Stop Intellitxt parsing certain elements

2 Mar
2011

I have a few clients who use IntelliTxt, the “in-text” advertising platform from Vibrant Media.

Mostly, you simply add some JS at the bottom of your page and IntelliTxt scans your HTML for keywords that advertisers have bid on. But I had a situation today where I wanted to exclude a certain section of the page from being spam injected. Looking at Vibrant’s Implementation Guide, it didn’t appear possible, but I don’t give up easy.

Using Safari’s activity monitor, I looked for all Intellitxt files being requested from the page

Opening up the highlighted file, gives you a list of exluded or skipped nodes in your HTML. According to the Implementation Guide, this set is different for each client, but it looks pretty generic to me. For this particular client it is..

1
'<select>','<a>','<noscript>','<img>','<b>','<h1>','<h2>','<h3>','<h4>','<h5>','<h6>','<h7>','<strong>','<em>','<i>'

So, you could take your div or p node and change it to an h6, for example. But what interests me is that on my client’s skipnodes set, we have a ‘nointellitxt’ setting, and simply putting:

1
<p id="nointellitxt">lovely spammy words like 'dresses for sale' or 'cars' or 'lady gaga singles'</p>

Means that this particular node will be skipped by IntelliTxt (although not excluded – it seems IntelliTxt will still parse this text, but not serve contextual ads on it).

So there you have it – I’m not sure why IntelliTxt don’t highlight this simple exclusion process on their site, but I’m here to help…

Comment Form

top