Today’s guest blogger is ReachLocal CTO Jeff Hagins, who seeks to dispel the “Mirror Myth” regarding ReachLocal’s Reverse Proxy technology once and for all.
One of the elements that makes the ReachLocal platform powerful is our ability to track the conversions (phone calls, emails, web forms, etc.) generated by the advertising campaigns we run for our clients. We do this through our patent-pending technology based on a Reverse Proxy architecture.
However, over the years, there has been some confusion regarding how ReachLocal’s Reverse Proxy actually works, with some people stating that Reverse Proxy actually "mirrors" an advertiser’s website (it doesn’t!). This misconception has been quite persistent, and I want to correct the inaccuracies and misinformation I’ve seen posted across the Web.
Let me start by stating that at ReachLocal we do not mirror our advertisers’ websites. That is, we do not create an identical copy of the advertiser’s website and serve it from a different URL. There are many downsides to mirroring such as:
- Content gets out of sync. When the original site is updated, a mirror site will take some time to update and can range from minutes to days, depending upon how mirroring is set up. In this case, an Advertiser who updates their site on February 14th with their Valentine’s Day specials will have a legitimate issue with their Internet marketing company who didn't pick up the changes.
- It is difficult to hold a mirror site accountable for reported traffic. The original site does not see the traffic coming through to its mirrored twin. If an advertiser received 100 visitors to the mirrored site, the advertiser could not independently confirm the visits by reviewing the logs with their webmaster.
- Server-side interactions cannot be mirrored. Actions such as logging into a site or submitting a form generally requires the server running the website to capture the form data or validating the login. These functions cannot be duplicated on a mirror site. All the mirror site could accomplish would be a matching look and feel: the visitor might think he/she is submitting a form, but it actually goes nowhere. So much for that conversion.
So given all the issues with mirroring, we took a completely different approach and leveraged a Reverse Proxy architecture to build our patent-pending system which enables us to track conversions without the nasty side-effects that mirroring brings with it.
How does ReachLocal Reverse Proxy work?
Reverse Proxy functions very differently from mirroring in that no content is hosted by the Reverse Proxy servers. Instead, for every request from a browser to the Reverse Proxy environment, our Reverse Proxy servers in turn make the exact same request from the origin site. The content from the origin site is scanned and phone numbers replaced in real-time then sent on to the requestor’s browser. It's like two friends who are so mad at each other that they insist on using a third person (i.e. Reverse Proxy) to talk between them. The exact message gets passed on, but sometimes the friend-in-the-middle (Reverse Proxy) changes the content for something more appropriate for the person receiving the message ("I hate you!" becomes "I'm not very happy with you at the moment", etc.). In our case, instead of adding tact, we change phone numbers.
A great example of how Reverse Proxy technology works is with the Google Chrome translation bar extension. If you land on a page that is originally written in a language you can’t read (e.g. German), you can select your language of choice (English), and the Google extension will automatically translate everything for you. The content is still coming from the original German web page, but the new translated text you’re reading looks as if a web developer went in, translated the entire site, and republished the page just for you (and did it all in under a second).
So with Reverse Proxy, the advertiser’s original site still functions as the host of the site’s content, sees all requests, performs all server-side interactions and is able to immediately serve updated content. In this way, Reverse Proxy allows ReachLocal to serve content as if the advertiser created the site for targeted traffic without having to change their website one bit, while also enjoying the benefits of native site traffic by actually serving it themselves. The browser gets the same html file sent by the origin site, but with a new trackable phone number that enables us to capture the conversion data from the ad campaigns we run for them.
Finally, we do all of this with a fully redundant, high availability, geographically distributed environment (across five different data centers worldwide). That means our Reverse Proxy environment is always up and running, and the user’s browser connects to the Reverse Proxy site that is closest to them, delivering the best possible performance!
Simple. Uncomplicated. Powerful.
That’s ReachLocal’s Reverse Proxy.
To learn more about the technology that powers the ReachLocal Platform, visit our website. Have questions about this post? Leave a comment and let us know.