Is it possible to detect if an email newsletter has been forwarded?

S. Dale Morrey sdalemorrey at gmail.com
Wed Feb 6 16:52:38 MST 2013


It really depends on the accuracy you want and what specific purpose
you are measuring this for.

The way I see it there are 2 reasons to know and in effect you are
really measuring 2 different things but trying to do so using the same
metric.

First position (These newsletters are secret subscribers only info and
we want to know who is violating their agreement).
There is no real way to do this in email except possibly a custom DRM
scheme wrapped around a .PDF that authenticates against a server you
control.
However if you do that you would ensure that 1 email = 1 user.

Second position (We want these newsletters to go viral and we want to
track whom is sharing with whom)
Easiest way is to send a unique email to each client with a unique URL
to click on.  This does not tell you if it was forwarded "per se", but
if 2 or more people click the same link, then you know it must have
been shared.  Another possibility is a web-bug (1px transparent gif)
and just count the unique IP addresses, if you want to get fancy you
could do the whole unique URL thing for each web bug and then it's the
same as above, but more automated.

 You could also look into requesting a read reciept, I've seen it
before but I don't know enough to know if that's a part of some
standard or it's just an MS thing.

On Wed, Feb 6, 2013 at 4:37 PM, Ken Snyder <kendsnyder at gmail.com> wrote:
> @John - Cookies are a good idea too. Looking at user agent, cookies and ip
> address do suffer from a common drawback: a user who opens their email at
> home and at work will appear as two different users.
>
> @Jonathan - I know exactly what you mean. Our clients are generally
> Outlook-entrenched corporate users who don't know what the phrase "plain
> text" means, but some do have images disabled and so we never know they
> open it. And of course the bigger problem is that seeing an email in your
> preview pane does not mean you read it :)
>
> Showing email stats to users has at least two major benefits: 1) It makes
> them feel good and 2) It allows them to gauge popularity *relative
> to*other emails they've sent. 1 is funny to me but sadly so many
> applications
> these days are more targeted to making people feel good than to giving them
> good features and content.
>
> - Ken
>
>
> On Wed, Feb 6, 2013 at 2:15 PM, Jonathan Duncan <jonathan at bluesunhosting.com
>> wrote:
>
>> I am probably not their target audience, but if I were, they would never
>> know that I opened the message because my client is set to text only.  If
>> their message entices me, I will click a link, if not, it gets deleted.  I
>> am probably not a good use case though.
>>
>>
>> On 06 Feb 2013, at 14:10, John Shaver wrote:
>>
>> > I've never worked with these type of emails much before, but if these are
>> > links they're clicking, couldn't you just save a cookie to their browser
>> to
>> > tell if they've been there before?  Associate the cookie ID with their
>> link
>> > ID.  Then if you get multple cookie IDs matching the same link ID, then
>> > it's probably  a forwarded email, or they opened it from multiple
>> > computers, or they deleted/didn't save the cookie.
>> >
>> >
>> > On Wed, Feb 6, 2013 at 1:43 PM, Lonnie Olson <lists at kittypee.com> wrote:
>> >
>> >> In a web request you also have access to the User Agent info.  You can
>> >> combine this with the IP to get a pretty close approximation to a
>> >> single user.  Subsequent clicks from different IP and/or User Agent
>> >> will likely be a different person.  Though not foolproof, it's closer
>> >> than IP only.
>> >>
>> >> On Wed, Feb 6, 2013 at 1:10 PM, Ken Snyder <kendsnyder at gmail.com>
>> wrote:
>> >>> We have an email newsletter that includes a tracking image, plus each
>> >> link
>> >>> contains a hash so that we know which recipients opened the email and
>> who
>> >>> clicked on what link.
>> >>>
>> >>> The client is asking if it is possible to detect a) if that newsletter
>> >> has
>> >>> been forwarded to someone else, b) if that new recipient opens the
>> email,
>> >>> and c) if that new recipient clicks on the links.
>> >>>
>> >>> My only thought is that we could make a loose inference of second-hand
>> >>> opens and clicks by tracking the IP Address of each open and click.
>> E.g.
>> >> if
>> >>> there are two clicks from two different IP Addresses it *could be*
>> >> another
>> >>> person.
>> >>>
>> >>> Is there any other way to detect second-hand opens and clicks?
>> >>>
>> >>> Thanks,
>> >>>
>> >>> Ken Snyder
>> >>>
>> >>> /*
>> >>> PLUG: http://plug.org, #utah on irc.freenode.net
>> >>> Unsubscribe: http://plug.org/mailman/options/plug
>> >>> Don't fear the penguin.
>> >>> */
>> >>
>> >> /*
>> >> PLUG: http://plug.org, #utah on irc.freenode.net
>> >> Unsubscribe: http://plug.org/mailman/options/plug
>> >> Don't fear the penguin.
>> >> */
>> >>
>> >
>> > /*
>> > PLUG: http://plug.org, #utah on irc.freenode.net
>> > Unsubscribe: http://plug.org/mailman/options/plug
>> > Don't fear the penguin.
>> > */
>>
>>
>> /*
>> PLUG: http://plug.org, #utah on irc.freenode.net
>> Unsubscribe: http://plug.org/mailman/options/plug
>> Don't fear the penguin.
>> */
>>
>
> /*
> PLUG: http://plug.org, #utah on irc.freenode.net
> Unsubscribe: http://plug.org/mailman/options/plug
> Don't fear the penguin.
> */


More information about the PLUG mailing list