Forum Moderators: DixonJones
1. Do email clients predictably send referers when links are clicked in an email?
2. Has any one used "web bugs" (1x1 gifs) to track whether emails get opened? Is it a dependable method - or is there something better that would track text-only email clients as well?
>to track whether emails get opened
I use a specific graphic for a specific e-mail so I can track, as close as possible, the number of times it's loaded. In addition, obviously, I use specifc click through addresses to track the success.
It's interesting to find that many people on dial-up do read the mail, but, the graphic will not load as they offline when they read the e-mail.
By keeping specific graphics and specific click through addresses, it does make it easier to measure success, although it's not perfect, for the reasons I've outlined.
That's what I'm after.
Our entire Newsletter is online, and the email we send is only a short tickler message. So I'm thinking I'll just add a variable to the end of the URL and then pull the numbers out of the server logs.
etc.
Site logs/stats will show you the info you need - just make sure you don't link to ../trackissueoneclicks with anything else or any other use, and stop spiders from accessing.
In the ../trackissueoneclicks you can include anything you want, but, my recommendation is to keep it simple and link to your online newsletter.
Also, see this thread [webmasterworld.com...]
My purpose is to track the effectiveness of the email copy in drawing the click. But this newsletter is blessed with some avid fans who take the link from their email and post it on their own site, and share it in other ways, as well (forums, their own emails, etc).
That will inflate the numbers, and I don't see a way around it. I certainly wouldn't do anything to discourage the practice.
Filtering for the date would probably help -- most email clicks would come soon after the mailing, and the clicks from the viral marketing would come later.
You should see the e-mail click with the "mailbox" or from a "disk drive." Also, the forum click will be easy to spot, if you focus on simply ../trackissueoneclicks
The graphic will usually load at the same time as the click. A url click from a forum will not show the graphic that was included with the e-mail newsletter.
Also, I drop my logs into a db and can sort and search the data easier than staring at raw logs, especially if you've got a complex structure or many clicks to track. You could probably, just as easily, drop the data into a spreadsheet.
I encode the time of sending and MD5 encode the user account, then hash it. I can quickly get the time out, and can always MD5/hash user account names when they click.
Once the link is activated the refer is logged to see the source, as engine suggested it, I also flag it that next time it is recorded as a separate record. This will tell me if the link was pasted into somewhere else.
I also use the above method to verify user signup and valid e-mails. To activate an account they must click in a link they recieved in an e-mail. Not perfect but does work better then trying to decode an e-mail.
