Skip to content
This repository was archived by the owner on Aug 27, 2021. It is now read-only.
This repository was archived by the owner on Aug 27, 2021. It is now read-only.

Click tracking docs does not explain plain email behavior #5611

@brynek

Description

@brynek

Expected content

A description that eludes the actual click tracking link substitution behavior for plain emails.

After investigating issues we had with bad activation links being sent to new users, we found that plain e-mails were the source of the problem. If link tracking is enabled (link tracking in plain e-mails is included by default), all links in plain e-mails seem to substituted (it looks like your parser is picking up everything with https:// or http:// in plain e-mails). Including a clicktracking=off in the HTML has no effect on the generated plain e-mail, and the substitution is still made.

Furthermore, there is no description of the "enable click tracking in plain emails" toggle that you have made available in the UI. This luckily served as our fix, without having to turn click tracking off entirely.

Actual content

No mention that using clicktracking=off in the HTML has no effect on click tracking in plain e-mails. Right now, the wording makes the reader believe clicktracking=off also would have an effect on the plain e-mail.
No mention of the "enable click tracking in plain emails" on/off toggle in the UI
No mention of how link substitutions are done on plain e-mails

Link to page:
https://sendgrid.com/docs/ui/analytics-and-reporting/click-tracking-html-best-practices/
https://sendgrid.com/docs/ui/account-and-settings/tracking/#click-tracking

Why did this cause an issue for us?

For Sendgrid developers:
Your link replacement process for the plain e-mails seems to html encode the links, which, when we are using multiple query string parameters leaves the redirected link invalid. For instance, the verification link https://foo.bar.com/verify?id=foo&token=bar is substituted with a sendgrid click tracking link redirecting to the html encoded https://foo.bar.com/verify?id=foo&token=bar (which resolves to the invalid https://foo.bar.com/verify?id=foo&amp%3Btoken=bar). This means that anyone who opened the plain-version of our transactional e-mails, only got invalid verification links. Of course we could implement an html decoder in our router, but this behaviour on your side seems like a bug to us. Same with the clicktracking=off clauses in the HTML - this should also lead to non-substituted links in the plain e-mail.

Metadata

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions