We manually update our XML feed when we publish an article on our
website.
I presume you mean an RSS feed here.
Can we add a javascript tracking pixel (from phpadsnew) into
the XML file to track how many times our feed was accessed?
This is a lot of questions, so I'll try to deal them one-by-one.
RSS is a syndication protocol, although it's usable (and used) in two
ways. One is to supply a "summary" feed, where the feed is used as
teasers and links into a site. The other is to send sufficient
"content" embedded in the feed items that users read everything they
might wish to from reading the feed itself (usually long text articles,
but images are still puled from the source site). Obviously both of
these applications of the technology have quite different meanings for
"user" and "to read", meanings that affect the relevant stats you want
to count.
In a pure "content" RSS feed, you just can't count the "readers" unless
you embed deliberate tracking, because you've already published and
permitted redistribution of everything they need to make the content.
Now the tech issues.
You can't "add JavaScript"
You might get the same results as adding JavaScript by finding what
that JavScript does, then doing that directly. A "JavaScript tracking
pixel" is just a tracking pixel that's embedded in a page by using a
<script> tag rather than am <img> tag (it works no better, but it can
be easier for content authors to set up). If you find what the URL to
the image is (probably complex and with a site id embedded in it) then
you can add that pixel to the RSS feed.
Adding an image to an RSS feed item like this usually involves sending
some encoded HTML in the description elements for the feed items. That
itself isn't trivial -- some blog publishing engines do it well, other
quite serious RSS creators do it badly or almost not at all. We'd have
to know just how your feed is created to advise further.
If you're really capable of doing this, then (IMHE) you can set up your
own "web bug" pixel tracker, hosted on your own server, and do it that
way. This avoids paying money to a tracking service, loses the ability
to cross-reference users with their browsing habits on other sites, but
best of all it's likely to still work for privacy-sensitive readers who
block web bugs from known tracking companies.
Just to get a rough idea how manyh subscribers we have?
What's a "subscriber" ? Subscriber to the field, or someone who later
reads your syndicated content?
If you want to count the feed subscribers, then look at your server
logs. If you want to count the reading eyeballs, then embedded web bugs
are a practical way.
