The backlog for an agency on its own sites is typically quite long… and my MarTech publication is no different. Last year I signed up for RevResponse, a site for whitepaper, trade publications, and ebook publishers to extend the reach of their publications by paying for each registered download of them. With the number of marketing whitepapers out on the market, I thought it would be an outstanding means of both monetizing my publication and providing value to our audience.

Unfortunately, while the selection and quality of publications are extensive through RevResponse, the service itself is limited. Either you can go in and find the links you’d like to share, or you can download a full XML library of every publication they have. They provide no filtering capability or API options – just grab a full 13Mb file.

To complicate things even more, the RevResponse images and XML file are hosted on a non-secure host. Since our site is hosted securely, that means that I needed to copy all assets over to our site to work with. I didn’t want to slow down our publication site, though, so I decided to host it on a third server under a wildcard SSL certificate so that we could easily access it.

To further complicate things, RevResponse updates their library continuously, so I needed to check it every hour to find new publications that are in the rotation. Once a publisher runs out of funds, the whitepapers are removed. I don’t want to annoy visitors to our site by pushing them to a whitepaper they can’t download!

So – I had to build a middleware system that acquired the library and all image assets and hosted on our site at least once an hour. Then I had to parse that data and put it into a manageable format that was easy to query and work with. We wound up doing a number of things:

  1. With the assistance of my food friend Adam, we built jobs to download the library each hour.
  2. I built a parser to only save the publications in categories that were applicable to our publication.
  3. I built an API to access the data from our publication, with the capability to pull a specific quantity, or filter the selection.
  4. I built a custom WordPress Plugin (that won’t be published publicly) that added a shortcode, sidebar widget, and filtered content for the posts and our feeds.

The best feature of the system is that every single post on our site interacts with the API. The plugin passes the tags from our article, compares them with the keywords of the whitepaper, and then the platform returns the most relevant whitepaper to our audience based on the keyword overlap!

I’m looking forward to seeing how the system performs over the next few months. And, of course, I’d love to help RevResponse build these features out for all publishers – no doubt they could increase their own revenue and extend their reach!