davis closed 1 week ago

Tom-Davis commented 1 month ago

Google Analytics Implementation Request

Please provide the following details for this request

Please leave the following blank

Acceptance criteria

Additional information

Please refer to Platform Collaboration Cycle or the Analytics Request Touchpoint on Platform website for more information about the Collaboration Cycle.

johnny-jesensky-adhoc commented 1 month ago

Thanks @Tom-Davis -- our team will review this, and let you know what the best options will be here. We'll communicate further updates in this ticket.

FYI @jestutt, @ChristineYoungZA, @Blivaditis

jestutt commented 1 week ago

Hi @Tom-Davis There are two parts to what you've asked for:

  1. Finding the sources in GA
  2. Creating tracking links for external placement to reduce the unknown/better identify external sources

Finding the sources in GA

Before we start, this is a direction analysis and we won't always be dealing with real numbers.

When looking at external and internal traffic sources for a single or group of pages, you have to come at the information from a few angles in GA and manually combine them. I've created Debt Resolution Sources [A&I] Explore report to get you start.

Tab 1 Total Views

Total number of page views generated across all pages with URLs that start with www.va.gov/manage-va-debt

Tab 2 Internal Sources

Page view count for all www.va.gov sources

  • based on Page referrer, where the user's browser says they were immediate before viewing pages with URLs that start with www.va.gov/manage-va-debt
  • excludes any referral traffic from the same section (sometimes called self referrals)
  • excludes any referral traffic from auth login pages -- to finish the login process on a page, the users has to have started the process on the same page which means it's a self referral

Tab 3 External Sources

Page view count for all non-www.va.gov sources

  • also based on Page referrer, where the user's browser says they were immediate before viewing pages with URLs that start with www.va.gov/manage-va-debt
  • Session default channel group is a high-level grouping system for external sources based on a number of values, mostly including Page referrer and UTM tracking URLs
  • Session source / medium provide the more specific source information like which search was used or which site they came from

Tab 4 Link Clicks

Event count for all tracked link clicks on www.va.gov to a /manage-va-debt destination link

  • you can add other metrics to find placement information if needed
  • excludes any events that happen on pages with URLs that start with www.va.gov/manage-va-debt

Creating tracking links for external placement

On Tab 3, you'll see that the Direct, or (direct) / (none), makes up more than 40% of the traffic sources. In these cases, direct = unknown and generally come from a non-HTML source like a bookmark, link typed into the browser, link click from an email or document, etc. To help reduce these unknowns in GA and provide better information on any external link placement we can control, we can add UTMs to our URLs to make them trackable.

Google provides a tool to assist in generating these trackable -- https://ga-dev-tools.google/campaign-url-builder/

Image

You must complete the fields with an asterisk (*) when creating these URLs or risk incomplete information in GA.

There is one optional field I recommend using if you need to differentiate placements like the type of letter a qr code is in or the pictures used in a social post.

Tip: I strongly recommend keeping a record of any tracking links you create somewhere. It will come in handy when you're ready to start reporting on any traffic these links have generated.

cc @johnny-jesensky-adhoc