Ghosts from Websites Past – Updating a 15 Year Old Website
A story about a man, a creative agency, and website that hasn’t been refreshed in over a decade.
Websites in 2004 were a fun thing. Actually, they weren’t. Everyone with Dreamweaver or Frontpage called themselves a webmaster. The web was still littered with GeoCities sites. Creating and updating a website often meant writing code and uploading files via FTP.
And websites were super ugly.
If you don’t agree, head over to archive.org and do a quick search to see how your favorite website looked 15 years ago.
During this time, I was operating a comedy website called Broken Newz. The “Z” was very trendy at the time and a domain with the “S” wasn’t available. I built the backend from scratch. A content management system that oddly enough wouldn’t recognize quotation marks, and I wasn’t smart enough to understand why. But prior to that, I was creating new pages every day by creating a new .ASP page and uploading it to the server. From there, updating links and images on the homepage was a bit of a process but I was dedicated.
Today, I wrote this article in a .doc. I logged into our website. Pasted the copy with a pretty picture and hit submit. We don’t have flying cars but we have WordPress. Close enough. The whole process of posting takes minutes. Fortunately, everyone can do this today.
Well, not everyone.
This spring we gained a new client who has been running a website since 2004. He needed a refresh. It was fun talking to him about that time. Back when we still all checked Classmates.com, when we were stoked because Newgrounds.com did a link exchange with us. How hard it was to update a website. Except this wasn’t the past for him. He was still living that world. He spent the last 14 years uploading pages, over 4000 of them, one file at a time.
I connected to his server via FTP and there they were. Over 4000 ASP files. It was like opening time capsule. So many memories. Then the anxiety kicked in. How am I going to do this?
The first attempt, after the initial redesign of the site, was to input all the pages one by one into a new database. Our first idea was to cut and paste. Two of us made this attempt. After a day of opening files, copying the relevant info, pasting it in the right fields, uploading new images, and hitting ‘publish’… we did the math. We realized we were going to spend the next three weeks monotonously copying and pasting. We needed a better system.
We took the next step and wrote a nifty piece of code that eliminated all data but the headline, images and story and converted it into an XML file. From there, managed to nearly flawlessly import the data and created 4500 new pages.
The pages, as I mentioned before, were ASP files. The site ran on an old window server that ran at the speed of a packing mule. We need to generate new URLs and not skip a beat with Google. When running an SEO report, I learned this site already had over 14k keywords in the top 100. If I screwed this up, 15 years of work went down the drain. I have a lot of confidence in 301 redirects, but still… 14k top-level keywords. We can’t miss anything here.
This same SEO report showed me something pretty interesting. 62% of his site was accessed via mobile. But his site wasn’t mobile friendly. In fact, it was a giant mess when I looked at it on my iPhone. Who is taking the time to try to read this from a smart phone? The whole experience would be “enhance, enhance, enhance”.
We also wondered how his site ranked so well since Google now ranks your site mobile first. Google really couldn’t give 2 clicks about your desktop version. If you’re not killing it in mobile, you’re dying.
Things were coming together. Pages were created. Bits of old code in the stories were flubbing up the design. I used my favorite piece of CSS for that.
From there we trimmed down some image sizes, set up the SEO and launched it.
Everything went just as hoped. The site switched over like a dream. The redirects worked. The site was faster. The new design up beautiful.
And the 62% of the people visiting the site could now read it.
This was the smallest/biggest project of the year and easily one of the biggest learning experiences for us. We took a relic and turned it into a flying car.
*note to readers: currently we still do not have flying cars.
It’s been two months since the launch of the site. Traffic is up over 40%. 2500 new keywords in the reports. And the owner of the site can now update his site from his mobile phone. And in record time.
But the client is not the only one happy. We’re pretty fricken stoked. We took a project on that could have easily been a time-consuming mess. It could have blown up in our faces and lost the trust of a veteran webmaster. But instead we banded together and came up with a pretty nifty solution. Now we’re a little smarter and ready for that next stellar challenge.